Calix, Inc. Aktie

Greetings, everyone, and welcome to the Calix Second Quarter 2026 Earnings Conference Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Nancy Fazioli, Vice President, Investor Relations.

Nancy, please go ahead.

Thank you, Darryl, and good morning, everyone. Thank you for joining our second quarter 2026 earnings call. Today on the call, we have President and CEO, Michael Weening; and Chief Financial Officer, Cory Sindelar. As a reminder, today, after the market closed, Calyx issued a news release, which was furnished on a Form 8-K, along with our stockholder letter and is also posted in the Investor Relations section of the Calix website.

Today's conference call will be available for webcast replay in the Investor Relations section of our website. Before I turn the call over to Michael for his opening remarks, I want to remind every -- but on this call, we will refer to forward-looking statements, including all statements the company will make about future financial and operating performance, growth strategy and market outlook, and that actual results may differ materially from those contemplated by these forward-looking statements.

Factors that could cause actual results and trends to differ materially are set forth in the second quarter 2026 letter to stockholders and the annual and quarterly reports filed with the SEC. Calix assumes no obligation to update any forward-looking statements, which speak only as to their respective dates. Also on this conference call, we will discuss both GAAP and non-GAAP financial measures. A reconciliation of GAAP to non-GAAP measures is included in the second quarter 2026 letter to stockholders that was posted yesterday. Unless otherwise stated, all financial information referenced in this call that will be non-GAAP.

With that, Michael, please go ahead.

Thank you, Nancy.

Much like the dot-com revolution before, the AI revolution is changing our society functions and will transform business model in every industry. The key difference between the dot-com era and the AI is a racing -- the AI rate of change is unprecedented, and those who move fast will lead the industries they serve. The second quarter was the beginning of Calix, the AI leader. -- as we began realizing the value of our 15-year investment through the first full quarter of our AI mated Calix platform being live. .

Calix is access to data, insights and the ability to autonomously or through team member augmentation improve operations, marketing, support and subscriber experiences for our service provider customers regardless of size. These expanding Calix 1 capabilities to enable our customers to address the threat of broadband commoditization through differential experiences resulting in winning new subscribers growing revenue, higher retention across consumer, business, MDU and the medical segments they serve, while improving operating costs with a predictable implementation of AI capabilities.

Customer interest in Calix exploded in Q2, driving record RPLs, record software and services revenue, while the strength of the platform was evidenced by the significant jump in software and service margin as we had all customers live. We also tripled the number of customers that signed up for agent workforce cloud. More important is the makeup of those customers. They were not only the invaders. We partner with Calix early in the product life cycle and are committed to the value that can be realized in their business model by being first to market -- the customers who signed up in Q2 span the entire adoption life cycle from early adopter, which is expected to late majority, which signals a significant Agent Workforce cloud-enabled ship across our base.

The shift is that every business leader knows they must have in the strategy or they will be at risk. The late majority adoption crude that Calix 1 is a secure, trusted and predictable approach for every customer to adopt AR. It is secure and trusted as we have invested [indiscernible] years into the platform and domain now to meet our customers' needs. Since November 2023, we have evolved the platform to be AI native while ensuring that our processes, culture, security and governance expanded to meet the needs of the AI era for our customers.

It is predictable as our AI native platform architecture allows us to use any AI model and is our belief that hardened open service models meet our workflow, use case and industry needs. That means Calix for the largest issue that is raging across all industries. How do you use AI predictably as one cannot calculate an ROI on a cost that can quickly run out of control. In our architecture, we have a true tokens through hardened open serve AR models and the acquisition of Pure Compute.

Paliton customers are adopting a platform that delivers AI capabilities that can be trusted are secure and are offered at a predictable cost which will yield an that will grow every month with their subscription as we acquire new subscribers a low revenue, reduced churn and lower operating costs through the power of Calix agent workforce cloud. At Investor Day, I shared a slide that showed our rate of innovation. On the first generation of our platform, our annual feature rate peaked at 181 in 2018.

Our second-generation platform featured sorry, our second-generation platform feature rate, PD18 in 2024. With the third generation of our platform launched, our customers can expect that our past innovation velocity will be quickly suppressed with Agent Workforce filing, which brings me to the Calix team. This team continues to win awards as one of the best cultures in any industry. Having outlined a subset of awards in the investor letter, including Fortune recognizing Calix and the 100 Best Companies to Work -- for list. The strength of this culture has never been more important.

While AI is a powerful technology, it requires teams to get the most of it. Our customer success organization will help our customers leverage agent workforce cloud to transform and inside Calix, our leaders are taking a human-centric approach to AI to ensure that Calix transforms internally by understanding how we work so that we can identify those portions of our business that will benefit from it. We are using AI to gain operating leverage across Calix.

In summary, we are committed to transforming our customers and ourselves through the power of AI. With that, I'll turn it to Cory to cover the details of an amazing quarter. Clark?

Thank you, Michael. We saw continued strong and broad-based demand in the second quarter, delivering record revenue of $293 million, a 5% sequential increase and 21% year-over-year growth, exceeding our guidance range. Importantly, this quarter, we experienced a return to strong software growth. Service and -- software and service revenue grew to a record $50 million, up 7% sequentially and 16% year-over-year. as customers began to realize value from Agentic workflows on Calix One.

Last quarter was the completion of our platform migration -- we set our focus with pivot to identifying our BSP customers, and RPOs would accelerate...

Sorry, Cory, can I please cause you. Daryl, on this operator, I understand from some of the people on the call that the call has been dropping. Could you please come on and let us know if we should restart.the call.

I am here, and I have heard everything thus far.

Okay. Apparently those who are participating by webcast or having the portions of the call that dropped. And I'm wondering if we should start the call once again.

I can confirm as well the webcast has been stable. And I've been spot checking and I have heard as well through the webcast.

Okay. So you can confirm that the webcast is acceptable and there are no issues on -- for the listeners?

As far as I can tell, there are no issues.

Okay. Pala, Corey, you can go ahead with where you are continuing.

Importantly, this quarter, we experienced a return to strong software growth. Software and service revenue grew to a record $50 million, up 7% sequentially and 16% year-over-year as customers began to realize value from agent workflows on Calix One. Last quarter, with the completion of our platform migration, we set our focus with pivot to identifying our BSP customers and RPOs would accelerate in the second half of 2026, Our customers moved faster than anticipated as we saw record RPOs of $386 million, up 3% sequentially and 11% year-over-year. .

Current RPOs were $162 million, up 3% sequentially and 21% year-over-year. Based on the strong momentum exiting the second quarter, we continue to expect RPO growth to accelerate in the second half of the year as we deliver additional identic workflows and demonstrate the value of Calix One. Furthermore, as we discussed last quarter, we would be running on a single cloud infrastructure this quarter. As such, having garnered the corresponding cost reduction, -- and when combined with the Gentek platform-driven demand, we yielded an 810 basis point sequential improvement in non-GAAP software and service gross margin. And to give you a sense of the continued leverage in the model.

I will break the precedent and say that we expect software and service gross margin to set a new record in the third quarter. As the AI leader in the broadband space, we are also focused on being the leader in human-centric AI deployment inside our own company. We've demonstrated meaningful operating leverage in the second quarter of 2026. Non-GAAP operating expenses were approximately $122 million or 42% of revenue, down from 45% and in the prior quarter, reflecting both leverage in our growth model and early productivity gains from our human-centric AI investments as well as lower incentive compensation and timing of certain expenses.

Turning to appliances. Appliance revenue was a record $243 million, a 4% increase sequentially and a 23% increase year-over-year. Non-GAAP appliance gross margin was 52.9%, representing a decrease of 460 basis points sequentially and 170 basis points year-over-year due to higher memory costs, which was partially offset by memory surcharges. As a result of the above, non-GAAP net income was $31 million or $0.47 per diluted share, above our guidance range. and we generated free cash flow of approximately $12 million.

Our balance sheet remains strong. We ended the quarter with cash and investments totaling $194 million after deploying $69 million to repurchase 1.6 million shares. DSO was 42 days, and inventory turns were 2.7%, reflecting deliberate investments in inventory to secure supply and meet continued strong demand. Turning to guidance. For the third quarter of 2026, we expect revenue between $301 million and $307 million, up 4% at the midpoint over the prior quarter. This reflects continued strong broad-based demand even as customers are more tightly managing their own inventory in response to higher memory costs.

For 2026, we expect annual revenue to grow at the higher end of the 15% to 20% growth range provided last quarter. Our third quarter non-GAAP gross margin guidance is 2% at the midpoint reflecting higher memory cost impact. As we navigate this industry-wide exogenous event, we remain focused on supply to ensure our customers can continue to meet strong subscriber demand. while we maintain a footprint aggressive stance. As you are well aware, there are many inflationary cost pressures across all industries. In our space, memory costs represents the most extreme of these pressures.

As we partner with our customers on surcharges, they value certainty. Certainty on the costs and most importantly, certainty of supply. Our surcharge program is structured to deliver exactly that with the goal to recover the incremental memory costs without adding profit. This means over the long run, memory surcharges will be gross profit neutral while remaining a headwind to gross margin. Given the implementation of our memory surcharge program, we expect appliance gross margin will bottom in the third quarter of 2026.

The third quarter of non-GAAP operating expense guidance is $124.5 million at the midpoint, a sequential increase driven primarily by the timing of expense and higher incentive compensation, partly offset by continued productivity gains from our human-centric AI investments.

Michael, back to you.

Thanks, Cory. .

And last year, there's been much debate about AI agents and the effect it will have on customers and the markets they serve. We believe that Calix is uniquely positioned to take advantage of the AI market opportunity in the broadband market based on our 15-year investment in the Calix platform and our work since November of 2023 to evolve the platform and our processes for the opportunity ahead. Our customers all know they need AI, and we are best placed to deliver subscriber or revenue growth, churn reduction and cost improvements with Agent Workforce Cloud in a trusted secure and predictable way. predictability of AI cost is the biggest issue gaining AI adoption.

And our architecture has solved that problem for our customers. trusted, secure and predictable costs are now possible for all customers regardless of size, and we began to see the results of Calix 1 and Asian Workforce Cloud in Q2. tripling Calix 1 contracts while delivering record RPOs and record software and service revenue is just a start. This is the beginning of Calix, the AI leader and now that the platform is live and running, we are excited to the opportunity for our customers to grow in the markets they serve and for our Calix team members as they leverage AI to transform how we do business and deliver operating leverage for our shareholders.

Nancy, let's open the call for questions.

Gary, you can open the call for questions.

[Operator Instructions] Our first questions come from the line Joseph Cardoso with JPMorgan.

For my first, if I may, I just wanted to touch on your expectations for gross margins to bottom here in the third quarter. totally appreciate it's a difficult operating margin -- or operating environment. I guess what's just driving the confidence here relative to calling the floor? Is it largely related to cycling past the grandfathered backlog relative to the surcharges -- or are there other levers you're seeing relative to inventory, product redesign, et cetera, that's kind of driving your confidence here in terms of calling the floor in 3Q? And then I have a follow-up.

Yes, Joseph, it is exactly as you outlined. It is the fact that we have grandfather a certain portion of the backlog and as we go through the next few quarters, that backlog as a percentage of the total will shrink. And so new orders are being -- the surcharges are being assessed that kind of our current cost structure. And we're also adjusting those now on a monthly basis as opposed to a quarterly basis. And so over time, we expect that gap to getting to gross profit neutral to improve. .

And I should say that our goal for the whole surcharge boring is to get to gross profit neutral. And as we partnered with our customers this quarter to address these higher memory costs, it became clear that they value certainty and certainty of cost and more importantly, certainty of supply. And so we had to modify our program to address this. And so consequently, we did not raise prices on backlog for a second time.

And meanwhile, new orders, we will we will adjust to reflect our actual costs and adjusted monthly instead of quarterly -- so as more revenue comes in from those new orders, we closer we come to being gross profit neutral. And that's why I believe in Q3 we will bottom in the client gross margin.

Got it. Makes sense, Corey. And then maybe as my follow-up, you raised the full year guide to the higher end of the range, which if I take literally at the high end implies a pretty solid sequential ramp here into the -- well into the upcoming fourth quarter. First, am I thinking about that correctly relative to how you're framing the exit rate for the year? And if so, could you help us think through the drivers given all the moving pieces? Obviously, you've guys gone into detail around the new platform momentum. I believe bead should start kind of flowing through here into the fourth quarter in terms of a tailwind -- and I assume some benefit from the surcharging pricing there.

So maybe just help us think through if I'm thinking about that framing as well as maybe what are the drivers here in terms of the acceleration into the fourth quarter? -- well, I guess, third and fourth quarter, but maybe what's implied in the guidance?

Yes, -- yes and yes, I think is what's the answer to that. If we take a look at what we're seeing from our customers, the demand environment remains strong and understanding that we can see all their activity through our cloud. even though we're seeing these higher surcharges and then changing their ordering paths. The underlying demand environment remains very strong. So that has not changed at all in this environment. And if you think about it, the quickest return on investment is adding a new subscriber to an existing network. And it doesn't really matter what those higher memory costs are when it comes to connecting a new subscriber to their network, they're going to move -- we're going to go ahead and buy that premise is equipment and move forward. So number 1 is the underlying demand environment is strong. .

Two, you saw a return to our software growth. Exiting the second quarter, the momentum is very high. And so we reiterate the fact that we expect our software to continue to reaccelerate into the back half of the year. And you're correct that we expect to have some dead revenue pick up in the third and fourth quarters as we exit the year. No real change in terms of that demand environment. Certainly, in the beat environment is extending a little bit in a lot of areas, but as it relates to the impact to revenue for us in the second half, no changes in that environment. And so we're looking at a back half that is strong across the board.

Let me contextualize with a couple of customer interactions. I've been on the road quite a bit over the last quarter. talking to customers. And so the closest point I'm not adding a subscriber, it's a really salient point because we've been saying this at all times is that as organizations move to their build cycles, and, for example, they come to end of building fiber. Then their entire organization pivots away from being a really efficient construction company to how to become a really great sales and marketing organization that can win subscribers.

And so the first thing I would say is that I continue to hear that theme in a lot of places. In fact, I had 1 CEO conference and the panel I was on, he basically said he can see the end of really the baby builds in fiber for his company and everything that he's thinking about is how do I add subscribers. And for Calix, that adding in subscribers means that we install a new subscriber, we provide new services, and we have an incredible strong revenue opportunity and our customer success organization and the tools we're putting place are really great at helping on the marketing side.

The other side of it, though, as I said in my opening statement, which cannot be understated is that we've cracked the code on how to make AI predictable. And this means that customers all know I need AI-- and in fact, we, as Calix, as we look at the change of how we get leverage out of our operating model and how we transform our teams, we know we need AI. The biggest challenge that we had, frankly, is how do we do it in a predictable way. And this predictability cannot be understated where we believe that Calix will become the easy button because not only do we offer a predictable cost model we had 25, 30 CEOs in our office a couple of weeks ago, and we provided a very predictable road map over the coming quarters where we show workflow by workflow, here's how we're going to identify their business and where we see the payoff ROI and output.

And frankly, those CEOs all left feeling incredibly enthusiastic about the opportunity to transform their teams and how AI will very clearly with the strong ROI change their business, and that's just going to snowball as other companies see this implemented within their peer group, and then they adopt quickly to you because -- it's inevitable. We you have to adopt it. It's just about what's the fastest path with the most predictability and that's what we've cracked the code on.

Our next question is come from the line of Scott Searle with ROTH Capital Partners.

Cory, maybe to just dive in quickly the impact on the memory charges in the third quarter. I'm wondering if you could quantify both the sales and EPS impact. It looks like just from a cursory glance that EPS would have bracketed or maybe even been higher if you had fully implemented memory surcharges across the board in the third quarter. And then looking to the prior guidance from the Analyst Day for 27, 28.

I wonder if you could give some early thoughts in terms of how are you progressing in terms of the comfort on the revenue outlook? And particularly on the OpEx front, much lower in the second quarter. It sounds like you're going to be able to continue to leverage internal AI capabilities and efficiencies to carry that forward into the future. So does the model start to change or accelerate a little bit of when we should start to see operating leverage?

Great. So let's talk about that. We'll go ahead and talk about that first. So I think the revenue outlook is firming up. I think there are plenty of demand drivers as we look into '27, '28. Software is reaccelerating the power of the agentic workflows is resonating with customers. And so we see that we'll continue to drive our software and our software gross margins higher. We've got the tailwind of bead will happen at some point in 2027. And in the meantime, customers are continuing to add subscribers. So we're seeing no letup in terms of the demand environment. And so to reiterate our 15% growth targets for '27 '28, they're on track.

In terms of the OpEx leverage, we are -- we are a leader in terms of the broadband space. We're obviously going to continue to be an AI leader in the human-centric deployment of AI and sialic -- you're starting to see some evidence of that work in the quarter. We expect more of that to come as we're committed to that as a strategy. I would say it's too early for us to accelerate that OpEx improvement, but I understand it's a keen focus inside the company to accelerate that. But I'm not changing guidance at this point, which is we will drive OpEx at a lower rate than revenue growth next year, leading to operating leverage. And we'll see what we can do to get to half or better growing at half rate revenue or better leave that as a goal, not a commitment.

And Corey, just the memory impact sales and EPS in the third quarter, that would be helpful if you could provide some context. And Mike, if I could just quickly, the competitive landscape, there was a lot of talk throughout the second quarter about the StarLink impact, et cetera. I'm wondering if you could just update your thoughts in terms of, I'll call it, the terrestrial competitive landscape and then throw in the satellite impact of how that's impacting your customers and what they're doing.

Yes, it's Scott. Essentially, it was a flip, right? We overperformed on EPS in the second quarter and the impact on not changing surcharges on the backlog had an impact into the quarter -- the third quarter. So we picked up a $0.05 in the second quarter. We lost a nickel in the third quarter. So for the year, we think we're EPS-neutral. -- that was part of what we thought about in terms of partnering with our customers and trying to get through this event of higher memory costs.

On the competitive landscape, excess fully like overtime competitor versus a fiber competitor. -- that changes market to market time by town. So competition exists, and it really just depends on how many people are building in that market and what their offerings are. Our position remains the same is that if you're in a market and you deploy the full car solution, you're going to be uniquely advantaged because of the fact that -- now again, this comes from a number of CEO who have proven this term over and over a guy, if you just are in market as a consumer and broadband provider, your at risk, if you are the broadband provider who takes the approach that we've built with our customers to being the dominant provider with the dominant brand in a town or an allocation you'll destroy your competition because you're selling consumer, you're winning the businesses, you're in the MDUs and that entire footprint comes together to provide enrollment capabilities through Smart Town that allows you to be the dominant brand with regards to talking to the PTI and the local school board around how do you provide WiFi roaming for all students who are underprivileged and we just want to have access to broadband to do homework.

How do you provide roaming as an augmentation technology for police for inlands. They currently have their public safety radios but they could be in areas where there's worse coverage or they're using an iPhone that you've given them. And now whenever they actually hit a roaming network and whether there's WiFi coverage. Placements can actually have faster speeds and better downloads on their iPhones and the P25 radios. And all these things come together to you building a great partnership with the Mayor with the Head of Parts and Rod with all these different groups, the fire chief the police chief, and then lay become the dominant brand in your town and that brand dominance makes you local and helps you win subscribers.

With regard to Sterlink, as we've always been saying forever, and the satellite providers, whether it's Starling or Amazon have a place. So if you're in a super area, it can put $500,000 to run fiber then Starlink is a great option if you're out on the boat, Starlink is a great option.

And so where it won't be an option is or where our customers know that if they lost the Starlink and they're in the town and they've got a higher connection, then they need to really look inside their business and decide what's going on because no logical customer is going to pick a sterling over fiber. The capacity differences are as phenomenal. And so fiber is going to destroy them from an experience point of view. And if the service provider does a smart thing. And the smart thing, for example, is they have to offer outdoor WiFi. And if you have the tax rate that's 40% on outdoor lipid, that customer is going to be wildly sticky because you've now put WiFi in by the pool, you got WiFi by the backyard by the garage, all those different places.

And there's going to be no impetus even if StarLink uses their significant capital availability to run this at a significant loss as we expect Amazon to do that cheap off or will be beaten over by the fact that I have a great experience, I have incredible speeds I have Wi-Fi everywhere I need it. I have virus protection that protects me every day and hacking pre protection. And therefore, plus when I go to the local parks as my phone picks up and it wanders around town, why would I ever go anywhere else? So satellite has a place that 5% to 10% of the market, the matter really comes down to customer execution and they should crush.

Our next questions come from the line of Christian Schwab with Craig Hallum.

I just want to be clear on the no surcharges on backlog. It appears to me that customer certainty on pricing and the decision not to surcharge backlog was made some time during Q2 as gross margins now are implied a little bit different than the Analyst Day on the 23rd and -- in addition, we also brought a substantial amount of stock at higher prices than where we're going to open up this morning. Is that fair? Or was that the plan all along?

No, Christian, it was not the plan all along. Our plan all along was to be -- to do cost recovery right? That was the plan we outlined at the start of that was to get to gross profit neutral. And as we partnered with our customers, it was important for them to have certainty around price. So we've made a couple of changes. So backlog was adjusted in the second quarter. where we implemented the surcharge from the second quarter. So to say there's no surcharges on backlog is not an accurate statement. There are -- what we said is we're not going to adjust at a GaN as we move forward, right?

So we gave them that certainty because customers were upset about changing price after the fact. So that was an acknowledgment of that in our partnering with them. And so the second side of that is that we've gone to adjusting surcharge pricing on a monthly basis versus a quarterly basis. so increase the frequency at which we have a chance to adjust the new orders coming in. And so that's what we've done. It had an impact to Q3 gross margins -- but ultimately, our goal has not changed, which was to maintain a gross profit neutral stance. And over time, we will narrow that gap.

Yes. Okay. That's clear. And then on a go-forward basis, kind of reiterating the fact of 15% annual growth through '27 and '28 plus. If memory prices continue to increase and we have surcharges that 15% is an organic number, correct?

All our growth is organic. Are you saying -- it's included surcharge.

Right, right. So as we get to this time next year, if memory prices continue to increase than the top line growth rate of the company should be, to some degree, greater than 15% is my question.

We'll see about that. right? So at the end of the day, they have a -- let's break that up between access and prime. So on the premises side, I see no impact to the higher memory costs. as we talked about, as they bring on new subscribers, particularly in an existing built network, it's the greatest return on invested capital. They're going to do that regardless of whether there's incremental memory costs. On the access side, it's really driven by CapEx budgets. And as you know, we are only just a small part of the overall rising cost environment. Right.

There are rising costs on fiber, labor, construction, fuel, et cetera. And so our customers, as they look at their CapEx portion of the business, we'll reevaluate how much they actually spend. So there can be actually demand destruction with the higher costs as it relates to the CapEx. But fortunately, for Calix is that the large majority of our revenue does come from the premises side. and access is a small part of the overall business. So to say that all the surcharges is additive to our revenue growth rate would probably not be an accurate statement. -- fully, that helps.

That makes it very clear. No other questions. .

Our next questions come from the line of George Notter with Wolf Research.

I just had some questions about the Agentic Workforce Cloud progress here. I know you guys made a decision to bundle that with the other cloud offerings and not really charge customers on an a la carte basis for that. And I think the view was that you would monetize via increased subscriber penetration over time. I'm just curious like where are we now? .

You've got some early feedback, I presume -- do you feel like that's still the right decision to drive for increased subscriber penetration? Do you think that model is going to work well? What's the perspective, again, a few months into the Agenticrollout?

Yes. Great question, George. If anything, yes, we think that's 100% right. And if anything, actually, we're even more confident in that approach because of what our customers are doing is what we basically did was we took everything with her. So you would go and buy everything like you buy 1 of the 1, 2 or 3 of the clouds. -- add on 1 of our smart components, you bundle it all together and you do it all bespoke. -- which meant that customers had a lot of complexity.

So on 1 side of it, our customers when we actually haven't it all in, they would say, well, on to buy it a la carte. And then when you make it out carsales -- and so by going down this approach, there's a couple of reasons why this is right. The first 1 is that you can't actually deploy effectively unless you're across the entire enterprise. So if you look at the 3 functions in the business, operations, marketing and then everything that you're doing around service or call center and field support -- every time you interact with a customer, it requires all 3 of those elements to come together in a workflow to deliver an outcome.

So at the first level, we believe that this has to happen because we have to have everything -- so that wasn't really a choice. So we went forward with that and that's what we did. So that's a further part of it. The second part of it is now that we're -- we've been in market for a full quarter, our customers really know that they have to do AI and this provides them with a very clear approach to actually transform their business. Because what happens is whether or not like if I want to change my marketing team, I may or may not be ready to change it at this moment in time. But I now know that I have the dental workflows to allow me to improve marketing when I am ready, and I'm ready to couple implementing AI in marketing with a change of my team.

And so this also makes us so that our customers can go at a pace that they require. And then the third part of it is our greatest growth driver is when customers add subscribers. -- making a few incremental pennies on top of a subscriber is not really the growth area. But when they -- as we said, we're $1 to $10 per subscriber, they have a new subscriber we go from 0 to great cash flow. And as you saw with -- that our margins are going back to where they were before and we're going to achieve incredible software and cloud margins, that means that that's going to be at a very, very profitable rate. And so which brings me to the last point is it's very predictable for our customers. And frankly, it's now that we've had a full quarter. It's very predictable for us.

So we can deploy AI in a highly predictable way. And our intent is not to just win a little bit here and there. But as you saw by the tripling of the cloud contracts, our intent is to roll up market, and we have become -- and this will be proven with every press release that starts flowing out like crazy, which our customer success stories that we are the easy button for AI and for a path to winning more subscribers, increasing revenue per subscriber and then reducing churn, which is how they grow.

Got it. And then just as a quick follow-up to that. Do you have metrics or any perspective now on customers that are actually using the genic feature? What's the feedback look like there?

It's good we just started rolling out the workflows. And so we're starting to see the productivity numbers. We are now to -- it's a good question. We're now in the process with our success organization going through and doing clear measured ROI with customers. So everyone sees the value of it, but have I measured the -- we know it's there. It's just about actually getting the physical measurements done, which takes time because, hey, I know what it costs me today, what is that workflow and that process cost me 90 days from now is what we're focused on measuring.

And so what you're going to see through the quarter through Q3 and as we ramp up for connections in Q4, started seeing a lot of workflows as we get that data which allows us to share with customers. That's what our press releases are focused on, is that a customer who sees a return on investment of that Agentic workflow, whether it's 10% or 7% we'll start seeing those shared out of press releases at a rapid rate through the end of the summer and into the early fall.

Our next question comes from the line of Tim Savageaux with Northland Capital Markets.

And you mentioned a return to record gross margins on the software side, which I think prior were maybe 66%, so call it, 67 plus. I guess my question to start with, and I have a follow-up is -- where can that gross margin go? I think we have discussed the potential for that to have a 7 in front of it at some point in the future. But I'd be interested in your expectations for the trajectory of software and services margins a little bit farther out and whether there's a ceiling that we should be thinking of or how to think about that?

We clearly see a pathway to have a 7 on the front of it. So we'll continue that mark. Ultimately, where Asentotes to is a little bit of uncertain because it depends on some of our approach related to large customers. We could actually end up in a situation where we are driving a lot more software revenue. at incredibly high gross margins and especially with private clouds, if we're doing a private instance on a large customer, the margin in that environment 100 points. .

So it's hard to say kind of how fast that will accelerate or where it has and to out to but 70 is in sight, and we'll just see ultimately where it assets out to. But -- there's a lot of headroom still left to go on expanding that software and services gross margin. 70% is just a way station on to where it ultimately ends up.

Great. And over on the appliance side, I guess, was my next question, which was -- it seems like you're looking for that to bottom and maybe tick up a bit in Q4. But again, kind of the broader question, and I don't know that you're ready to guide here, but -- is it reasonable to expect appliance gross margins to make their way back, I don't know, to the low to mid-50s by the end of 27, I guess? Or what sort of slope should we be looking at there?

Well, Tim, I would love to tell you what that looks like. But if you could tell me what memory costs are going to do, then I can probably help you triangulate on that. But if you believe the hyperscalers and their capital deployment, I think that next year will be harder than this year. but that's only if you believe those CapEx numbers. And as soon as you break on those CapEx numbers, it changes the entire pricing dynamic.

So it's hard to say. So all I know is we put together a program which, over time, will get us to a gross profit neutral stance. And so kind of regardless of what those surcharges are, we're going to continue to march to that path. And we'll just see what ultimately happens next year as it relates to memory costs. And that's the best we can do at this point.

Our next questions come from the line of Michael Genovese with Rosenblat Securities.

Great. I guess given where we are with some Calix one's rollout, and that's just started, and you've kind of got visibility to that driving an acceleration. I guess if we just tie that into RPO, my question would be sort of -- is there a way to kind of say how many quarters in a row you think RPO could accelerate from here given the visibility that you do have to the Calix 1 acceleration and going out to more and more customers? Like should we just think about RPO going up through the end of the calendar year? Or I know RPA is usually tougher seasonally in the first half of the year. Is there reasons to think it would also be accelerating in early '27 as well?

Yes, it's going to accelerate for sure because if you think about the year that we've had, right? So Q1 was a hamper of sales because of the fact that we were in the process of converting 1,200 customers from the previous platform to the new platform, and it was highly disruptive. -- while we had exactly 0 subscribers go down, we had some challenges with regards to some of our workflows on, frankly, on customer implementations. And so we -- and as 1 would expect that's a disruption to sales because the sales teams then are working with customers and helping them get through that.

We crossed that at the end of March, which then meant Q2. While there was some cleanup that had to be done and the product team was turning off the old system, the majority of the effort got to, okay, where is the value? So we've got customers back on track, and then we started to have the conversation around, okay, what's this going to deliver? And customers are asking the same thing. What's in it for me. Why did we do this? And you saw the impact. The impact was the tripling of contracts.

And I made a really important point in my comments and in the letter. And that is with regards to what were the types of customers who bought Asian workforce cloud contracts, very different. In the past, we would normally go into this cycle. Q2 would have been you have the innovators and the early adopters who are jumping all over it. Why? It is I believe in Calix, -- they know the business value that we offer. They're trying to change and improve how they run the business. They're trying to get out of it. right? While the middle majority, late majority and the laggards all sit on the sidelines and say, "I'll tell you what, when you come with ROIs like a great question, George, shown on the ROIs, but I want to see 30 of them before I actually buy because I'm the suspenders and belt decision-maker. And that was the traditional cycle, which bluntly I've been living with as a leader inside Calix since the day I started across 10 years in May. And that's been probably the biggest gating element in our business.

This cycle was radically different. We actually had a customer who I have personally been trying to close for almost a decade on smart home a decade, I have called on that leader, more times than I can count to convince him to transform its business and actually deploy our virus protection and malware and all the capabilities of what we're doing in smart home and on a really good sales leader, and we could not get them over the line. that customer, who would then be called a late majority in our traditional -- if you take what we learned in Marketing 101 in University he would be in the late majority. He signed up for Calix Cloud and in fact, asked us to extend the contract beyond our normal 3-year term. Why? Because he knows that artificial intelligence is not a nonnegotiable. This is not a nice to have. This is not -- I provide my customers fiber, and I would like to differentiate my value proposition by adding virus protection and malware.

This is everybody on the planet needs to deploy AI in their business or they're scored. They are going to get crushed by the competitor that does it. is that level of pressure. And every CEO is under that same pressure. I need to sit in front of my board, and it doesn't matter if my Board is sophisticated or unsophisticated. They're all -- they all know that when I come to the next board meeting, the only question I need to be asking it where very high plan and how fast is it going to drive results. We did this in a predictable way. we have become the easy button if they want to deploy AI effectively because we have a 15-year track record of demonstrating that we can be trusted bars with customers, that we do it in a secure manner. And by the way, the scariest thing about AI is how dangerous it can be from a truck and security point of view.

And last, we do it predictably. And this predictability that cannot understate how important this is, it is going to be a huge inhibitor on deployment for companies who cannot provide their solutions in a predictable way. And I can tell you, inside Calix, it is stopping a bunch of our AI projects. because without the predictability with regards to what AI is going to cost and with the -- the craziness of tons -- it's very hard to -- when the team comes and says, "I'd like to do this $100,000 project to implement AI to change this element of my business. Oh, by the way, it's $100,000 to do the workflow and people transformation -- but I estimate that the token cost is going to be this, but it could be 30x as much I really don't know, which means am I losing money? Or am I actually getting a return on investment -- and I will tell you that in a specific example, we currently use copilot with all of our employees.

We have a co-work proposal in front of us and that over my dead body, am I going to approve that when it's all token-based. And frankly, we looked at our pilot project and the costs or through the root -- and the ROI is nowhere to be seen other than, hey, my employee actually did a bunch of stuff better. If I can't see what the head count gains are or the productivity gains -- it's just like -- and because employees are throwing everything in the kitchen sink into the AI engine, hoping that their job will get easier. So our predictability in the form of cost -- but also, we sat down with our 1,200 customers.

We know there are billions of workflows they're running, and we know what can be identified easily and what cannot, which leads to a quick ROI means that our customer velocity is going to go through the roof especially as, again, back to what George said. Great question, which is where is the ROI, well, I've got a customer success army that I talk to every single day, and they know that their #1 mandate is find out what the current KPI is, implement the workflow and get the change in KPI so that we know we can basically say, you implement this workflow, here's your ROI -- and the power that we are going to bring into the broadband market is unmatched because we have access to our customers' data in a trusted way.

We have access to all of their billions and billions of workflows and we are the best place to actually turn those into AI workflows at a rapid rate. And that's why we spent 2.5 years building it out. so that we can now -- we went slow. And believe me, it was painful to actually go slow because everybody is going III and our product officer was under constant pressure from me saying, yes, that should ever it up, right? But he refused to because we knew that if we did this in a uncreable way, then the AI would start loosening and we do a all job. So everything that we're doing is trusted, secure and more importantly, predictable, not only from a business outcome point of view, which is the delivery of the ROI. So they know that they turn that works long, they're going to get this output and this improvement but also from a cost point of view, which was the architecture that Shane and his team brilliantly implemented, which allows us to use hard and open source and eliminate costs.

And so all these things come together a long way of saying, at less on make money for our customers and then in turn for ourselves and our investors.

Awesome. Last question, just to follow up quickly. It just does seem like the third quarter guide sequentially is a little bit below historical. And so I'm just I just guess I'm asking specifically on bead, was there any kind of change from the second half of the year into the first half of next year? Or any other reason that's kind of holding back specifically the third quarter guide from being a little bit higher?

No. I look at the third quarter guidance is in line with what we outlined in terms of expectations for the year, moving to the higher end of our guidance range that we provided last quarter. So I think on the under line, everything is tracking according to our plan. .

Our last question will come from the line of Ryan Koontz with the Needham & Company.

Maybe just start with some housekeeping. I know you're not reporting on customer tiers here. But could you give us any color that you saw across your different segments there and maybe the source of your 12% customer concentration in the quarter and update on any new Tier 1 engagements that may be looking at your private cloud options.

Yes, we're not going to provide kind of any color on customer breakdown on mix. And the 12% customer in the quarter were liberally to disclose who that is. But the expectation is, yes, they were a 10% customer in the quarter, but they likely will not be a 10% customer for the year. So kind of understand that was a blip in the quarter and you wouldn't expect to see that happen again.

And then on customer engagement, I kind of gave a lot of color in my last answer with regards to what the engagements are like and that goes across the board. That's all sides of the customers. The contracts that we closed in the quarter were actually -- the majority of them were smaller customers, which is pretty normal lease. You see that, but we're broadly engaged in sales cycles across all customer sizes. .

And the great thing now is that we're about to sell in because we've done a 2.5-year implementation and we can actually demo it and show it. So with larger customers, they have longer deal cycles. So the average -- on a large customer, a 12 to 24 months on deal cycle. So we're now deep into it. And more importantly, we're not talking about PowerPoints perhaps showing things. So great momentum.

Makes sense. And then maybe on your RPOs, you talked about expecting acceleration there. We did see a divergence in 2Q between current picking up and kind of that long term, not growing as much. Can you maybe expand on that in terms of the long-term RPO expectations going forward? Do you think that will kind of normalize going forward if you get some renewals.

Yes, Ryan, it's always a function of the tail on those RPOs and subject to customers coming back up for renewal. My expectation is that with the Carson contracts, we'll see an early renewal cycle. So you will probably see that total RPO growth will continue. But obviously, the current RPO is the 1 is eliminates kind of the effect of the shrinking renewal, the tail coming in. So strength in the current RPO is really what you want to focus in on. That being said, with the power of the Callison platform, I would expect customers of all sizes to renew their contracts or relate to move on to the platform. And so I think you're going to see total RPOs grow. .

Makes sense. Great. And then maybe lastly on -- you talked about token costs and your use of open source and you've migrated to GCP. Can you maybe expand on that a little bit in terms of how you compare competitively with other options with your open source approach.

What other options in what regard ?

Well, just if you would implement this on kind of an off-the-shelf frontier model or something like this, I mean, what type of cost savings are you seeing relative to...

So let's talk about that Right. Okay. I get what you're saying.

So when a company goes and use our intelligence to decide to build it in a bespoke manner. -- right? That's the first thing is that the challenge they have is that they also are now entering into software life cycles and so they need to consider that. With regards to us using a Frontier model, we're using our hard notes, the differences that we don't really see we can see the gains of it. So if you compare the gap between what is a frontier model and what is not source model, it's a very small gap, especially if you look at the workflows that we're doing. -- we're not asking this system to go and quote a new back-office system or do incredibly crazy things.

We're asking it to take what is a well-defined workflow well-defined context that exists in our knowledge layer and then apply that workflow with agents to execute against clearly defined boundaries and drive great outcomes. And so our use cases, our workflows and the models that we're going after, actually, they lend themselves perfectly to this approach. So I don't need to paint you flying on a unicorn over mountain and maybe it was 3D Ascenti, right? This is not the use case. So yes.

You may disconnect your lines at this time. Enjoy the rest of your day.

Good afternoon, everyone. For the next section, we have Calix CFO, Cory Sindelar. Cory, thank you for being here. Appreciate the time spending with us.

So maybe just starting off, the migration to Calix 3.0 was clearly a massive undertaking, and congratulations on getting it done. Now that we're on the other side of this and customers are seeing what the AI native platform can do, how are those early conversations going about Calix One adoption?

Yes. So about 1.5 years ago, we made a determination that we needed to re-architect the platform. And so that culminated in the first quarter where we migrated off of AWS onto Google platform. And the benefit of us moving to Google platform was not only to improve the scalability of the platform in terms of cost, but it gives us the ability to enter into some new markets. Part of which was to help us to do a private cloud to go into large customers to go international, but it also enabled us to bring AI functionality into the platform and bring automated workflows to the service provider marketplace. And so that's what just happened.

So in March, we finished the migration from AWS to GCP. And as you know, we release our software every 91 days on the second Tuesday of the second month, which happened to be last week. So we've done our first major release with the new cloud platform, and we are now at that point, engaging with the early customers, those that had already bought the 3 clouds from Calix. And I kind of joke around saying, these are the customers that are all in on Calix. If we built a margarita machine, they would buy it.

And so these customers are now taking advantage of the 3 clouds that they had in the Calix One contract, and we are working with our customer success teams to go demonstrated the value and the productivity enhancements associated with the agentic workflows. And so it's going in a very controlled manner, meaning we're going in and measuring kind of where their productivity is today. And then as they deploy the new functionality, what improvements are they seeing.

So that's kind of where we're at. So still very early days in terms of bringing it along. And as you see, as we build to our show of connections in October, more and more of those outcomes will become more visible as we do press releases and share that with the public.

Got it. And maybe the flip side of this is one of the debates we hear from investors is around the defensibility in a world where AI tools are becoming more accessible, AI disruption, right? You talked about security trust and 0.5 billion workflows of proprietary data as is your moat. Maybe just extrapolate on that a little bit and expand like what is it about the way you've architectured the platform that makes it really hard for someone else to replicate even as AI tools become more commoditized?

If you're going to allow agents to actually work autonomously in the network, you're going to have to construct them in a way that they don't hallucinate and that they are in a controlled manner. And so building that trust and security layer puts the provisions in around them. And then you are using the collective data collected from our 1,200 customers to train the LLMs in order to allow them to take action inside of the service provider network.

The platform is obviously have to bring context from other data sources. So through AA and MCP, we're actually pulling all that data together, combined with the knowledge of our customer activities that allows us then to provide these agentic workflows in a way that don't hallucinate and get to the outcomes that we're looking for.

Got it. Maybe another aspect of this is you made the decision that customers can't access any meaningful AI features unless they sign up for Calix One and collapse all their contracts into one. Maybe can you just give us an update in terms of how customer discussions have been trending since that decision? How has feedback been on that front?

Yes. So customers do not have to go to Calix One out of the gate, right? They can still purchase our clouds on an individual basis. And maybe I kind of explained that just a little bit. So prior to Calix One, we sold our cloud offerings on a persona basis. Think about it as a job description. So if you're in the call center, you're using Service Cloud. If you're in operations or a network role, you're using Operations Cloud. If you're a marketeer and you're doing the commercial arrangements and you're looking to bring on new subscribers, you're using Marketing Cloud.

All of those solutions were oriented to a role to a person. But if you're using agentic workflows, it's really a horizontal workflow. It's touching all aspects of the enterprise. And so to get to that, you really have to have all 3 clouds. It makes no sense because if you take an example of any of these workflows, it's going to touch every bit of those different 3 clouds. So what we've said is as we migrate forward with these agentic workflows, we're going to sell it as Calix One, which is taking all 3 clouds, putting it together into a commercial model that then looks at it as all subscribers are equal, and we can move forward. It doesn't preclude you, you can't. You can still buy it the old way.

So those customers that don't see the value yet in terms of agentic workflows can still buy the old way. For the ones that are early adopters, it was easy. They were very happy to sign a Calix One contract. They're looking forward to seeing the value being created through agentic workflows, and we're in that DV proving stage at the moment.

Got it. And so maybe if we think about 80% of the base that does not leverage all 3 clouds today, how are you thinking about them transitioning then? Is it more going to be -- at least in terms of your viewpoint to date, do you expect them to move to Calix One as they think about procuring another cloud? Or do you see them doing more of the piecemeal until they get comfortable in moving to Calix One? And then I have a follow-up there.

Yes. So the way we want to monetize the AI is not through selling more clouds. That ultimately is not the goal. The goal of Calix AI is to increase their productivity so that they can do more and/or do things faster. One of the issues we have with service providers is their ability to do more things. Calix has been able to deliver a lot of different functionality and capabilities and the way at which they consume it is gated by know-how and/or capacity.

So the promise of AI is to unlock that, allow them to do more and add more subscribers at a faster rate or introduce new services that they aren't today. And so for us, that's the way we monetize it because all of our operating system platform managed services are on a per subscriber basis. And so the quickest way for us to drive more software revenue is to increase the rate at which they're adding subscribers, not so much selling that third cloud.

Yes. That makes sense. And so maybe as a follow-up to this is like what's the milestone that investors should be looking at? Is RPO the right number to be looking at? And then as a second part to that question, how should we think about the cadence in terms of a potential inflection in that milestone?

Yes. Out of the gate, RPO is the telltale sign is probably a leading indicator for the progress on our software contracts. You're also going to then see it in the service cloud margin line as we make more progress there. From an early day perspective, we're still now building and proving out to folks the benefits of the AI through the DV creation. I expect -- and like I said on the last earnings call, we expect that RPO to reaccelerate in the second half, particularly as more of those proof points come into the existence and being shared with others. And ultimately, it leads up to our Connections show in October down in Vegas where we typically do very well with our customers. And fourth quarter, I would expect to be a big quarter in terms of signing contracts culminating kind of in that experience.

Got it. And then maybe just on the pricing side, like for the average customer who's going from 1 or 2 clouds to Calix One, like how should -- help us frame how we should see the uplift look like?

Yes. I mean -- so each one of our offerings is fit into a price volume curve. So the more subscribers you add, the better price you're going to get. But generically, I'll speak kind of at a high level to give you a sense for what you're talking about, let's say, a cloud is $0.50. So if you had 2 clouds, you're paying $1 and you want to get the third cloud, you add a $1.50. Put a new subs coming in at $5. So like I said, chasing the $0.50 is really not what you're after. You're looking forward to bringing forward the $5 per sub forward as opposed to trying to sell that third cloud.

Got it. And as you bring in new logos, does the bundle approach actually make the value proposition clear for them? Or does it feel like a bigger initial commitment?

Haven't got to that point yet. But like I said before, the commercial arrangement is such that they do not have to buy it all at once. So if they aren't convinced that Agentic workflows are for them at the moment, they can start wherever they want to. They can start with just deploying our premises systems with EXOS. They can start on the access side with AXOS. They can take one of our clouds and start there.

So inevitably, we aren't gated by them having to buy the whole package initially. They can start where they feel appropriate. And you've seen that in some of our prior letters where we try to outline for folks those platform journey. And you typically see that they will start with support cloud. And ultimately, as they move forward from support cloud, they're then going to add managed services on to them. And so we would expect that a customer can come in and start their platform journey at any point.

Got it. So maybe changing gears a little bit here. When we talk to investors, the Tier 1 opportunity is one of the more exciting parts that we're hearing now from the investors that we speak to. You mentioned expecting to land an anchor customer this year. Can you help us understand what that engagement looks like? Is the entry point more likely through an MDU or small business where the barriers are lower and then it expands from there?

Yes. So we were moving forward with our Calix One platform in agentic workflow. And Michael went off to Mobile World Congress in March and thought that the conversation would gravitate around the use of AI and workflows with those large customers. And his takeaway from coming back from Mobile World Congress was that these conversations with large telcos inevitably move to MDU and small business. That's the pain point that they are experiencing today. And so we are extremely excited about the opportunity to develop our MDU offering into the marketplace.

We see it as a significant TAM expansion to the work we're doing today. It's another $10 billion market TAM. And we see that the profile of that business very similar to the existing business inside of the company with our regional service providers. It's a highly fragmented market where it's being served today by enterprise solutions coming down. And our approach will be bringing a consumer-led managed service offering from the bottom that is focused on tenant subscriber experience.

It's approached by the same way in terms of a sales force in terms of a direct model. And so we think we have the ability to double the size of our company by approaching the MDU and the added kicker is by approaching that market is that we can then also service Tier 1s. So if that's the way we go into a Tier 1, as Michael would say, we're going in the side door, solving a pain point, that's what we'll do. Our platform, once installed, makes it very easy then to extend anywhere else inside the enterprise.

So the fact that we solve an MDU pain point, our technology stack is fully stood up. They can then by definition through our platform, move into small business. They can add outdoor WiFi. They can add all kinds of things in matters of days and weeks as opposed to months and years because there's no incremental IT work involved once our platform is put in place. So that's our approach. We've demonstrated that in a couple of locations where we've been that easy button for large service providers. And as long as we continue to show that we are the easiest way forward, I think we'll have the ability to penetrate these larger telcos.

And so maybe can you just touch on the differentiation versus the competitive market? It sounds like there's less of a rip and replace motion if they go with you and implement it. But I guess why are they displacing the existing incumbent...

In MDU specifically.

Yes. Yes. Exactly.

Yes, exactly. Because it's, again, the way that market has been put together, right? They are serving it with enterprise class technology being pieced together. And our approach has been taking the residential package that has been consumer-oriented and bringing it up with our anchor customers like Zentro, who -- that's where they started, right? They were serving the MDU space. They were looking for us to come in and provide them a little bit more functionality to round out the solution. And ultimately, we can provide a solution that is managed at a lower price point than what is being offered today. And through the management, give that tenant experience a better option.

And you might ask yourself, well, why today? Why are you approaching MDU and why didn't you do that 2 years ago or 3 years ago? And the answer is the MDU is a microcosm of a small town, and it takes all of the technology that we have developed over the last 10 years to bring it to bear into the marketplace. A community in an apartment tower, for instance, has our apartments. Well, that's the smart home application. When they're in the property, they want to be able to roam freely. They want to go into the parking garages, the workout facilities, the tennis courts. Maybe this apartment has retail on the first floor, they want to be able to go down to the retail shops and their WiFi just works everywhere they go. That's SmartTown.

The property is going to have paywalls and different networks and that technology comes right out of our small business application. So the landlord property management company is going to have its own network. You're going to have a tenant network. You're going to have a guest network that has retail. They'll have a couple of different networks themselves. Managing all those networks and kind of the one cloud is what comes out of the small business application.

So you're putting all of that technology now together into an MDU package. And so it is here more so because it is there. It's now that we put it all together to have these pieces. And so we have to augment it with a few more pieces in terms of like the landlord app for managing tenants as they come and go. But we're at the point where we can actually sell this application into the marketplace. And so we're excited about being able to do that. So today, we've got tens of thousands of apartments under management, and we see our ability to get to millions in not too long a time frame. So pretty exciting. The economics are similar to our regional profile. So we are pivoting 100% and trying to make a lot of inroads into that marketplace.

Very interesting. And as you think about the potential because you described it as kind of like a land-and-expand motion once you're in, now it opens up the door to other avenues. Like is there a low-hanging fruit there that you think is going to be first up at that in terms of being an addressable opportunity at the Tier 1s? Or is it still a little bit opaque at this time?

The 2 adjacencies from MDU would be the small business. And then after that, then we'll have to see kind of where that evolves to. But ultimately, I think that's the easiest way in is through MDU at the moment.

Got it. And as we think about kind of the broader opportunity there, like what are the key milestones investors should be looking out over the next 12 months?

Well, going into Tier 1 are longer sales cycles. So the next 12 months might be an ambitious statement given the fact that they are usually 18-month sales cycles. But ultimately, it will culminate in greater RPO growth. Tier 1s typically do not like to have their name in press releases. But somehow analysts are smart enough to figure that out and somehow it gets out there. But that's how I think you would see it. I don't know that we would do a press release in particular, unless the Tier 1 was willing to share it, which they typically don't.

Got it. Maybe shifting gears to the international market. You mentioned Google Cloud partnership opens up the ability to do private instances in hosted market. How are you thinking about the sequencing there? Is Europe the first focus? Or are there markets in Asia Pacific where you'd further -- you'd look to? And how should we think about the revenue contribution time line?

Yes, that's great. We love our partnership with Google. Calix will be an orderable line item in the Google marketplace because we had talked about how GCP allows us to create a private instance within Google for our cloud. And if you look at us going into a large Tier 1, they're going to have their own contract with Google. Their volumes are going to be greater than what we have. Their pricing is likely to be better. And if you're talking to a CIO of a large Tier 1, they're going to want to be able to burn down that Google commitment by utilizing these workflows on their own private instance. So we facilitated that.

We will then just be a line item within that Google marketplace. The great news about that is the margin profile turns out to be 100% because large Tier 1 is actually buying the compute and the storage and everything else that creates our PaaS in terms of the cloud operation from it. So as we enter in the Tier 1 market, we are engaging Tier 1s around the world. It probably starts a little bit closer to home here in the North America market. But conversations are being held around the world with large telcos. I don't think one is more aligned with Calix than another, whether you're looking at -- when you say Asia, I'm using more of the Five Eyes as opposed to some of the other countries that may not be geopolitically aligned.

Yes. That makes sense. And when you think about that motion now going into these international markets and trying to kind of spread out there, how are you thinking about the associated investments to go after it?

Yes. We're making those investments in the selling and marketing resources. It's a small footprint to start in terms of people. You're going to have less than a dozen people that would be focused on adding those Tier 1s who have the relationships that are accustomed to hunting in that environment.

Makes sense. Maybe just sticking on that investment. I think you spent 80% of R&D on the platform migration in '25, which obviously constrains feature output. And now that capacity has freed up, how quickly does the innovation engine ramp back up after kind of that allocation there?

That's a great question. We had said about last fall that we would be ramping up forward investing in R&D to make these investments on agentic workflows. We did get the cloud migrated over and the funnel for new features is off the charts, so much so that we're actually looking at doing monthly releases, something that we've not done before. So I told you that last week was our first major release post the go-live on GCP but features are coming fast and furious. We'll do a June release, a July release, another major release in August, all the way probably through November. So sit back and watch, you're going to see a ramp in terms of the innovation engine of Calix as now we're -- all our engineers are focused on delivering on these product sets.

Sticking to Tuesdays?

Sticking on Tuesdays, second Tuesday of the month.

No, makes sense. And maybe just on the migration. You've been transparent that there were a few bumps, particularly around installer workflows. As we sit here today, are those fully resolved? And should investors think of the platform as fully stable and ready to scale from here?

Yes. That's true. I still think we're working inside the new model. And so I've explained to folks that it's a brand-new platform. We're optimizing it. Expected like, for example, the software and services margins to snap back a little bit more. But I think the optimization of the current platform will take a little bit more time than that.

So our current expectation for that is that we would be somewhere between last year's Q3, Q4 margin profile. And as we look to Q3 of this year, we should be setting new records for software and services. Obviously, we didn't go through this pain without expecting to see margin improvement on the overall platform and scalability. And so we should be able to grow from there as we move through Q3 and into the future periods.

Got it. Let me just pause there and see if there's any questions in the room. If you have a question, please raise your hand.

Can you walk through the factors that are impacting gross margins from the standpoint of the price uplift to account for the memory offset by the not having to run dual clouds and where all that kind of levels out?

Yes. So I think I just talked through the service and support gross margins. They bottomed, obviously, in the first quarter, and I think we get to new highs during in the third quarter moving forward. As it relates to memory, that's our favorite topic of the week, month and year. We are being encumbered by increased memory pricing, no different than anybody else in the industry. We have chosen to pass that along to our customers in the form of a surcharge and we did that as it changes, we can then concomitantly pass that on to our customer.

So if it continues to go up, surcharges will go up. If they go down, they will get the benefit of any relief that we see in terms of memory pricing without having to change price. Our estimate for margin impact for the year as a result of memory was 200 basis points to gross margin. Since we aren't margin stacking it, you're effectively pushing through a surcharge at 0 margin. And so that's having that headwind to it. At the same time, we took our revenue growth profile up as a consequence for that increased pricing through the form of a surcharge. Where does it go? I would say, ultimately, your crystal ball is as good as mine. I do not think we're at peak pricing yet in terms of memory. I do, based on from what we understand from our supply chain is that it will continue up into the third quarter and then flatten out a little bit in the fourth.

We do think in sometime in '27 that you'll start seeing additional capacity come online that actually helps up then with the supply. At the same time, companies such as Calix and others are optimizing their memory footprint, shrinking the software to help reduce that price. When memory was $1 a gig, you didn't care actually having an extra gig in your appliance. At $25, it's a whole different story. So you're going to make that change and you're going to reduce that footprint.

So I think those supply-demand dynamics will continue, and we'll start seeing something in 2027. Ultimately, I think this plays out that we'll get to a new normal for memory. In the beginning of 2028, we'll probably start seeing the first bits of WiFi 8. And I see this playing out in 1 of 2 ways. We'll simply take that new normal for memory pricing, embed that into Wi-Fi 8 pricing and as customer migrate from Wi-Fi 7 to WiFi 8, it just goes away. If WiFi 8 pushes out later in time or the new normal for memory accelerates earlier in '27, not what I'm thinking is going to happen, but if it did, well, then we would just simply increase the price on Wi-Fi 7 and put the surcharges behind us.

So this headwind that we're facing here in the short term actually unwinds itself, right? Because ultimately, what will happen is the gross margin, the revenue at 0 margin will dissipate. That 200 basis points will reverse itself. The incremental revenue growth will then contract a little bit. And it will be our goal to just continue to grow through all the other avenues we talked about in terms of BEAD and through the MDU, through the footprint expansion that we have in the business that will oversend that slowdown in revenue growth. That's how I see this ultimately playing out.

[indiscernible] memory pricing and other macro factors might be impacting your customers' cadence of deployment of fiber passing and fiber connecting? Are they -- are your -- is your customer set getting a little bit more reluctant or unable to deploy as much as they would because of component constraints?

No. I mean -- well, let me frame that a little differently. There isn't -- the component shortage hasn't led to a supply chain shortage. We can still produce and supply. The increased price has an impact on what they may buy. For example, if they had a plan to go deploy $2 million worth of CapEx during the year, fiber costs are higher, labor costs are higher. We just increased our equipment costs on them. They may not be able to pass 1,000 homes, might be 800 or the competitive landscape may be such that somebody is coming right behind them and they need to get it done and maybe they go back to their Board and increase the capital to $2.2 million. Kind of hard to say. But these price increases have an impact of some kind.

And so at this point, it's not because of the lack of supply. I mean our #1 priority is for us to be in a position where our customers have a new subscriber to turn on, they have the equipment to do so and start generating revenue. And of course, we benefit then from the recurring revenue from having that new subscriber on the network. And ultimately, we would like to be in a position such that if there's an opportunity to take share, we can. So we are aggressively in the marketplace, making sure that we have adequate supply not only to serve our customers, but take advantage of the situation if we can.

Any other questions? No, I don't see any. Okay. So maybe shifting gears from there and maybe your second favorite topic, BEAD. So I think is it fair to say that the peak numbers that you guys were estimating for BEAD has come down a little bit, and I'll caveat that with, I think some of the analysis that you guys were doing has changed a little bit. But I guess where we're getting a question from investors is around has peak numbers come down? Or has the timing changed in terms of when deployments are occurring now?

Yes. At the end of the day, we've been trying to caution investors not to really get in the stock because of BEAD. We don't need BEAD to grow. BEAD is a nice tail, breeze, tailwind if and whenever it happens. We don't feel any less bullish or more bullish about it. Our estimates haven't changed. Unfortunately, our stock trades quite a bit with BEAD announcements and sentiment, right? Somebody says something negative about BEAD and we're being traded down, and I don't understand that dynamic at all.

BEAD is a favorable line item that will happen in time whenever it happens, we'll benefit just fine. So specifically, nothing has changed from our estimates. Our estimates for low tens of millions of dollars of revenue this year hasn't changed. Backlog got coverage. I think it ships this year. That's great. Other people have their BEAD money slipping out into 2027. So be it, I don't think that affects us at all. And as we're looking out in 2027, nothing has changed from the tops down number. We just what we're working to is looking at what the backlog is, how it's developing. And I suspect that it will continue to mature.

I mean 2027 is a long ways away. They're still working through a lot of different things within the BEAD program. And I said I'm not any more or less bullish on the program. I think we'll just do fantastic as it goes along. And I don't really care when it starts.

Got it. Just on that same topic, though, you've mentioned that if BEAD does surprise to the upside, you'd want to reinvest some of that into sales and marketing for MDU and international opportunities that we just talked about. Should investors think of BEAD upside as effectively self-funding the expansion into these new TAMs with operating leverage still improving on the core business underneath?

Yes. I mean that was a statement of really our OpEx model. And our OpEx model, as we outlined at our Investor Day was one of, as we move into 2027, we will slow the growth of OpEx such that it is a portion of the revenue growth. And as we get to 2028, you're going to start seeing more of the full effects of AI being adopted internally at Calix in terms of productivity.

So the areas that you're going to see us grow is we're going to have to build a little bit more into the sales capacity to attack the 2 new markets that we're entering into, but it's in the context of inside the revenue growth. Ultimately, in 2028, we think the operating margins get into the 20s. That OpEx itself rolls down, drifts down to 40% or less. The margins stabilize and get back into the 60s overall. And so I think that's ultimately where we get to in 2028.

BEAD is just another source of revenue. It just fits into the model. And like I said, I mean, BEAD is a tailwind, tail breeze maybe, in which case, we don't really need it to grow. It's just -- it's there. When it happens, we'll do well.

Got it. And then maybe just on satellite technology as a competitive threat. Obviously, it's a pretty discussed topic. So maybe in the last 30 seconds here, help us think through how you guys think about that narrative?

Starlink is great. I think it actually helps. So for some of our customers, they need a competitive push to help them move along. Why would you try Starlink? You have a crappy Internet experience. If you enjoy your Internet and satisfied with your Internet, you aren't calling Starlink, right? So you take somebody like Tombigbee Fiber, who's down in Mississippi that has 30,000 subscribers with a 92 NPS, they aren't calling Starlink. However, if you're a service provider and somewhere in rural that you're the only game in town and you haven't been servicing your service providers and you suck, well, yes, you should have your margins compressed. You should have an ARPU battle. You should have a Starlink come in there.

Now what should that do? Well, I should invite them to come to Calix and talk about how you can actually improve your game, focus on NPS, focus on subscriber and keep Starlink get paid. So it's really incumbent on the service providers to provide a better service. And if they don't, gladly that Starlink is there to provide that competitive nudge to make them go do something. And the only other place we see it is really around a backup play.

So got an example of where my boss lives in an affluent community, and they had an 8-hour outage. Well, what do you think that affluent community is doing without their signing up Starlink as a backup alternative because they can't live 8 hours without Internet. So it doesn't matter when you're coming from that perspective. So Starlink is a good thing. We embrace it and know all the Internet traffic is not going to space. The economics just don't make sense.

No. Thank you, Cory. I think we're out of time. I appreciate it. And thank you, everyone, for joining us today.

Thank you, everyone.

Good morning. Thanks, everyone, for coming. We really appreciate you joining us here on Investor Day 2026 at this incredible time in our company and also in the industry. So that was the video that we showed at Connections back in October when we were actually on this path to where do we go from a company and how do we actually transform our customers' business. And that's what we're going to talk about today because now it's real and it was accomplished in the end of March. So for today, what we're going to walk through is a number of things.

First, we're going to walk through our mission in the future. Where is Calix going to go and what's the opportunity at? Then we're going to go through that opportunity and identify what do we see as the advantage for our company? What's the platform innovation that we can deliver to the industry. And then John is going to come up and he's going to talk about how do we accelerate customer success. Because as we said to our investors day in, day out, our success is predicated on the success of our customers. When they add a subscriber, we win, when they reduce churn, we win. And it's all about how do we invest to help them do that at a faster pace.

Then we're going to go to a customer panel. We're very fortunate to have 4 incredible customers who partner with us on how we innovate and how we change markets. They're going to be sitting up here, and we're going to walk through the different elements of how they run their businesses and how we're partnering to drive their success. We'll then have Cory come up and do the financial model, and we'll close with a leadership Q&A where you'll have the opportunity to ask us questions. And so that's the agenda for today. So let's start, the mission and our future.

On March '26, something big happened with our company. It's something we've been working towards for 2.5 years, which is the launch of Calix 3.0, which is our next platform. And I, 2 weeks from now, will actually finished my 10th year at Calix. And we've been on this journey and nothing worthwhile is easy. And the journey started back prior to 2017. I joined in 2016, and we started out with a recognition that the company that we were is not the company we needed to be. This is founded with our Chairman, and he talked about this, how do we transition ourselves to our only company. If you go back to where we started, it was a networking company that had a single Wi-Fi product. We have kind of a cloud, but not really, but that was Calix 2.0. And we envision a very different view where we could actually transition ourselves into a platform company.

And in 2019, we made that switch. It was hard for us. It was a huge transition in people. And in fact, we always talked about this, is that we always thought that the transition from a people point of view inside Calix would be relatively small. Maybe 20% of the [indiscernible]. And actually, it was the exact opposite. We found that over that period transition not only from a product point of view, from a culture point of view, we actually changed out 80% of our people, and that launched in 2019. And that was the start of a very different company. You can see it in our financial performance. You see it from a revenue point of view, where from 2019 to 2025, the year that we just ended, we landed at -- went from $400 million up to $1 billion. You also see it in the margin profile. The 34 points when in 2017, I actually go back to 2016, 2015, it was even lower. And so that margin profile has gone up significantly, and that's the power of the platform.

How do we drive at scale, a highly profitable business that customers love partnering with to drive great outcomes? And those great outcomes were evidenced by the end of the 2025 year $1 billion in revenue, great margin profile, incredible cash but also significant innovation. Today, we're going to talk about how do we innovate with our customers to help them be more successful over time. And that innovation cycle, the chart up there, we're going to go through that in depth. And in March '26, we were talking about this at the end of last year. We talked about it at connections with the 3,000 attendees who were there is when 2.5 years of work came to culmination and our elastomers left Calix 2.0 and we're added to the platform. And so I'm proud to say we put this in the investor letter. We talked about this very briefly last night is that all of our customers are now on our AI native Calix platform.

This represents a massive opportunity for us. Obviously, when you talk to Cory, he's going to say the most important thing to him is we're not running to clouds. We're not taking the cost affiliate with those. Thank goodness, we can close [indiscernible] but for our customers, this represents an opportunity where artificial intelligence really comes for, and they're going to be able to join the benefits of that capability at a faster and faster pace, which we'll walk through today. And look, this represents the next start the next period for Calix. And we entered this period from a position of strength on many levels. First and foremost, innovation velocity.

We are an innovator in our industry, an industry that is plagued with speed and lack of innovation, where that we're coming into it and [indiscernible] how we do business and so you can see this in the way that we've been recognized. Over the last 12 months, we've achieved 72 different awards regards to how our culture, how our team members, our products are changing the way we do business, and we'll show you some of those examples. We've invested $2 billion up until this point to get to it. This is not something where you grab out of technology. We hope it works. This is actually a long journey that we've been on for 15 years, like I said, $2 billion invested to get to this point, go slow to go if you look at the launch of Calix 3.0 Agentic platform, it was 2.5 years of hard work. And I will tell you throughout that process, and we're going to talk about innovation velocity, our customers are saying, what's the payoff? Why are you moving me over? How does it impact of my business?

We knew and you cannot walk away from the fact that artificial intelligence is going to change everything in every industry. And so we saw that back in November '23, and we started that investment. So at this point in time, we can send up here on stage with you and say, we have an AI native platform. We can walk you through all the steps [indiscernible] and we're ready to go and take advantage of it. And the last part is we kicked off a strategic relationship with Google. I have stated this openly, and I will say this in multiple times, and I'll actually talk about on some slides, the greatest threat to our service provider customer is Amazon. We were on Amazon Cloud, and we made the decision to actually move off because, frankly, they're the enemy. They're the enemy of our customer. And so we moved over to Google, who's an incredible partner.

They have a great footprint around the world, and they enable us to completely change our business model, one where we're no longer a [indiscernible] company, but go into any market in the world, but also we can actually do private instances that allow our customers, if they're a large customer to host everything internally. And we'll talk about that expansion. The second component from [indiscernible] is everything that we're doing with customers. You hear that when they come up on stage. One of the things we're most proud of is the fact our customers have a deep trust of us. We actually have a 94% renewal rate. We have over [indiscernible] for customers on this next-generation platform but more importantly, if you've been reading about artificial intelligence, what is the most important part of it? It's not an agent. It's actually the insights that you get from [indiscernible].

If you don't understand how a business works, you can't effectively apply AI business to transform it and drive rate outcomes. Last year, we did over $0.5 billion workflows. We run the customers' business every single day. We understand how they run their field service, how they run their marketing, how they run their call centers, and we take those insights, and we are uniquely positioned with this new platform to help our customers understand where will artificial intelligence have a really profound impact on their business on a go forward. And so that 0.5 billion workflows are critical to allowing us to go fast in this next day. Third, talent. We're very proud of our team. And I firmly believe myself that as we go into this genetic era, we need to stop with the value of AI is only how many people I can cut and say, how do I power my team members and really focus on human-centric AI. How do we empower the team members to do significantly more with less. And I'm going to show you some examples of that today.

But how do I actually get that talent to think about how do we use AI. We can go on and be dispassionate about the business and say I'm going to cut this many heads. Or what I can do is say, Help me understand what you do every single day. Help me understand those workflows and how do we reimagine work and go to a different level. How can one people do the job of 10 and expand the business and grow revenue? And so for us, 98% of our employees are adopting AI in one shape or form. In fact, they wrote an article with fast company, which I actually talked about, we ran over 700 packs through the year with our team members to actually explain and explore how we use AI. And John will be coming up our COO, and I'll walk you through how our IT organization is working and partnering with our employees to identify how do we reimagine work.

On top of that, over the last couple of weeks, we had announced with Fortune. There were top 100 company to work. We really do believe in this next stage, we want to continue to attract great talent. As we optimize our business, which we'll continue to do, we want to be the company where people want to work. And this award, amongst the other -- the 48 other cultural awards [indiscernible] last year are important to us and help us build a great team to do great things. And last, Cory is going to talk about this at the very end, our financial performance through what I would call very tumultuous times over the last 5 years, goal of pandemic and all the other things that are going on, we continue to execute through it.

The margin expansion, the free cash flow and the way that we return value to investors in the form of share buybacks is something that we do methodically and execute every single day, and Cory will talk about that at length. So we enter this next stage of our company in a position where we can invest. We have a great team, and we have great partnerships with customers to do wonderful things. And I cannot talk enough about how important is the relationship we have with customers. We entered this period with a deep trusting relationship with around how they view success. What are the things that they need us to do to change industries, change the way they go to market and help them build their business. And in fact, we believe in this so passionately and we see the opportunity to move faster than we changed our mission statement. We evolved our mission statement. And the mission statement is now this, enable customers to transform operations.

We moved it from Simplify. Why? Because Simplify doesn't capture what's going to happen in AI. The future of AI is not I take an existing workflow and I just pop in a couple of agents and I simplify things. That's not the future. The future of artificial intelligence is, here's the outcome and to achieve an outcome, I have 2 choices. I can do what I said, I can simplify it and drop some agents in and eliminate some head count and make it simpler or I can take that outcome and completely reimagine how I work. And in fact, I'm going to show you a couple of advertising examples that we did, which allow you to understand just how transformative this is from the scale point of view. And so we're not in this next stage going to be simplifying. We're going to help them transform how they operate sales, marketing, call center, all those different component parts through the power of AI.

The second part is accelerate experiences. We've been delivering experiences since we launched the second-generation platform back in 2019. But now the big impact is how do we accelerate that at a rapid pace. Not only from an innovation point of view, with [indiscernible], our Chief Product Officer, is going to talk about, but also how do we accelerate the adoption of those experiences, which John is going to talk about because at this point in time, we only have an attach rate of about 45%. How do we get that to 100%? And Scott, from Tom Big's going to be up here, one of our customers, and he's going to explain why hasn't he done our marketing product. Well, he didn't have the capacity and capability in his market to hire someone to do that.

With artificial intelligence, we now have the capability to take a junior person and turn them into an expert marketer and transform and accelerate their delivery experiences to differentiate in their market. And so you can see the outcomes. We put these up at connections every single year. We have an incredible list of what our platform does versus competition to enable the success with customers. You see 90% market share for Berlin, 73% ARPU growth, 60% reduction of OpEx. This is what we do. But this next stage is going to be a completely different scale with regards to what we can do for our customers. And I'm really proud to say that these 4 customers will be up on stage to actually give us an insight into how they see the next generation of where we're going. An example would be allow who deployed 23,000, now 23,000 students in Lincoln, Nebraska have SmartTown have the ability to get free WiFi as they roam around town every single day regardless of economic scenario. That's one example.

But then the other part is, at the same time with Brad and his team, we did 14 proof of concepts around artificial intelligence in partnership with them in the last 4 months. So how do we really accelerate their outcomes? We'll have Lumos up here, Tombigbee, and Bluestream Fiber, who's a leader in smart MDU, who has been with us over the last year, building that product out and telling us, here's the legacy vendors and what they provide in an MDU scenario Here's my business differentiation and how I need you to change my business model and drive great outcomes. And so they'll be up on stage with me. And look, this comes at a critical time for customers. Competition is ramping up. It's getting faster and faster. So we know what's happening with regards to satellites. Starlinks launching, they're adding more capacity. They're throwing capacity everywhere. Amazon just bought Globe SAR, $11 billion investment because they were falling behind. And so they know they want to get satellite capacity. And really here is a happy Amazon customer. I am, right? If you don't have your hand up, you're just lying, right?

So I'm a happy Amazon customer. Every day, every couple of days, there's Amazon boxes at my house. They have a great NPS. They're convenient, easy to use, and they have a direct consumer relationship. Amazon Aero is a relationship with the customer and all that's going to happen is that for everyone who resold Euro in the past, is a quick switch of the wand in the back when Amazon comes and says, here's prime broadband. And how do you stop that? We see that as a huge opportunity with us with our customers as that competitive pressure ramps up for them to get it that if they don't have the best experiences in their market, they're at risk because they can just be swapped. And so this experience opportunity is one that we've been pontificating about for 5, 6, 7 years, maybe a decade and now is the opportunity because the competition continues to ramp up and that we believe that for Calix as we're the leaders in this market with regards to experience. For those who haven't gone on the bandwidth, they better get going faster.

And so -- and that's going to be John's mission to make sure that happens. The second part of that is commoditization continues to be rampant. Everyone uses this fancy word called Convergence. Let's be really clear. You know what convergence is, Convergence is a bundle. I was at [indiscernible] mobility. I understand what a bundle is, right? It's nothing fancy. It's just broadband with mobile at a cheaper price. That's commoditization. They don't want to use commoditization, but that's what it is. So again, if you're not differentiating in your market with regards to experience, the customer has a real struggle. This is what we do really great. And we'll talk about how we expand that out in a moment. And then last, look, artificial intelligence, anyone who doesn't believe this is going to change human kind is missing out. It's going to change every facet of life. It's going to change manufacturing, it's going to change every industry, and it's absolutely 100% going to change our industry.

I gave the example of Tombigbee Fiber. So Scott told me, I talked to him just before Christmas and said, he said the biggest reason why I haven't used Engagement Cloud to deal with my customers is because that don't have the capacity and capability, the people in my market who can do that. And I can't hire that person because they're going to be in a big city and how do I bring them and what can I afford them? That's why I haven't gone forward. And so transformation and marketing is a great example where everything is going to change, and customers need to be thinking about not only -- I'm not going to be running campaigns in a linear approach anymore. I actually have to think about how do I 10x 30x, 100x, how do I run 1,000 campaigns with one person a week? Everything is going to change and artificial intelligence is going to make that scale. And I have a simple example to share with you.

So if you've been following our company, you know the [indiscernible] videos that we do. And we've been doing these [indiscernible] videos for 5 years. And the concept was we would do the video content with our decision council partners and [indiscernible] we put the content out. And then what we give it is we provide that to them and then they can put their brand on it, and they can use that as a marketing campaign against our products, right? To build a Jerry D campaign was generally a 2- or 3-day shoot, and I remember the very first time I went to it, it was tractor trailers. It was people building out, like there was like 50 people there and our cost over 3 days was $1 million to do 3 videos, a huge cost. And that's excluded like there's all the preproduction time. There's -- and you've got all the union workers, all those kind of things, right? A huge impact. What I'm going to show you is artificial intelligence that was done for allo in a couple of days for $20,000. Shoot the video.

[Presentation]

That was done with $20,000 using artificial intelligence and win in days. This is a simple example of how profound of a shift we face. Every job is going to change. And as I said, we can think about simplification, how do I optimize the workflow or I can think of, here's the outcome, how do I completely change it. In that scenario, we're also looking at technologies where An individual can run 4,000 banner ads direct out a single person where they used to actually only be able to do a small fraction of that on a monthly basis. And they can do that in a day. There's a profound shift coming in all industries, and we're going to be at the forefront of it. And so as I think of this and go into 2026, there's really 3 discussion points.

So the first is what's the Calix advantage? Where do we sit currently and how going to make the most of this opportunity for you as investors and for our customers? The second is our platform innovation, how do we actually speed up the innovation that has a lot to do with the new platform that we put out. And then third, core to our business is how do we make customers? If they do not succeed, we do not get paid. It's that simple. If they lose a subscriber, we lose revenue. If they win a subscriber, we make money. It is customer-based success, which is our revenue and so how do we accelerate that at a rapid pace and that was what John is going to come up and how he transforms his business both as a customer success organization, but then also do we, Calix transform ourselves internally in everything that we do.

So let's start out with the Calix Advantage. First, I firmly believe that success starts with people. I believe in human center which is how do we empower really great leaders to understand what the business opportunity to leverage these capabilities so we can change our business and grow at a rapid rate at a lower cost. And so I'm proud to say we have an incredible team in place, and that team leads an organization who, as I mentioned before, has transformed themselves into an award-winning culture. So if you go back to 2019, our glass doors were 3 out of 5. We were not attracting talent. We weren't getting the best and the brightest. We weren't having people from big companies like Meta and places like that coming to Calix now they are. And the reason why is because we really embrace empowering incredible people to identify what to do with these technologies and how to take our business to the next level.

We've kicked off with a strong year [indiscernible]. We're really proud of the fact that Fortune recognizes us [indiscernible] one of the top 100 places to work against some of the big companies in the world. And we believe that, that ongoing investment in culture is going to be critical to us because not only are we going to help our customers transform their business, which there'll be a lot of resistant change. But inside our company, we will change as a company significantly and John is going to talk about that.

The second advantage that we have is how we innovate. All of our innovation is done with customers. We work really closely with them. We have understanding into how they run their business, and we've been doing that for 15-plus years. We've invested over $2 billion in the platform and we have those 0.5 billion workflows where we can actually parse out everything that they do. And in this next phase, what's going to be critical is that we think the right investments in the right places to drive the biggest outcomes. We process at this point, over a petabyte of data every single day. And that comes from our customers trusting that we will help them understand that data so they can run a better business. So that massive investment from an R&D point of view has gotten us to this point. But the other part is that, that R&D investment is going to pivot significantly. If you look at our innovation cycle over the last few years. These are the number of futures that we released every single year since 2017.

So prior to being a proper platform company, 77 features, 180 features. We were a hardware integrated company that didn't have a platform. Then we released [indiscernible] what you see a massive acceleration of the rate of innovation up until a peak in 2024, where we build from 406, almost 1,000 features coming out. And so what happened in '24? Well, in November of 2023 is what we decided as a company that Shane came into myself, and then he and I went to the Board and we sat down and we said artificial intelligence is real. It's finally at a place from a security and privacy point of view, where we can leverage it safely with our customers and this is going to completely change every industry, and we need to start investing aggressively now because if we don't, we'll be behind.

So in November 2023, [indiscernible] a massive project to actually start building out the third generation of our platform. And Shane will walk you through the levels of the platform, but that kickup are embedding AI into everything that we do, becoming an AI-native platform. And what happened? Well, as we progress through 2024, we got those projects out from a features point of view. But in 2025, we slowly pivoted our entire organization into one thing, get that platform out. About 80% of our resources through 2025 went towards getting the next generation of the platform out which meant you saw feature velocity drop pretty significantly. And what's really great is out. It was hard nothing worthwhile is easy. It was really, really hard. But if you want to invest for the long term and change the way your business runs for customers, you have to think about this way before.

The companies who haven't done this, they're already way behind. And there -- if you haven't been watching the pace and the acceleration and the speed of change that AI is embedding into every industry, if you haven't made those expense, you're going to follow behind incredibly fast. We are very fortunate that the product team identified this opportunity back in November and the threat. And we went hard and now we're out. We can use any LLM. We have knowledge, we have all these capabilities ready to go. And so it was out last customer was ported on March '26. And for us, with that AI native platform, we are now going to see as a return to massive acceleration of our broader business. And so in the past, we led the market. We had the market as we opened up new capabilities. We've led the market from awards point of view to go back to when we were a pre-platform company, again, 2 awards in the entire year. It's pretty sad.

I know we were really happy with when we [indiscernible], right? So this year to date, we're already 5. We won 18 awards last year. People in the industry are recognizing us for the innovator that we are. This is nothing. What we're going to do in this next phase with AI is going to transform how we innovate. We have the trust with customers, but the pace of innovation is going to be mind boggling. And we see that from all the news that comes out every single day, and we're going to be at the forefront to take advantage of it. And so what is that platform? I'm not going to dig into that. Shane is going to come up and he's going to walk you through what the Calix [indiscernible] platform is. But at its core, is a platform that takes customers and their teams.

So everything you do in sales, marketing, network operations, field service, call center, and it aligns them to the markets that they serve on a single platform. It allows them to serve consumer small business, MDU. And as I'll show you in the TAM expansion, new markets, including hospitality and other areas with no incremental complexity. It just works. And so that's how we bring it all together to help our customers do for things: grow revenue, add subscribers, reduce churn and in the end, deliver the best NPS. And that's one of the things that we're constantly proud of. People are shocked when we hear from customers who are going to sit up there. Tombigbee is going to sit up here and they can say, I have an NPS of 92 for the last 4 years. [indiscernible] Al can talk about the fact that they're NPS in the mid-70s, and they've had that for years. That's why they have 65% market share in a lot of the markets that they serve. NPS drives revenue. And so for us, the platform has allowed us to expand revenue.

Have we just stayed in North American regionals? It wouldn't be at the revenue level that we're at. More importantly, we won't have the opportunity that we have in the future. So if you look at the revenue growth, it correlates to us releasing products. We launched Smart Home. We won Verizon, which was the launch of our next-generation platform with regards to networks. Back in 2019, you see revenue accelerating with the expansion of SmartTown, which is our role and the small business. And as we start on Calix 3.0, you're going to see a massive expansion of the potential TAM for our company and where we invest to us to go after new markets. First of all, it starts out with agent work for everything we're doing with AI, but there's a massive expansion or an MDU, hospitality, events and all these other markets. And for us, they're adjacent markets, they don't require -- they require an expansion of our product, not a complete rerate of our products using the resources of our customers.

And then on top of that, we have the ability to go after global service providers using Google Cloud as a partner. And so as we think about the expansion of TAM, it's a huge opportunity ahead for us. If you look back to where we started out, it's kind of like a is was the TAM in North America, right? And as we take the next step, global Tier 1s, it's a software-only play. So that's the TAM approaching $10 billion. And then just in North America, the multi-dwelling unit apartment buildings, just that alone is a $10 billion TAM [indiscernible] North America. That includes hospitality, that excludes events, that excludes all the other areas. There represents a significant opportunity for us to disrupt these markets and grow at a rapid place. And so with that, then pivot over to Shane sitting at the back and walking up here faster.

So Shane, our Chief Product Officer, Shane Eleniak, who's going to walk through our platform and innovation and how we make that real for customers.

So if you think about what we've built from a perspective. It's a vertical platform. It starts with appliances and operating systems, the data path, delivering the actual services that subscribers consume. It's also the source though, if you think about the edge and you think about the subscriber edge, it's telemetry, it's instrumentation, it's understanding quality of experience, so understanding unmet needs, behavioral analytics. And it takes the form of a petabyte of data every day for us that we then want to look at and pull out insights traditionally with our second-generation platform and would have been insights that we'd be talking about. But we have context. We understand the workflows. We understand the actions that need to take place at the service provider from a marketing perspective, from an operations perspective, from a customer support perspective.

So we've evolved from thinking about insights to thinking about knowledge, thinking about machine learning and embedding LLM, it's context. What's the action that needs to happen? What's the material event? What's the unmet need and the expansion from data to knowledge to say I not only know what's important, what's urgent, what needs to happen. I understand the workflow. Traditionally, we would have built static workflows with dynamic data. Now we're able to go to workflows. Those are outcome-orientated dynamic workflows. So if you think about the stack, it's the sources of the data, the data itself, the knowledge, it's these agentic workflows. And ultimately, what we're trying to deliver experiences to the consumer, to the small business, in the MDU, WiFi roaming, venues, events, hospitality.

Those are the types of experience people want every day that differentiate, but ultimately meet the things that you and I want to do every day as a consumer, everything that a small business owner wants to do. And that's a vertical tech stack. And that's really the big change. If you think about platforms in an Agentic world, they're vertical. Traditionally, what we live through with cloud and SaaS was very much horizontal. I'm going to focus in on a horizontal and do one workflow really well. Now it's about understanding the outcomes, understanding dynamic workflows, actually having that contextual knowledge to understand what needs to happen and in what order and what sequence. So that's all platform and the hard work that we've undertaken in the last 2 years to evolve from a second-generation cloud to an agentic-orientated cloud. There's a high aspect to products, and there's a what and a win and a why aspect to products.

This slide really speaks to the how. How you do things matter. The shift from systems and OS to a platform and a platform effect that you're building on top. But it also speaks to us becoming a cadenced company and doing time boxed to agile and looking at a shift left in how we did testing in automation and simulation to drive velocity of features. And that's really the key goal here was to get a platform in place with Calix 2.0 put the right how process in place to be able to drive innovation, to be able to go faster. And really, what we're seeing now is a shift back to being able to focus on features after having built our cloud. And really that previous slide was a large shift away from a traditional SaaS-based cloud to an agentic-orientated contextual knowledge, dynamic workflow orientated AI platform. And that's the work that we've done. But that says now we're able to focus products back on doing features and use cases and getting back to that velocity. There's also an element to this, though, and we'll talk about on the next slide about how you build products, the product development life cycle and how AI changes that is also an aspect of this.

So we feel very comfortable that we have the right how process this is in place that we built with Calix 2.0 to be able to innovate and go at scale. But we think there's a rich opportunity to go even faster with Calix 3.0. The second part of this is working with customers. Having how going fast doesn't help if you're building the wrong things. So a big element of this that starts way at the left in time is getting the use cases right. Understanding first use case, and that tends to be our focus. We'll pick a subset of customers with a use case that we understand really well. We'll work with them as our anchors, and we'll get that one right. You didn't go through a phase, though, where you're not only hardening that use case, you're expanding it because now all of a sudden, you've got a market. So you're going away from a small group of customers to a broad market and all of our customers.

So you're doing use case expansion, your hardening, but there's an ecosystem around us. In this case, this is smart MDU. You start to talk about billing gateways and different commercial models. I don't want to do a bulk. I want to do pay walls. You start talking about tenant management systems and all the hooks and handles that start to fit into a complete solution and ultimately a platform grade solution. And we do this with everything we do. We work with our customers, we work really closely with them not only to understand the change in the marketing workflow or the operations workflow or the support workflow, the subscriber need, how does that experience need to change.

But what are their broader ecosystem we need to work with? Where are the hooks and handles? What else do we need to work with? What are those business systems? How do those change? What are new business systems as we get into that we need to work closely with? Ultimately, because they're delivering a set of experiences to the property management company and ultimately, the tenant, right? The other thing that's changed though is everybody is now talking about vibe coding and code generation. We have a huge opportunity and we're leveraging it. Today, we're now able to do rapid prototyping.

Before what we would have done is big wire frames and sat there with slides and sat there and said, hey, what do you think about these [indiscernible] Gemini input. Now you can actually code a prototype. It's grounded in your code base. It's grounded in your data model. It's actually accurate. And I don't know about you, but I work a lot better with a [indiscernible]. When I can see it and I can see what's going on. It's really easy to give feedback going. That's good, but this would be better or you missed this or yes, but if you thought about that. So now all of a sudden, you're shifting that prototype to the left. You're getting those first, second, third and fourth use cases, solidified very quickly.

You'll lose the lost in translation between product management, talking to architects and designers who are going. But I read your Epic, but is that what you meant? So now all of a sudden, we've got visual. We've got rapid prototyping, talking to code generation tools that are generating the tool. But this is a CICD pipeline. I need tested code. So how you do simulation, how you do testing, how you actually get the code that's deployable that's stable, that's of high quality is really where we're focused and we're able right now to leverage these tools in our product development life cycle.

So we feel very comfortable when we say we built an engine on how to the -- how aspects in Calix 2.0 today, but the tools that we have today are significantly better than the tools that we had in 2024. So we feel really good as we shift now into the platform that we have the platform place to do this. We've got the processes in place to do this. But we also have better systems and tools than we did 2 years ago. And whether you look at kind of what's possible in the video world, what's possible in the tested code generation rapid prototype world is pretty cool where we've evolved in the last 2 years.

Okay. And then if you talk about, okay, you've got a great platform. You've got a great set of processes on how you do it. You work closely with your customers on what and why and when and getting sequencing and priority right. Tell me more about how you got here with this agent workforce. So effectively, we started as we would with everything, simple use cases, POX, close customer relationships. And we started packing, hey, we've got ideas. We understand workflows, we do 500 million workflows a year. We've had 5, 6, 7 years of run time on platform with Calix 2.0 understanding how people use these workflows but when you shift to Agentics and dynamic and outcome orientated, it touches multiple personas. Everybody is involved. Marketing has a role to play, engineering, operations, support the needs of the subscriber, all of that changes.

So we started testing our proof-of-concept. Q1, we released assistance. We have -- we want to help people. That was really our first agent. Q2, though, you start to get into these dynamic workflows that touch multiple people. You start getting into upsell, cross-sell, which is a marketing workflow, but operations needs to know about it. Engineering [indiscernible] customer support needs to know about it. It touches subscribers in different ways. Same thing with churn workflow, same thing with acquisitions. So you're going to see us starting to see acquisition workflows, upsell, cross-sell workflows, churn prevention workflows, loyalty workflows. Then you start getting into customer support workflows, how do I help the customer shortest time to happy customer. And then you start getting into subscriber workflows and how do I interact directly in the subscriber with the app.

We start to deliver those in Q2. And then you're getting into call center workflows and network operational workflows. And those are going to be the swim lanes that you're always going to see us delivering in. We're putting the first ones in people's hands in May. We then augment with additional ones. We start moving closer into the home device anomaly. So we move from, hey, we understand the network. We understand the demark of the network, now we've moved into the home itself and into the business itself when we're trying to understand anomalies there. We start getting into critical incidents, which get into notifications, how do you interact with subscribers and tell them, hey, there's an issue. We're working on it. Here's the meantime to when we're going to resolve it, how do you continue to update them, then you start getting into optimization workflows, AI-enabled network optimizations, planning type of use cases.

Upsell, cross-sell also starts to evolve from, hey, I'm doing customer support. The issue is really the person has the wrong package. They're not doing anything wrong. The network is working the way it is. They've got an unmet need. How do I upsell cross-sell at that point of contact also. So that's with the types of things that we start to deliver in our August really.

Okay. With that, I'm going to turn it back over to Michael. So hopefully, I've given you a sense of -- we have a vertical perform. We understand the how. We understand what and why. We work closely with our customers. And we're at a point now where we're delivering dynamic workflows that change the way people work today. Thank you.

Thanks, Shane. And at the same time, he's going to transform his organization. If you didn't see on the last slide, the productivity gains as you get from rapid prototyping from an accuracy point of view and a use case point of view, but also cogen is a simple thing. It's actually how do you go from use case to test to code and product out the door that transforms the customer. So this is the road map and you're going to see this accelerate at a rapid pace. And when I was at Mobile World Congress, this is the first time we've really been there, we've got invited up to do a keynote. And so the keynote that I did was what's the AI future [indiscernible] AI architect session that we did this in and I talked through how do you rethink how we work marketing. And in this scenario, this is one we've been talking a lot to our customers about because they generally and our -- they buy our platform end-to-end.

And the scenario becomes I have a subscriber agent and agents are task-based entities. They can think and reason and they can evaluate data. So they look at a subscriber. They see things that are going on. Working is optimized and then as the person moves from inside to outdoors, they move out into the backyard to do some work, they see something bad happening, which is their experience gets challenged. And so that is a WiFi degradation because they don't have coverage in the backyard. Maybe they want to go work in the Pagoda. But then at the same time, the subscriber agent can grab insights with regards to that customer, taking external data and look at, for example, they know that there's a pool in the backyard because they have a warranty for a pool pump. I actually am registered somewhere with Pentair for my pool pump and so you can do that correlation. I saw them leave the house.

I saw the degradation, oh, by the way, they have a pool pump, chances are they have a pool in the backyard. This becomes important context in the marketing scenario. But this is all automated. In the past, can you imagine an individual trying to go through and sort through all the external data to say who's got a pool pump and then who has an issue. That's a huge amount of work that is unfavomable. You cannot scale that. So in this case, it's genetic work doing it at an individual level. Then it passes on an operations agent says, in this case, Brad and his organization also sell mobile phones. They sign up for an allo MVNO. So it also looks at that home and sees the devices. Well, there's 2 iPhones and those are on my MVNO. There's 2 iPhones that aren't, okay? That represents a mobile opportunity for Brad and his organization to upsell, there's 2 devices to be one there.

More context. It then passes over that workflow to a service agent who does all the correlation to what's been going on in that home over the last 30 days. It's moving back and out. And so sees things like he's been working sales force and different applications without breaking privacy, we just want to know what they're doing because we want to optimize their experience and then it bundles that all up and it passes over to a marketing agent, which builds a campaign. And in this case, the campaign is contextual. It actually grab and as we showed with that video, there's no reason why you can't go create video based upon someone hanging out in the backyard at a pool doing work. Put the video. It builds that context. It puts the offer in place and then it chooses, based upon their social media preferences, where should I surface that ad.

So I buy an ad from 8 to 10 p.m. on YouTube. Should I add surface it on their mobile app if they're using [indiscernible] branded by the customer. And then when they go into [indiscernible], they click to buy. Doing this type of a workflow at an individual level by subscriber is impossible unless you actually scale it with AI. Now that's how our regional customers would use it because the fact is they use all of our tools. As we migrate into different markets, Calix from an Agentic point of view becomes a different entity for those companies. We become a toolkit. So in each of those scenarios, we recognize that they have all kinds of other systems. Brad and his organization use other systems because they're a big scale company.

So whether they're using Salesforce or Adobe or pick your system, what we can do is as they go through their workflow, we can then become a component part. We're in smaller customer, we're going to do it end-to-end. In a bigger customer, we're going to become a component. So in that scenario where we do cross-sell, upsell into the pool and drive the data. We may want to identify as the pool, we may not be. That might actually sit with another agent in a data science AI agent, and it's actually servicing that data. We provide the subscriber insight. It goes into a workflow and off you go. So our role changes and Shane and his organization from an architecture point of view, part of our rearchitecture was also to ensure that we can serve any size of customers.

How? Because in the past, it used to be APIs, which we're great at. The new terminology for Agentic is more context protocol, MCP, which is you put a server in front of it and we share our context without moving data or A to A, which is agent to agent. So it's our agents talking to their agents as part of a workflow, which we've -- we're very good at an industry standards organization. And so for a company that has all these back-office systems, no problem, we can add value from an AI point of view with a in a very different way. It represents a significant opportunity. The other part of this Agentic workforce is that it also gives us an opportunity to help our customers rethink how they do a -- this is a transformation of your business.

This is also a transformation of business in a different way and to understand how we help our customers go faster, accelerate differentiation and the experiences that they deliver into the market win more subscribers. I'd like to invite John to the stage, our COO, who's going to walk through how we accelerate their customer success. John?

Thanks, Michael. [indiscernible] So I want to share, first of all, how do we think about success, right? And at Calix, we deliberately align with our customers' life cycle around build, around launch and around optimize. What does that look like in practice? In practice, that may mean that we consult around our strategic funding. You have questions around that and maybe around setting up their Wi-Fi or their network design. It could be around subscriber activation, right, or projects driving market growth. And lastly, it's around optimized, right? What's the most important thing is this performance or accelerating growth or thinking about what is that subscriber experience. And we'll get engaged with customers in the subscriber experience that NPS scoring, we actually help our customers drive that. And why do we want to know what that MPS is not just because it's a great score, but there's opportunity in that score.

How do we continue to improve the experience and change that relationship? We're relentlessly focused on 3 things with our customers. Helping them to attract new customers, retaining their existing customers; and thirdly, growing that customer base. And when our customers are successful, we're wildly successful. And this is an example in terms of this visual in terms of our ability to help our customers grow, right? Because we win when our customers win. And innovation isn't just a product focus at Calix. It's a company focus. And how did we help our customers grow their attach rate. Well, we listened to them. They came to us with challenges. And we've created a marketing acceleration program, which helps them think about their offer strategy think about persona-based offers and really change how they go [indiscernible] to really focus on that experience. And we continue to them. And they said, "Hey, we need to help in this area." And we create a workforce transformation program and that was intended to help them attract talent, retain talent grow their existing talent because they have an aging workforce that they needed help to make sure that they were ready for the future.

And then they came to us 2 years ago and said, "Hey, you know what, as we get into smart business, it's a really different selling to, and we're not that experienced in that. And so we created a sales acceleration program that thinks about what does the sales structure need to be? What is the incentive compensation need to be? And how do we create a sales culture, right? And so we're doing these things, we are partnering, right, with our customers in that continuous life cycle. Now the most exciting part is around AI opportunities in front of us. And we can talk to our customers in an academic way, but that's never my style. I want to get out there and be a practitioner. And so everything that we are going to work with our customers, we've done already, right?

We've activated our organization. We've done a great job of driving adoption, but not only adoption but actually building agents, right? My team, in particular, I'm pretty proud of that, has a majority of 400 that have been built out there that we're using to transform how we do certain activities. And when we did those, we found huge improvements in how [indiscernible] to go and build something for a customer, right? We really automated that and as Shane and Michael both talked about, the biggest opportunity is around that end-to-end workflow. And we're doing that internally as well as we're thinking about our 3 biggest workflows and how do we reagent those so that we can apply AI not in an episode or on a functional basis, but across the entire thing to eliminate rework, right, to automate processes today to reduce the number of handoffs, and that's fantastic. And then none of it matters if at the end of the day, that you can't actually measure and get a realization.

And we've done that. We had a project going on around our supply chain, right, next-generation supply chain and we're looking at signals to build [indiscernible] interested in that, Jerry will be up here at the supply chain. He can talk to you all about how he looked at the signals, automated that and reduced the number of people that he needed to go actually run his business which was fantastic. Then we're going to now take all of these learnings and we've built them into something we're calling the AI leadership playbook. And that's from our first-hand knowledge of how are doing this? How are we governing? How are we activating pilots? How are we doing the change management to get everybody behind this? Because many employees could be threatened by this.

But we think there's a huge opportunity. Why am I excited about this? Because we are working with our customers on ideation of things to do. Their biggest challenge is capacity, right, as all of us have a capacity problem. And so how do they do more with less? I'm excited about how our agents are going to fill that need so that when we come up with an idea and they want to do this, we can get them all the way there. And now there's an agent to do the work for them to go faster.

With that, I'm going to turn it back to Michael for the customer Q&A.

Go ahead. Queue the video.

[Presentation]

[indiscernible] Thanks for being on stage with us. All right. So let's start out with -- we've been talking to all of you about what's possible with artificial intelligence, right? So why don't we start right here, Brad, because back last year, you came to me in May and said, "Hey, my team wants to go build out a whole bunch of things. We really need to start collaborating on how we can partner together. So why don't we start out with kind of your view of what the potential is ahead and how we partner together?

Well, what's been done several times between our organizations is we really have a mutual challenge, opportunity problem, whatever you want to call it and so what I found is we were working separately. And every time we work separately, we waste money. And so what we talked about and got everybody in the room and said, here's what we're developing, here's what you're developing, Hey, we have other systems and other complexities. But what we did is we came to view. And with that, we switched to the Google Cloud.

That was a huge lift for us. But now what we see going forward is not just the first 14 things that we talked but now the ability to work together at a true partnership or stakeholder level to solve problems because it can't be that the network performs better. It's -- every piece of our customers' journey has to work with colleagues, with some of our other team vendors teammates and great solutions that I can't even imagine. I hate to say it, but what our teams are working on was never imagined 2 years ago.

Right. So Brian, yourself, like you -- how do you think of it from a competitive point of view and where we're going to go as the next stage because you've gone through some big transformations in your business, too, right?

You really have. And it's been the partnership with Calix. Calix allows do this. And I'll be the first to say, we're not really the first follower. We're more of a fast follower on the AI. We leverage the work that the brand and you guys do leverage the things that you Team does. But it really helps us focus on our business outcomes. And I think what people don't fully appreciate, you talk about our NPS score of 84. 4 years ago, it was 77%, 78%, but we've grown -- I mean, we're building houses that we worked 4 years ago. We've added financial number of subscribers. We have not added new customer care agents to grow over and not have to add customer carriage just because of what we're doing and what you're doing in the background.

So what I would say is we're still evolving. We're still learning. We're just kind of getting there, but it's really your focus that's allowed us to focus on business outcomes.

So your point is, is that Brad and his team was going to start doing the AI. So he partnered with us in that way. Yours was, I don't need to invent it. I just want to actually keep my head count flat and expand profitability and grow NPS and customer and then yourself, the conversation was around marketing that we had at Christmas, right?

Exactly. So we're the ones that -- with the smallest ones on stage, you may not look to be the small [indiscernible] so we're coming from a world where I have 6 CSRs and one marketing girl who just graduated on this a year ago. She's wonderful, hope she's watching. She's doing a great job. But we can't pull the data. We can't gather the things. We couldn't do any of this without you. So without...

And that's why you didn't do Engagement Cloud, you couldn't actually -- you're doing very basic marketing because of the fact you didn't have the capacity to do that.

We don't have the time. And so this is a game changer from us. So we're the opposite where some people are afraid of downsizing because of AI, I view it as upsizing. I'm going to take my [indiscernible] because of your help because of your partnership. We can do that, and we can reach people in ways we ever could. And as there was a gentleman ran for [indiscernible] running against somebody who had been an officer in the military and he said, all you [indiscernible] him. It's all your profits out for us. So I represent [indiscernible] your customers, of the 1,200 customers, you've got a lot of people who look at like me, [indiscernible] is amazing. These guys are awesome. But we're the meeting potatoes of what Calix does, and this makes us have the resources that [indiscernible] amazing. That's a game changer in roll Mississippi. .

Well is the democratization of AI, right? We are in this phase where technology is going to democratize at a pace we have never seen before where you not the big guy doesn't necessarily have access to things that you don't have access to. And that's, by the way, the same for Calix because all the LOMs out there are now readily available to us as AI becomes the biggest commodity out there. We have the opportunity to use the latest and greatest swapping in and out and bring that capability to you. Now Gavin. So your conversation with us is a bit different when we talk about what we're doing there because we built product together. So why don't you walk through kind of how we actually partnered with regards to the MDU market.

Yes. And look, our story with you all is really the last decade and almost started out with our business, a little bit different than the folks on this stage. We offer community solutions. We do long-term contracts, we work with HOAs or [indiscernible] new master planned communities. And we first started to see as you look at that product curve that was up here earlier around some of the innovation. We're able to leverage SmartTown for, you think, master planned communities, the pool, the gym, the pickleball court. All of that goes with that and really sort of have that ubiquitous as we started to really verticalize and focus on some of our different segments, MDU and you saw that TAM that's up there, we think that's a major growth engine for us as a company.

And what we really try to do is step back and say, think about it these are long-term contracts or truly partnerships with these communities, what is going...

[indiscernible] to her 10-year contracts, right? Like [indiscernible] and a lot of those HOAs, you're doing like long term -- it's not a consumer who's deciding month to men, right?

No, they want to know long-term customer experience. I want to know about what I'm getting today, but what am I going to get the end of that contract and think about a 99% renewal rate. This is a long-term relationship that continues to build on it. So we work with Calix, we really focused on the SmartMDU solution, and we've now rolled this out through a number of our communities, a number of our backlog, and you saw the scores in the video, some of the early communities are 4.99 out of 5. So even every single one, it's been a 4.9 out of 5 as we deployed this.

We think that differentiates and it gives us something from products that we can go out to each of these larger portfolios, property managers, asset owners and talk about this as a key differentiator for us in the market. Well, and the way that you used to do it was you would cobble together technology.

So maybe you can explain to the investor group, right, how you actually used to do it.

Yes. I mean I think that's right. About 3 years ago, we went on this journey, we called it as a company, customer experience excellence. It's about consistent scale, predictable processes, technology and talent. A lot of the similar themes you heard in today's presentation. And as we look at that, we wanted to make sure that we had product and brand standards reach the segments we're focused on. And we started to work with Michael and his team to really get Alantra around what makes the most sense and going back to the pure use cases.

And what we found was a number of these Calix solutions, particularly in the MDU segment were truly differentiated vendor or wide vendor and allowed us to have a solution sell that could ultimately scale for the test of time. And so -- and in your markets, -- so as you look at it, Scott, you -- one of the ways you differentiate. We talked about how you use SmartTown, Gavin to go across and HOA, so you can cover the eventing place, you actually went to a very different level because you actually went and covered all of the football fields, the baseball fields, you really focused on because you're very community orientated around how do you use that technology to kind of cover everything, right?

We started with football fields and because the ease of the platform had that in a 5-week period covered 9 schools. And then we took it from football to track to softball to all sporting events to all schools to all parks, all downtowns, all community centers, all volunteer fire departments at all part and we've hit our whole service area with Community Wi-Fi. If you sit in the stands and [indiscernible], Mississippi, a 12-person town in a football game, you have better service than the competition has in your home. Why wouldn't you switch to [indiscernible] Why wouldn't you want experience with us? And the whole Smart Home platform, we were -- Calix partners with you. We were one of the innovators of that. We were a lot what it was as well.

But we were out front. We were the first few to try that. We were the first few to turn 19,000 units on 1 day. And we have a first responder network from that. It's just amazing what impact you can have in your community. And I've sat in a room with a customer from another company that was a first responder and he looked at me in the eye almost crying and says, why would I do business with anybody that wouldn't care about me like you do. I said you shouldn't. And that's the difference that we can make in our communities because we work with Calix, a company that cares about us that ultimately care about communities. And that's what makes it special.

Well, let's expand out a little bit on that. So the power of the platform is that you didn't have to go. You talked about you only have 5 CSRs and a marketing person, right? So you didn't actually have to go like what Gavin said was, in the past, you cobble together a whole bunch of technologies, you could just turn it on, right? Maybe you could expand a little bit more about how simple that was for you to differentiate in your market.

So on 2 occasions, we've done that with you with SmartTown and with [indiscernible], very easy. We turned [indiscernible] than 7 days. So if you make a decision to go down the road with one of these platforms, one of these products, to scale for the community. It's just a matter of working with your team, my team, put them together and talk within days, within weeks. It's up and running. It's seamless. It's easy. It's simple. And I'm not doing this with [indiscernible]. I'm doing this with a team that's good, hard-working American team because you've got the rocket scientists for there. We just watch the rocket go off, and it's a great show. And I appreciate folks like autopiloting that with you. I want to be like Gavin and get in your contract. He's not new here. How do you get a 10-year time -- that's a good deal.

So we're going to talk after. But that the ease of that is because your people care, your people care about my community, your people care about our success because our success is your success.

Right. Well, so let's talk about how this can scale. So Brian, you got bought by somebody who wants them. So maybe you could share that. And then how do you think about scalability?

Yes. So you may have heard, we are now 50% owned by T-Mobile. It's a joint venture between T-Mobile, where we all T-fiber and EQT [indiscernible] yes, we've ramped up. We are growing it. Our builds [indiscernible] our pace and we're more than doubling our pace again on our builds and the installations are coming and have an ag service model and all that. But I would just -- I haven't even shared this story with you. So as part of the deal, the founding genesis of Lumos was [indiscernible] got sold to cash communication, Cox kept a B2B. They spun out the residential, the EQT. So we were exclusively focused as a residential business.

So that exclusivity has gone and so T-Mobile sales, the resident is responsible for selling residential, the joint venture, we're responsible for selling enterprise and complex business. So we got to do one thing, right? So we got to either say, okay, we're going to go build a DIA network, dedicated Internet network? Or we can take what Calix provides us on our XGS pod. We can put some SLAs around it and sell it as an enterprise network with XGS was outsell As. So that's what we did. So we just about a month ago, launched that, it's been a huge success at a higher ARPU big contracts like Avid, we're working on those. And it was the easiest product rollout that we've ever did in my entire life because we just took what you already did in the back end we wrap some SLAs around it and now sell it enterprise.

Well, and that's the same thing that we did with Verizon. So when we did the collapse with Verizon, it was all around how do you actually bring the different networks together onto a single fiber and put MPLS and all these other capabilities that would normally be -- I'd have to go build a separate enterprise put -- I'd have 2 fibers. I have my consumer fiber, I have my now because of the software, I can do it on a single and create a new business opportunity at very high margin because there's a lot of capacity in there.

Well, we'll get to capacity in a moment. So Brad, you do that all the time because your belief is that when you go into a market, you want to dominate the whole market from consumer all the way up to enterprise. So maybe you can talk about how we enable that because you've been doing that like Brian is doing now for a long time, right?

Well there's ways that you differentiate your business and one of our -- and maybe it's my fear. My fear of a competitive threat coming in and...

Healthy paranoia. I think that's why Andy from Intel always said right, the paranoid survive.

Well, sometimes it's unhealthy.

[indiscernible] you have to really out to get you.

That's right. And so what we did is said, the most expensive thing we're going to do is put in fiber. So why not make everyone that, that fiber goes by a potential customer. from grandma's house to the Corner Store, the small business, medium-sized business, State of Nebraska is our largest customer and then make the ecosystem work together. Well, when we started this, this was probably 12, 15 years ago, we didn't realize things like DIA circuits could be done by the current level of PON architecture. Oh my gosh, that collapses. But guess what? Mobile ties into our company just simply streaming. All these things and then the SmartTown, I met with an electric last week, and the CEO, and it's a pretty large electric utility said, you know more about our electric grid than we do.

Can we share information? Okay. Now that's the ecosystem. And can we work together and we do have this with some of the low income developments, how can we lower these financially challenged households, electric bill. Well, by lowering their thermostat at certain times when no one is there. Well, we know no one is there. And so we're doing some things that are pretty innovative, and it's helping society and it's helping education and it's...

Just while doing good is very profitable. right? Like honestly, there's a huge opportunity. We'll get to you because I imagine you're just biting your tongue seeing you also have an electric utility. So that will get to you in a second [indiscernible]

[indiscernible] is doing good, but operating in an exceptional fashion right, is less expensive, makes the customer happy, and it's less expensive. This idea of white glove, that's offensive to me. Why don't you just take care of everybody [indiscernible]. I think people like us. I think people like me, but no one wants to talk to their utility. We're a utility. And be like the water system. It always works. You always have plenty and it's good and clean. Well, they want their communications the same way. Well, and the platform makes it very simple for you to do the things that you used to cobble together. That was one of the things you and I have partnered on a lot is that part of our product road map has been driven by build this, build this, build this, you can eliminate the complexity of incremental vendors and actually collapse everything onto a single platform, right?

Small business, what we did with roaming and then where you go with MDU, right? So let's talk about capacity because in this crowd, there's a lot of people who actually think that Starlink is a competitor. And I know Amazon is a competitor. We've talked about that because they have consumer relationship, but let's talk about it. You have a ton of capacity. So why is it would that a customer would ever change? Well, if you do the experience, right, like you're talking about, they wouldn't, right?

Well, it's not binary. I have 2 subscriptions for a low orbiting satellite Well, one is in an aircraft and one is at a cabin that's down in the valley. It's the right tool for the job. My fiber isn't going to solve either of those challenges.

There's $10,000 or $100,000 you run it to that cabin.

And I'm not sure how you tether a plan but the reality is we work together. We all work together and we always have. But our industries came out of these monopolies, the cable industry, the telco industry. It's not a monopoly anymore. So if there's 10%, 15% that's solved in a better or different way cars. It's more efficient.

But at the same time, build the experience so that in you are where you have capacity, you never lose, right? Is that how you think about experience?

Absolutely. No, absolutely. And it's been a little bit different since the joint venture, but prior to the [indiscernible] we learned pretty quickly because I came from the telecom industry where nobody liked us and probably for fair reasons. So it's going in the...

[indiscernible], Brian. I just want to say that. So when we came to Lumos and we had a chance to create this new brand. Customers love our technicians, customer loves our customer care. And then -- and they just absolutely hate the cable company. So I said, okay, let's just provide a fair value. Don't -- our prices included taxes, fees and everything else, fair value, great customer service, don't chase the pricing game and just focus on customer rates and double down on customer experience, do everything we should have done at the telecom business, and it worked out great.

And that's why Calix is such a great partner because that's how you believe. You guys go above and beyond on the customer experience, you're with us almost every day, what can you do different? What can we do better? And that's the type of partnership that you have to be successful.

So now let's go back to the utility side. So you also run an electric utility, a broadband business and electric utility. And so there's a huge opportunity coming down the path with regards to utilities, too because you and either run a network all by itself, or you can actually say, as utility needs to go through a massive upgrade to deal with EV, there's -- I've heard the number 1 trillion thrown out just in the United States with regards to the investments, needed to go after those markets. So how do you think about your broadband business? And then also how do you evolve your utility business to bring those together?

Because your teams have been pushing us around how do you actually parse off that fiber to build out great utility experiences and all those parts, too, right? Like Brad mentioned, which then goes to a true customer experience of where you control on.

So in the utility, we saw it coming. We have no option other than move to where we have the fiber to the home, move to where we have the experience, where we can control the thermostats, where we can roll the water heaters. It's a capacity issue. It's a cost issue. There's only so much electricity. And every time you're adding all these EVs with gas going to $6, right? If every house has 2 cars and they're both EVs, that 1 transformer that powered 4 houses now you need 6 transformers, unless you manage the [indiscernible] so we need to manage the load for that. And that is a strong possibility on inability of what's coming. But load management itself is smart as well. If you look at electricity, it's grown very little in the last 50 years because you've had efficiencies.

[indiscernible]

[indiscernible] like that, right? So the next logical step of that is we need to manage the network, the management of the network needs to be consistent across all aspects. And it is going to be the Tesla managing the house or [indiscernible] we need to manage the network. The only way we can do that is build the fiber. Well, from my system.

And then put the software in the AI on top of it to make sure it actually really understands it.

Well, that's right. But the only way we could afford it is monetize the network by providing the service we provide. And that has opened the door, I built 2 fibers to every house, if I have a fiber fed a meter, it's there. I've built it. I serve 61% of my market, which is incredible. But I've built to 100% of that market. I can serve every customer I've got on the electric side and make a fiber connection, and now I can afford to do that. And so it will let us do that and control it and go places we never could have before that we had to go, but we couldn't find a way to get here, but for this. And so partnering with you has allowed us to do that well, has allowed that to be more than a service.

Internet is commodity. We sell service. And I sell service that you and Brad, pilot, my goal of doing business is the same as [indiscernible] want to give you that customer experience that you deserve. I think you can have it in a big city, and you have that in a tiny town in Mississippi, this is what we get with you. And your team as part of our team, it's family. We work together, we pull it together and we can take the next step forward, and we can untether that plane. If you [indiscernible] one of those is [indiscernible]

But from a profitability point of view, how has it changed your profitability across -- because your investors are different. Your investors are actually your members.

Absolutely. So there's a massive profitability gain here. So my electric coop is very unprofitable. We break even. We have a very, very thin margin and not far enough. I've been in the business for 6 years now. My fiber margins or 6x what my electric margins are with half the customers. And it's allowing to add infrastructure back. We transfer $1 million a month back to the electric go up or more that's putting money back in the system that we can grow and expand and do and it will allow us to have this modern grid as time goes on that we could not do but for this business.

So while I don't return the profits to my members under TV go up. I provide better service. I have better products, I have more employees and we invest in the system in ways we never could have before without this.

And so Gavin, as you look at your profitability as the business and where you see it going, how does kind of this level of automation, the elimination of complexity? How does that all kind of play into where you go and the future that you see?

Well, I think it does go back conversation we had to start out with around smart MDUs and some of the new verticals we're focused on. Classically, we've leveraged a lot of the experiential build references and HOAs and kind of associations. We're really focused now the MDU and asset owners a channel as well as new homebuilders and developers in order to really scale up to new markets. If you do it really well in this market in that market, it starts to be more and more again about just what is the value proposition how do you offer in that space. And in that motion, it's really around making sure that you have the right solutions, they scale across markets. And by doing that, you add significant more scale, more communities line, more MDUs that come online. And ultimately, that leads to a major increase in our EBITDA over time if we can maintain our fair share in margins.

So that market is one that hasn't changed for a long time. And in fact, I had an investor ask me that question saying, okay, so you're -- as you enter in the MDU market because you're early entrants into it. It's actually something that has a strong incumbency, right, and they haven't changed in a decade, right? So how do you actually see the step forward in partnership with us guiding us to how do we actually drive differentiation and spur that market on to allow you to win more?

I think it's twofold. When you go back to some of the -- just the core fundamental doing that. There is a lot of legacy solutions that have been there for a long period of time. If you're able to offer. It's no different than our road map with you as sort of every year, we sort of brought out the latest and greatest technology. It's smart MDU and similar platforms when you get to 30 of that one, too, they'll continue to iterate and we'll have more access control and the property managers can do new use cases that they could before. All of a sudden, that creates a value proposition that we're allowed to go in and speak about that as FlowStream collectively come in with a solution that makes a lot of sense.

Well, to your point, it's also a very fragmented market with a lot of different property systems [indiscernible]

[indiscernible] add also just especially with look, this is not a -- the MDU space coming back to reset some of these have been in place for decades plus as what they've leveraged. So when you come out and you say, here's the latest, there's all this AI and intelligence that we can also give to you about your residents, everything else allow you to offer a better service and a better resident experience, we think that's different.

So [indiscernible], what's next?

What's next? I'm excited for what we're going to do with what you're bringing us. So I'm going to expand that marketing base. I'm going to make my service better with my tech and we'll have better informed service crews. We're going to get to problems before they exist or before the customer knows they exist we're going to have better knowledge of our network. I think something that comes as part of that marketing discussion, I don't know that I said to you, so I'm an electric co-op. So my CSRs have never been in a business where they had something. It's hard for them when the grandmother calls and said I need fiber service. That base package is all you need. Once you may need more the grand kids may be gaming upstairs that I will be able to use this AI technology to know that they are and have a call with the grandmother.

Your grandkids are going to have more fun that can gain better, they can have less latency. If we do this greater package and my CSRs probably would never make that call, and I can use your [indiscernible] either prompt it or send a customized e-mail to them. That's going to be really cool. It's a game changer. I can get a more aggressive marketing stance from less aggressive people by this product that will make it comfortable. And so when the grandmother calls back in, they will have a great conversation with her because she's asking for it rather than they're having to sell it. And then they will be the persona of explaining it. and it's symbiotic. The AI gets us to that point. So that's what's next for us, continue to grow as best we can increase that service. I'm not happy with the [indiscernible]

I'd say, yes, well, by the way, that's almost a cult. As you know that -- it's almost a cult. And so Brian what's next for you?

Yes, it's a little bit of a complicated world right now, as you might imagine. And so I'd say this really wrap it into 3 things. Number one, we have one single customer. It's a very, very big customer. It might be one of the world's largest customers based on equity market or equity value that we have to provide performance, operational performance and we have basically contracts with it. We have financial penalties if we don't deliver that. So it's really building that world-class smart intelligent network behind the scenes to ensure that we are solving problems before anybody sees it before it happens. So that's kind of number one.

Number 2 is as we grow exponentially. Number 2 is we are now evolving into going to be getting more involved in the MDU space, getting more involved in the enterprise space where we are more opportunistic, that's becoming more and more of business core. And then thirdly, it's going to evolve as our new marketing agent learns the fiber sale business versus the wireless cell business different processes associated with that. I think you'll see this continue to evolve and get a lot more intelligent about how we'll go about doing it collectively. So probably the 3 pillars I would say.

And so Brad, you're actually coming out of [indiscernible] office in a couple of weeks. And we're actually going to do what we do annually, which is that the planning on how we do [indiscernible] innovation. So as you think about what's most important for you in this next step, especially with all the talk that we're -- and the partnerships that we've been doing on AI, what do you want to -- what's next for you? How do you see this evolving out?

Well, A lot of it is -- we're just crawling with AI and crawling with the partnership there. And hopefully, over the next year start walking. And then it's a run after that to torture the analogy. The real thought process is moving from efficiency, efficiency, efficiency. I don't know what efficiency means. We haven't added to -- even though we grow at 25%, 30% per year. We haven't added to our service and that. But it's making them smarter, being able to push more products through them. delighting the customer, I think the pace we're really missing is an industry that I think is maybe a bigger opportunity than all the network work and all that is what products should we be delivering to the small business or the home that allows another $10, $15 of our [indiscernible] and not in a transactional format in much like the mobile phone is developed.

There's a so-hard thing on the mobile phone. And I can't tell you what it costs me. But all of a sudden, as I got older, I learned I needed to start walking more. And so I look at it once in a while. I don't know what it cost me? It doesn't. It's included -- it delights me that I have it -- and I think that's where our businesses are going is we were very focused. We were electric. We were telco, we were cable, we were mobile. It's all coming together, and I think that's where Calix's real play is over the next 2, 3 years is enabling that and allowing us to be [indiscernible]

Great. With that, let's thank them very much for their time. We really appreciate it. With that, I'm going to welcome to the stage Cory Sindelar, who's going to walk us through our financial model.

Thank you, Michael. Good morning. I appreciate you spending some time with us today. So you probably realize that this is a really big story. We are building a company with the Agentic workforce that allows us to transform the service provider industry. It's on scale of like Uber transforming cabs or Amazon transforming retail. So my job here for the next few minutes is to wrap up everything you heard and how that's going to affect our financials over the next 3 years and ultimately, what it means in terms of return to shareholders.

You have seen us talk about these 4 financial objectives. We've been talking about it since we began our platform journey from 2018. As we go forward, nothing really changes with these other than we go faster and we go better. What you've just seen in terms of the prior presentations is that each one of these objectives is enhanced by what we're delivering with our Agentic platform. You've heard us talk about on the call that our visibility in term dollar business has improved and is near an all-time high. You might be wondering why we say that and what gives confidence in saying that. So let me share this with you. But this is showing a kind of a cohort analysis. It's the basis of our land and expand part of our model. Once we land a customer, they grow in time.

So what you see here is going back to 2018, we added 147 customers. And you can see how that just grows over time. As you know, as Michael said, all our services are tied to subscriber, right? We monetize on a per subscriber basis. And so we are maniacally focused on helping our subscribers win. Our customers win new subscribers. That's why we've developed our customer success team led by John, and it grows over time. You can see by this chart, it's up and to the right. Sometimes it grows a little faster than others, but inevitably, it is up. And because our customers are focused on delivering experiences they are winning in their marketplaces, and they are taking subscribers every single year, every single quarter, every month and every day. Now this is where we're at today, more or less.

Let's talk about what the opportunity is. This is the U.S. total addressable market. It's a fraction of where we're at today. We have an enormous opportunity in front of us. We are now beyond the elbow in a sustained growth phase. We have everything that we need to be able to go work against this. And so that's the opportunity we have. How do we get there? Well, largely the same way we have so far. First and foremost, we're going to grow our customers. Our customers are going to continue to add new subscribers. And with the power of Calix One, they'll be able to go at a faster rate. When you have the ability to market to the audience of one, there are chances of improving their success of landing new subscriber goes up.

You saw that the workforce agent workforce is going to free up additional resources. Those resources you've heard the panel up here today, talk about how they can redeploy those resources to go deliver new services. Maybe it's SmartTown, maybe it's MDU, maybe it's smart business. They're going to additional roll out additional new services. And of course, we're never going to stop adding new customers to our base. And with this genetic platform, I think we're going to be able to track more at a more rapid pace than we have in the past. All of that is accelerated by Calix Agent Workforce Cloud.

Let's talk about the numbers a bit. So on your left is kind of where we've been operating in terms of a target financial model. on the right is kind of where we're looking at in terms of moving forward. On the revenue line, we typically said, hey, we're going to continue to grow sequentially, and we still believe that's the case. And then we'll provide you a range. In 2016, we said we'd be at 10% to 15% yesterday. We just took it to 15% to 20% on the back of the kind of the memory surcharges. But for '27 and '28, we're going to call it right up down the line at 15% growth. It's based on all the visibility, our backlog, the reoccurring revenue, everything that we're seeing with our visibility being as high as it is. You might be asking me, well, can't you grow faster?

I'd answer to you, possibly so. But this is what we're committed to delivering in this time frame, and we'll see how it goes. On the gross margin perspective, we've been using and operating under 100 to 200 basis points for some time. And over the past 3 years, we've done really well. We've exceeded that target greater than basis points in each of the last 3 years. Unfortunately, in 2026, it's a little bit of a different story. We have some short-term headwinds that we're going to be facing. I'll talk about those memory and the dual cloud costs. But as we get beyond 2026, I think we have the opportunity to continue to get back on track to deliver 100 to 200 basis points of margin expansion -- so let's talk about memory. It's probably the one that's affecting us most. When you have a demand supply disconnect as large as the one we have now, the kind of price increases that you've seen over 10x increase in cost. Somebody is going to lose out.

The music is going to stop playing and somebody is not going to have a chair. Our #1 priority is to make sure that we are one of the people that have a chair. So Jerry and his team over here are world-class. You've seen us execute through COVID, one of the few companies that never missed a beat, and we went into COVID and came out of COVID stronger. We will take the same opportunity here, and we've been aggressively in the marketplace in terms of [indiscernible] memory because at the end of the day, when Brad and the rest of the team up here are winning their new subscribers, we better have units to supply them right? So for us, we don't want a top line problem.

It won't be a top line problem. So now you have to deal then with the cost side of this equation. So it's unfortunate. So what we've decided to do was to go ahead and pass along those costs in kind of the best way possible, which is we will just look for cost reimbursement. I'm not going to put a margin stack on top of it, okay? And so we're doing that in the form of a surcharge. Surcharge is deliberate. We don't want to raise price. If costs go down, we'll pass that through. costs go up, we'll pass that through. So this will be persistent likely into 2027. But as I said yesterday on the call, I size that headwind for 2026 out of 200 basis points. And that's just the fact that you have 0 margin business running through your P&L, right? You're putting a whole bunch of incremental revenue at no margin on top.

If we take a look at the second headwind, [indiscernible] dual cloud costs, the best new story there is, it's past tense. It's done. It's over. It was a one quarter phenomenon. And so that's probably the best thing I can say about it and it's short-lived, but yet had some headwind to us on our annual numbers. [indiscernible] looking at the operating expenses. You've heard how we're using AI internally. You can share some time over here with Jerry on how we're using AI and supply chain. We're using AI to help our customers to go faster? Well, we're also looking to do it internally. So you ought to expect us to see some improvement in terms of the way you run business should be some OpEx savings as a result of it. So as we look into 2027, we are still forward investing in some of the things that we wanted to do for AI, and there's a couple of new markets that we're going after in terms of large customers, MDU build into.

But we're setting that target in 2028 that we should be able to grow OpEx at half the revenue growth rate. And we'll make some progress into that path in 2027. So it's kind of a recap. The end of 2026, we'll be back in that old financial target model. So we told you we would forward invest but that we would be back in target model by Q4 of '26. From there, we'll actually make some improvements by growing less than the revenue growth rate. And then with OpEx, '28 making some more substantial progress. Let's roll that all kind of together.

Over this next planning horizon, we see our gross margins drifting back into the 60s. OpEx will drift down to 40% or below. So we can see clearly during this period of time that we can get to a 20% operating income for the company. This business is going to generate a tremendous amount of cash. So that cash is going to go to the balance sheet. So these are the metrics that we are gauging how we're going to operate our balance sheet there's reason to make no change to them. we can still run a very, very tight DSO. And by the way, we frequently do much better than the 35 to 40 days. For example, last year, we actually hit 24, which is livable Inventories, no change. Let's stay to 3 to 4, and we're running it on the higher side of inventory level for 3. And that's largely because of the current supply environment that we're in. We need to do more to be prepared to meet the demand equation and to be available for some of the memory and other components that we are buying ahead.

In this battle for memory, one of the things you want to make sure you do pay our vendors, maybe pay them early. You want to be one of the best customers to your vendors as possible so that when they have that memory and they want to allocate it, they want to pay one of their favorite customers. That's Calix. So the DPO is important. So that gives you our cash conversion cycle. No real changes to it. We'll continue to operate in a very clean balance sheet. So what do we do with all that cash? I personally would love to have it all stacked up around me. I have visions in my head of either think of scar face or pinky blinders or something like that.

But Carl tells me we're not a bank so I can't keep it. So at the end of the day, we have a very disciplined capital allocation process, and we've talked about that, but I might as well recap it one last time for those that may be new to the story. So we think about our cash in 3 buckets: our operating cash, our strategic cash and our rainy-day cash. And over the next 3 years, we would anticipate that we need some additional cash in that operating bucket, not only to grow the growth in the company in terms of expanding the balance sheet but to have those inventory investments. If I think about the strategic balance, we are an organic grower and have. But we've always set some money aside in our minds thinking about, hey, there may be an opportunity to go buy a technology or do some kind of tuck-in.

But we have recently just continued to build onto the platform, take them to you, for example. We would rather have actually built in the platform, creating greater DVs for our customers. than to go out and buy a piece of technology. I think over the next 3 years, that trend will continue. And so our needs for the strategic cash probably go down. So that means the cash that we're generating will likely drop to the third bucket, the need a cash bucket at a faster rate and will be the fastest-growing part of the bucket. So we do a disciplined model. We come up with a long-term model. We're going through it with the Board every quarter. And as I said many times that the larger the cash balance gets the less sensitive we are to price.

And so every quarter, we put a treating plant put together and the shares will be repurchased and you've seen that we've done a significant repurchase here in the first quarter, and it will continue to be that way. So for the next 3 years, I would expect us to continue to retire shares at a healthy clip. But there's another part of the story and it gets better. As we've moved from this early development part of the company and to where we're now into the early majority sustained growth, we are realigning our equity programs. So in the past, we were using performance stock options. And we had 2 employee stock purchase plans.

Going forward, we are pivoting to performance stock units and collapse those 2 employee stock plans into a single plan, which creates more and more of like a stock -- employee stock or plan closely aligned to long-term shareholder value the effect of which should be fewer shares being issued. We think the combination of cash being returned in the form share repurchases should more or less offset the share being issued through the equity plans when we become dilution neutral. Obviously, in 2026, with the large purchases that we did in the first quarter will probably dilutive negative.

So just recap. We are the only company that has built an genic workforce platform capable of transforming the service provider industry. The tau is enormous. We have a proven model that works. We are asset-light, and therefore, investors should expect significant cash return to them over the next 3 years.

Thank you. With that, I invite the leadership team on up here.

Shane and John are going to join us. All right. Can [indiscernible] the mic, put your hand up. Glad to take questions.

So maybe a couple of questions. You've been selected the managed services opportunity. And you've talked about it since like Calix stays as well about that sort of $1 to $10 per subscriber rate you made a lot more progress on that front. So now with the visibility that you have, can you help us quantify that a bit better in terms of where you are in terms of progress there including now the PI-led opportunities that you're sort of looking at, how should we think about what's a reasonable milestone for that to get to during this sort of model period that you're discussing? And then a follow-up would be maybe Cory for you is you talked yesterday as well about bed being an [indiscernible] on the model.

So when we think about the growth target that you're outlining today, which is 15%, is beat incorporated? Are you saying, oh, let's see what goes sort of how it plays out? Is that sort of what's embedded in there? Just wanted to understand because you did refer to it as [indiscernible] yesterday.

So on the first part of your question, which was around is there an opportunity to accelerate the business. John talked about [indiscernible] ore currently doing one of the issues that we always had was capacity of the customer. right? Because you heard from literally sitting in the seat, when Scott was talking about the fact that he only had this many employees. I didn't have the capability to expand out, couldn't use all the products, couldn't get them launched faster. So we think in our existing base, where John showed an attach rate of only like 40%, there's a huge opportunity for us first of all actually put the capabilities in place using Agentic to actually do the work for them. And so John's organization is going to change radically in how we support customers because -- and I've said this time and time again, this next stage for us and what we've built over the last 2.5 years is in the past, we would have to control the customer to please launch that product.

Please listen to us, please change your website into an experienced-based website. The next stage for us is going to be push the button, push the button. Here's the ROI, get it done. And so we think that represents a huge opportunity to radically expand at a faster rate with the existing business. And then the second part of that is, as we look at new markets, we showed some of the TAMs. These are very logical expansions of our existing platform. We have incredible partners. Again, you heard about from Gavin on what the opportunity is in the MDU market, which is embedded legacy companies aren't investing to actually innovate. And so we come in with a very different mindset.

They basically brought their technology interlace down, and we all know that things don't scale well down. We brought our technology consumer up, and we actually have an opportunity to completely transform and disrupt that business. And that's the ones like Gavin, which are the ones we partner with, why he's so interesting is because there and a challenger in that market looking to significantly grow market share. And so when you partner with them, what they do is they say, hey, I can try to be like everybody else but here's all the opportunities to completely change the business because I want to disrupt. Gavin doesn't want to go and be like everybody else because that's not how you make money.

If you're like everybody else, and you're just kind of who do I pick, right, versus partner with someone like us and be first to market. That's the other part too is he wants to be first to market. And so this represents a significant expansion of TAM. You look at the MDU market, that's excluding hospitality, international what we can do in inventing. Just the [indiscernible] MDU market alone is bigger than the global Tier 1 market for us, which would only software. And as frankly as big as our incumbent market today. And what we're really good at it is going into markets that are highly decentralized.

If you look at the U.S. regional market, what is the U.S. regional market. It is unlike any market in the world. It's not dominated by a single provider. It's like 3,000 companies who are highly, highly distributed. And what do we do in that market? We dominate it. So as we go into this MDU market, this is what we actually do really well. It's the exact same type of market as we're currently in regional sets going after one big company. And so we know how to do this. And what's great about the regional market and the way that not only our products are built, the partnership and trust we have with customers, but also our selling and success motion. That creates a moat around the business that we have because customers trust us, and we are going to go do the exact same thing in that market. And so it represents a significant opportunity for growth.

And then the second question is over to you, Cory.

So I'll just add a little bit on it. On the monetization of $1 to $10. All our offerings have been on a price volume curve. And as we go into large customer segment, obviously, there'll be a lot volume, I would expect them to be on the lower side of that scale of 1 to 10.

Well, that's why the TAM is different because of the fact that the other part too is they're also going to be running it on their own private Google instance. Therefore, we're not going to have an appliance element in there. and we won't actually have the COGS. So Shane's team is not going to be running it like we do with our current customers. It's actually going to be their Google instance with our stack and a pure software-only sale, which is why when you look at the TAM, what's so different about it is that that's 100% are pretty close to.

Yes. And on the bad side of it, the answer is some, right? So our forecast is being built on the orders that we have and the pipeline that we're driving and our customers have provided us with their initial set of builds. And so I would say that there is some beat in those numbers. It doesn't include all of it should it come to pass. And so it provides an opportunity to outperform that 15% should we get what we think we might in terms of share.

Next question. Mike?

Mike Genovese from Rosenblatt. A couple of questions. First of all, Michael, can you talk about expectations for selling software only to customers that don't buy hardware, how big that business could be. But also how does the telemetry work there if you're just taking the data from publicly available sources and not your own captive equipment, does your competitive -- how is your competitive advantage in that area? And then I'll follow up with one for Cory.

Yes. So by the way, it's the same. So for example, what we do currently is that we have millions and millions of non-Calix systems actually being managed by our system today using industry standards like TR 69 and those things. So while Shane talked about being in the data path, we're not in the data path there. We're actually using industry standards to collect the data. So that still becomes available to us. It just means that the customer can't do quite as much, right? But with the new business model, what we do is actually our operating systems can run on other systems. So we put our operating system on, let's say, you're a big customer, you say, I would like my volume SKU to be I'm going to go direct to a manufacturer right? So I go do that.

But then I still put the Calix operating system on that. And so it sits on top of it, and it can be an industry standard operating system. We've openly talked about how we're going down the path of purple, which is the standard. We will be purple certified. And so they pop that on. They then have containers where they can add to services. But when that OS is there, it is also collecting the data, right? It can link into our cloud. because what people don't really understand is that they're thinking about the OS and how to grab data [indiscernible] and I have talked about this a lot, you still have to have that cloud system that sits on top of private Google instance, and it provides access to what are the operations, services, marketing, all those capabilities.

And then in that customer has identified in the workflow, you're just going to consume it in a different way. For example, I am running ServiceNow, our sales force in my call center service environment at a view is Calix a genetic agent is going to be a component of that. It's going to feed data into the Salesforce or ServiceNow technology. And then what's going to happen is if I'm a call center agent and she calls in I actually am going to click through and I frame is going to pop up and you might be in -- if you choose, you can be in Service Cloud to diagnose our issue, then close it down and go to the bigger view customer. So this is something we're very good at.

The other part of this is that when you look at the evolution of open systems in technology, that's what model context protocol is, which is how you don't move data, you can actually use context or agent to agent so that we can actually have what is basically an API framework for agents. So we have had a long track record of standards-based, right? We're very good at integrating into disparate environments and we see this as just the next evolution and the ability to enable those customers to do those things. So I'm assuming you had a second question, which would go to Cory.

Yes. But if I could just follow up, first, Michael. So are you expecting to win other new large customers with this?

100%? Absolutely 100%. Shane and I didn't go to Barcelona because we like Persuto,right, and cappuccinos. We were in Mobile Congress talking to big customers. This is the start. We had to get past the -- like we went to Mobile Congress to kind of say, hey, we're here. We have some big customers. We have Verizon. You heard from Lumos, what we're doing with T-Mobile.

And the next stage for us is now to go after that big customer base worldwide. And so we kicked that off at Mobile Congress than now were there. That's why I was on -- did a keynote there to start having those conversations. Absolutely. What's different about us though is that we also don't have to go after a large customer in one way. And what you also heard from Brian with regards to where we are in MDU, a large customer as many facets and highly dysfunctional, right? Having been in the big companies, I know how this works.

It's silos, right? What do they function? You have a consumer, you have a small business organization, you have a multi-dwelling unit organization, you have a business organization. And all those segments in there are filled with executives who actually run their businesses. So what's also unique about us is that our platform does everything. It's just about where is the entry point. And as -- if you look at my career history, I've been in leading sales transformations and transformation of companies for this is my sixth, right? When you're in a sales and marketing role, the most important thing that you want is lots of opportunities to get into a customer.

If you're a one-trick pony and the only way in and they slam that door is you can't -- that one opportunity that you had is closed, that leads to a lot of disappointment. And in the end, a small pipeline against the TAM. If you can actually talk to a large customer and say and go to talk to all the different organizations about everything that we offer, what you can do is search for a beachhead. You're searching for is it and in the end those beachheads and large customers generally come from who wants to make a name for themselves, who's driven drive change versus who's on the -- who's actually just waiting for retirement.

You have to go find the people who are saying, you know what, I just took over this organization. It's the same as when I got promoted at Microsoft, the people before me, and I got -- when I went to bell, some of the people before me weren't driven to drive change. They were just like, I want to keep doing the same thing. But when I took over a business when I was at Microsoft, anyone who come to me with regards to I need to change my business. I want to go and take us from 2% growth to how do we do 50% annual growth, I'm ready to listen. And so the good thing is that we just have to seek within those customers who are the ones who want to drive change and then enable them with our platform.

And then what happens is once the platform gets in, you have the beachhead and everything is available now to the entire customer. They just have to turn it off because once you integrate into the back office, everything works. So for example, if I go into a large enterprise customer, a Tier 1 and we win the small business go to market, all of our capabilities in the platform are enabled for all segments of that customer. And so you're in. And then you can start saying, well, you don't have to do a 2-year OSS/BSS integration into Amdocs because it's already been done, just add it to the product catalog and you can be up and running in 90 days. And that's a huge change.

Okay. And then finally, Cory, I just want to get your confidence on the idea that the second quarter could be the bottom for gross margins and that will grow sequentially from here. And given the uncertainty about what's going to happen in the spot pricing for memory, how do you factor that into what you're thinking there?

Yes. So Mike, I don't know it is the bottom because the gross profit is going to be the same or better, but the margin might be down. Because if you think about what we're talking about, and I said, 200 basis point headwind, it's not even. We're just getting started with the memory surcharges in quarter, you're going to get a full quarter in the third quarter and fourth quarter of those memory surcharges. So how much 0 margin revenue I'm blending in will have an impact on my overall margins [indiscernible] but I can tell you the GP will improve right, does that make sense? Yes, you're going to keep seeing EPS growth, but you're not going to see -- the margin is going to wash out until we get through the memory. And I will say on the software side, though, let's just talk about that because that's a positive.

With having the dual cloud cost behind us, we'll spend Q2 kind of optimizing the new GCP platform. But I would suspect by the time we get to the third quarter, we'll be setting new records in terms of software service gross margins and from there, build on it. So you're going to have that as a tailwind as we move forward.

I'm actually going to make one other point on that topic, too. So when I was up here -- I'm sorry, I just got [indiscernible] a lot too, when we talked about the velocity of product -- so you see the dip, right? The dip actually in 2025. To be really clear from a selling motion point of view in the end of Q4 and all through Q1. It actually had a dip on our selling motion, too. It was all hands on deck. It was all hands on deck for Shane's team to get it out. It was all handled on deck for John's team to support customers through that transition. It also took a lot of our selling capacity in Q1 and off the bench from actually selling because they -- for example, all our sales engineers, we're working with customers to make sure that those customers had support as they went through this transition.

It was all hands on deck. So in the same way, I think that, to your point on software sales, the other part, too, as we go through Q2, Q3, we're back to regular cadence as an example being what Shane talked about with MDU, the big last release that really makes this platform grade is in May. We're 2 weeks away from that, and we go hard at that market. And so now those selling resources aren't supporting customers, which is you never want that. We never wanted selling resources supporting customers, but that's what it took to do something monstrous for our customers.

So with that, we can take one last question or not. We're out of time. Scott, one last question because we're over time.

I always have multipart. But just real quick, Cory. In terms of the outlook for '27 and '28, I'm wondering what you could tell us about your assumption memory pricing, kind of what the baseline is that's going into that forecast? And then Mike, going back to the private cloud. Huge opportunity there. I wonder if you could distill a little bit what the pipeline looks like, how big, how diverse the engagements are. And these are larger complicated sales. So kind of the closure time period of when we would actually expect new customers becoming onboard. Is that '27, is that '28? How are you thinking about that? .

Yes, Scott, the underlying assumption is that memory will find a new equilibrium sometime in early '27. I don't think it goes back to where the pricing was, right? It will be at some new higher level. But I think you'll start seeing -- and you're seeing the memory guys are already putting more fab capacity to it Micron has stuff coming on in early '27. So you'll have other people that are optimizing the amount of memory that they're consuming.

We also even saw what Google just did where they saw a 10x reduction in memory. So you're going to start seeing those optimizations. And as the fab capacity increases, that represents an opportunity to.

So I think it goes to early '27 and then we'll see a new baseline.

And so with regards to -- I'll close off with this, with regards to driving market, one of the great things is that competition is increasing for our customers. It's not great for them. But the great thing is that we're partnering with them, we understand their challenges, and we understand their markets and how they are going to compete and win in their businesses. So I would say as we go after these incremental TAMs One of the important things is that it isn't where we went back to and you had one provider per market, and there wasn't a lot of competition.

So you know what, as long as -- if I have a fiber and I show up, I win, those days are way behind us. There's a lot of competition. And so I would say with regards to the selling motion, the good thing is that customers see a need and you heard that from the 4 people who are up there, they're all like, I want -- Brad said, "I find white glove offensive. I want to have the best experience all the time." You heard them talk about how they dominate in their marketplaces. They talked about -- you talked about what Scott said with regards to everything that we're doing for the customer. In the past, John and his team went have spent a lot of time to try and convince customers to do these things, but with the threats and the pressure there, that represents an opportunity where people are starting to say, "You know what? I also need to do these other things, how do I build the best experiences. "

And so yes, these cycles take time. Yes, there's a -- you have to build out the pipeline. But the great thing is that there's demand because of the fact that there's threats, and we're best positioned to deal with it. So -- and the pipeline grows. And with that -- yes, we'll wrap gentlemen, off you go. Thank you very much. Thanks to the leadership team.

So now got to cycle right back to the end of the slides, which actually just says thank you. So to everybody in the room to the customers who took the time huge thank you for coming and joining us on the panel and giving color on what the business model is. And to all of you as investors, thank you for investing in Calix. We really appreciate your time, and we hope you have a great day. Thank you.

Greetings, everyone, and welcome to the Calix First Quarter 2026 Earnings Conference Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Nancy Fazioli, Vice President of Investor Relations. Nancy, please go ahead.

Thank you, Alicia, and good afternoon, everyone. Thank you for joining our first quarter 2026 earnings call. Today on the call, we have President and CEO, Michael Weening and Chief Financial Officer, Cory Sindelar. As a reminder, today after the market closed, Calix issued news releases, which were furnished on a Form 8-K along with our stockholder letter and were also posted on the Investor Relations section of the Calix website. Today's conference call will be available for webcast replay in the Investor Relations section of our website.

Before I turn the call over to Michael for his opening remarks, I want to remind everyone that on this call, we will refer to forward-looking statements including all statements the company will make about its future financial and operating performance, growth strategy and market outlook, and that actual results may differ materially from those contemplated by these forward-looking statements. .

Factors that could cause actual results and trends to differ materially are set forth in the first quarter 2026 letter to stockholders and in the annual quarterly reports filed with the SEC. Calix assumes no obligation to update any forward-looking statements which speak only as of their respective dates. Also on this conference call, we will discuss both GAAP and non-GAAP financial measures. A reconciliation of GAAP to non-GAAP measures is included in the first quarter 2026 letter to stockholders. Unless otherwise stated, all financial information referenced to this call will be non-GAAP.

With that, Michael, please go ahead.

Thank you, Nancy. It was another incredible execution quarter for the Calix team. Record revenue with strong demand continuing into 2026 with customers. At the end of March, we completed the migration of all existing customers to the third generation of the Calix platform. Launching on Google Cloud, thereby enabling the expansion of our capabilities and the markets that we target. As important, those customers who expand their partnerships with Calix on Calix One begin to see the benefits rapidly as agent workforce and our AI native platform comes to life.

The impact of AI will now start contributing to our customers' success by helping them transform their operations, allowing their teams to add capacity and capability with AI and accelerate experiences that they need to differentiate in the markets they serve, thereby enabling their teams to compete and win. Today's call is focused on the quarter and our 2026 outlook. Tomorrow at Investor Day, we'll go deeper on how Calix One expands the opportunity of our model with proof directly from customers who will attend the event and are ready to share.

With that, I'll turn it over to Cory to walk through the results and guidance, and then we'll take your questions. Cory, over to you.

Thank you, Michael. In the first quarter of 2026, Calix delivered yet another quarter of record revenue of $280 million, marking a sequential increase of 3%, driven by continued strong demand for our platform. This quarter, we welcomed 14 new customers, reinforcing our ongoing efforts to grow our customer base while supporting their expansion within the local communities they serve.

Remaining performance obligations were $376 million, down 2% sequentially and up 11% year-over-year. The sequential decline related to a robust fourth quarter comparison and our focus on completing the migration of customers to the new third-generation platform. Current RPOs in the first quarter were a record $157 million, representing a 3% sequential increase and a 22% rise from the same period last year. We anticipate that RPOs will reaccelerate in the second half of 2026 as we gain momentum with Calix One, underscoring the strength of our business model as customers focused on delivering exceptional experiences, adopt our platform, add incremental offerings and win new subscribers.

Non-GAAP gross margin was 57.2%, down 80 basis points sequentially due to investment in our dual cloud environments as we migrated customers to our third-generation platform. Compared to last year, non-GAAP gross margin increased 100 basis points. Our balance sheet remains strong. DSO at the end of the first quarter was 36 days. Inventory turns remain steady at 3, reflecting continued inventory investments to address robust demand and building supply continuity, and we generated free cash flow in the quarter of $7 million.

We also invested $171 million to buy back 3.3 million shares of our common stock at an average price of $51.34. Furthermore, the Board today authorized another $100 million to be added to this program. This investment speaks to our belief in the tremendous opportunity ahead and our commitment to creating lasting value for our stockholders. We finished the quarter with a strong cash and investment balance of $243 million.

Turning to guidance. Our revenue guidance for the second quarter of 2026 is between $287 and $293 million, representing a 4% increase at the midpoint over the prior quarter. This reflects continued robust demand trends and a modest benefit from recapturing a portion of the higher memory costs via a memory surcharge. For the year, we expect revenue to grow between 15% and 20%. With demand supply disconnect, so large related to memory components. There will inevitably be some companies that will come up short.

Our first priority is to ensure that we have adequate supply such that our customers can continue to add subscribers and take market share. Our advanced purchasing had allowed us to avoid higher memory component costs during the first quarter. However, that advanced supply has run its course, and we now face market prices. We are partnering with our customers to share in the higher memory costs, by initiating a surcharge, albeit it is a partial cost recovery and without adding gross profit is one way we can help our customers in this unfortunate memory supply environment.

Our gross margin guidance for the second quarter of 2026 is between 24 -- sorry, 54.25% and 57.25%. Reflecting the effects of higher memory component costs, the impact from surcharges and the customer and product mix. The decline in appliance gross margin is expected to be offset by improvement in software and service gross margin as the dual clog costs abate, and we optimize the current cloud environment. For the year, -- we expect our non-GAAP gross margin to decline between 50 and 150 basis points.

For non-GAAP operating expenses, we forecast $128 million at the midpoint in the second quarter of 2026, which is a sequential increase of $1 million. This increase is mainly driven by our efforts to expedite AI functionality and enhancements to the Calix One platform. Importantly, we continue to expect to return to our target financial model for operating expenses by the end of 2026. And improving our operating leverage and profitability. Tomorrow morning, we are excited to host our first Investor Day in 4 years at the New York Stock Exchange.

The event will look -- will provide a look into our strategy, innovation road map and the long-term prospects that drive our confidence for the future. In addition, we will outline our key targets for growth, profitability and cash flow, giving investors benchmarks to track our progress and understand how we are positioning Calix for sustained success in the coming years. Nancy, let's open the call for questions.

Operator, ready for questions. .

[Operator Instructions] Our first question comes from the line of Samik Chatterjee with JPMorgan.

I have a couple. Maybe if I can just start with you, Cory, on the gross margin guide here, just to fully understand the drivers. Once you put the surcharges through -- are you expecting a recovery recovery in the back half of the year in relation to gross margins on the appliances as some of these surcharges flow through on your revenue line. And then as if memory does continue to sort of go higher memory costs continue to go higher, what's the plan here because you're not passing through the margin on the cost increase. So what prevents more downside on the margin percentage as you go through the year or if memory costs continue to increase? And I have a follow-up.

Yes. Thank you for the question. So our plan is to recover the costs. And so if there are further cost increases, we would adjust the surcharges accordingly, the effect of the surcharges by themselves, put a headwind to the gross margin. I estimate that in 2026, the effect of the surcharges for here to the end of the year, represent a 200 basis point headwind because you're adding a large amount of revenue at 0 points of margin through.

Okay. Got it. And then maybe just on the demand side, you did mention sort of stronger demand from your customers. And I'm sort of trying to parse out, obviously, haven't been able to do the math yet in terms of you're raising the revenue guide from earlier talking about 10% to 15% growth on the top line to now 15% to 20% and you also are outlining stronger sequential growth for 2Q than you have in the last couple of quarters. How much of that is stronger customer demand that you're seeing in terms of orders versus the benefit from the price increase, which you're referring to as a sort of decently large price increase that's going to go through the revenue line as well. So just maybe help us break that down? And what are you seeing in terms of customer orders that's maybe giving you a bit more confidence as well.

Yes. Great question. The majority of the quarter-over-quarter increase is due to customer demand and a lesser portion is the surcharges. We are rolling out those surcharges now they'll take effect in May. So we're not getting a full quarter of recovery this current quarter. But -- so consequently, the majority of that increase in revenue is coming from increased sand and to a lesser extent, memory and price increases.

Our next question comes from the line of Scott Searle with ROTH Capital Partners.

Just wanted to dive in on the gross margins related to the dual cloud cost. It looks like it was in the $3 million to $4 million range. Just want to clarify, does that go away completely by the second half? How should we think about modeling that? And then I had a follow-up.

Yes. Great question, Scott. That's the good news story is that -- we've got all the customers migrated onto the new cloud. And so yes, the dual cloud environment is done. So it's done. As we sit here today, it's done. So -- so the penalty that we incurred happened in the first quarter. I think you got it sized about right. And so what you would expect to see on that line is for it to return back to levels that were previously at I would expect for the next quarter or 2, we'll be back at record levels and continuing the progress that we've been making on that line.

So in terms of my follow-up, I've got a lot as it relates to the 1 platform, but I'd rather than preempt tomorrow, I'll save it for then. So maybe if I just could fiber availability in general, how is that impacting demand? And it's been interesting to see some of the StarLink numbers that have filtered to the marketplace where they've gotten some traction in places that I really didn't expect that they would in terms of more dense suburban environments than you would ordinarily think that they participate.

I'm wondering what you're seeing in terms of your customer response on that front in terms of their demand, their rollout plans? Is this pushing them to accelerate? How is that kind of factoring into the calculus in terms of the overall market demand of your core customers?

I spent a lot of time with customers. It's Michael. And while the startling thing is there, you generally see it in real areas where there's -- when you have a 6-mile run to actually join a farm, obviously, Starlink is a good example there. I don't hear anyone saying I need to accelerate my rollout of fiber to compete. But for us, frankly, another competitive pressure is a good thing for Calix because if you think about which we'll talk about tomorrow, we try to think about the experience-based nature of what we're doing and how we help our customers differentiate and transform their business to win subscribers and grow revenue by delivering an amazing experience, whether it's a consumer small business or multi-dwelling units, frankly, that's good for our our business is because that gets them listening if in the past, they didn't feel that competitive threat. So it's all good.

Mike, and just fiber availability, what our customers saying? And maybe if I could sneak 1 other. I think you talked about guidance for 15% to 20% this year. 15% is really just kind of bumping along at the level you're at today. So what kind of visibility do you have in terms of deployments into the second half? Are you starting to feel pretty good about the lower end of that range?

I haven't heard anything with fiber availability have you.

There's been some talk, a little bit of like sub...

Yes. BEAD but we expected that as all that be money starts flowing, that there's going to be -- there's going to be some supply right? So Cory, any comments on that?

Yes, Scott, I'm either more bullish or more negative on BEAD at the moment. I would say my temperature is about the same. It's progressing as we would expect. We've got tens of millions of dollars forecast in the second half of '26 related to BEAD. We're starting to see states actually start receiving their money. So things are kind of working its way out. We are not hearing that fiber shortage is causing a significant impact to BEAD demand as we're hearing it.

Our last question comes from the line of Christian Schwab with Craig Cowen Capital Group.

Congrats on the good quarter. Just for foot clarity, our previous guidance was 10% to 15% top line and near the high end. And now we've taken it to 15% to 20%, should we just assume that that's the surcharges that is going on there's incrementally better visibility and continued strong demand, but demand has been accelerating beyond what you thought 90 days ago is it? Or is it?

It's yes and yes to that, Christian. So clearly, the effect of surcharges is going to move us up into that higher part of the range. But we're also seeing some of the best demand that we've seen.

Yes. And that's we'll talk about tomorrow on Investor Day, right? So for us, as everybody knows, when you're rolling out the next stage in the platform going through that evolution, there can be unexpected challenges. We had gotten pretty hard for being done in Q1, and I'm really proud to say that the team got through it. So having gone through that and not facing any incremental delays that allows Cory and I had to sit on this call and be very bullish about the future for 2026 because this is what our team has worked towards for -- since November of 2023, we've been pounding away at this for over 2.5 years.

And now our AI native platform is got 12 -- more than 1,200 customers loaded. And tomorrow, we're going to talk about how fast we're going to go and how we're going to go skiing out into that. And I think blue ocean of incremental TAM and compete aggressively to grow the company. So yes, you're hearing me be very bullish as you'll hear tomorrow during Investor Day.

Great. And then my just quick follow-up question is, as it relates to BEAD, congrats on rightly getting some of that dollars generated for the company. I think you said tens of millions in the second half of calendar '26. What is the internal plan for what year you think will be the peak of that program? And what type of annual revenue number should we be thinking about?

So Christian, we've talked about that in the past. We've done some high-level math. I think that you'll start to see this thing ramp more significantly in 2027, probably peaks in '28. And I don't think that I would want to kind of put a number on it, but it's it's potentially high tens of millions. As we said about BEAd, we see that as an accelerated on top of the core growth model, right?

Our next question comes from the line of George Notter with Wolfe Research.

This is Karen on for George. Just a quick question. Any update on traction with Tier 1 customers on the cloud side?

None that we're willing to share. .

Okay. Got it. And then -- any comments on the quarter-on-quarter uptick in appliances. I assume -- I think you guys have mentioned in the past that DVS takeouts are mostly done, if not all out. But what drove that is the quarter-on-quarter?

Customer demand for our products because of the fact that we're better than our competitors. Yes. And it was within our guidance that we provided -- so there were no surprises in the quarter. We added customers, and we'll talk about that tomorrow on Investor Day when we walk through the core drivers of growth.

Our next question comes from the line of Tim Savageaux with Northland Capital Markets.

Good afternoon. maybe a couple of questions. First, I don't know if you mentioned it upfront, but if you can go through the BEAD commentary again. Was it any more granular than tens of millions in the second half and maybe on a somewhat related note, there's been some news out of the FCC recently about foreign-made routers in the U.S. and some exemptions there. I wonder if any of that has any implications for Calix.

Yes, Tim, on the BEAD piece all we said is that we would we start to see revenue that I'm either more bullish or less bullish I'm even keeled through last quarter to this quarter. Things are progressing along as you would expect it to. We think that then translates into tens of millions of dollars in the back half of this year. And obviously, the ramp will start next year. In terms of FCC regulation it appears to be that the timing is fairly quick, kind of measured in weeks, not months. So we would expect to be receiving our conditional approval here soon.

Calix has sought received various government approvals over Calix's 26-year history, expect no difference here. But I would also point out that FCC approval alone doesn't really differentiate your product at all, where Calix wins is after you deploy with automation and intelligence and subscriber experience that lowers OpEx and drive business outcomes.

Which we'll talk a lot about at tomorrow at Investor Day is that, as Cory said, in the last 26 years, we've done this frequently through a myriad of government programs and as a proud American company. And we're actually going to show you tomorrow what the power of it and agentic and AI native, agentic ready platform, how that's going to help us transform our customers' business, and that's going to drive outcomes. So press release on FCC is kind of irrelevant. We do it.

Okay. But maybe just a quick follow-up there. I mean, is the timing of the conditional approval, having any impact on the business at all or others have...

Absolutely not. They're actually the -- the FCC is actually moving really quickly. So we anticipate no issues whatsoever. It's a nonevent. We will do this quickly because we're already well down the cycle. And so we won't press release it. We'll just -- well all of our customers we made aware quickly as that resolves itself.

So Tim to be a little bit more clear on it, any existing product that's shipping is not at risk. So there's not having any impact on current shipments. It is the next release in terms of new products coming into the marketplace. And so we have a number in the pipeline, and those are what we've applied for conditional approval. So expect to get those approvals. And it appears that the SEC is moving pretty fast, which is great, which we were surprised at how well how fast is the process is going. It's great. So we don't anticipate there being any problems as a result of that -- of those new rules.

Apologies. We have reached the end of our question-and-answer session. I would now like to turn the call back over to Nancy Fazioli, for closing remarks. .

Thank you. Calix will participate in several investor events during the second quarter, most importantly hosting our Investor Day at the New York Stock Exchange tomorrow as referenced. Information about these events, including dates and times and publicly available webcast will be posted on the Events page of the Investor Relations section of calix.com. Once again, thank you to everyone on this call and webcast for your interest in Calix, and for joining us. This concludes our conference call. Have a good day.

Greetings, everyone, and welcome to the Calix Fourth Quarter 2025 Earnings Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Nancy Fazioli, Vice President of Investor Relations. Nancy, please go ahead.

Thank you, Daryl, and good morning, everyone. Thank you for joining our fourth quarter 2025 earnings call. Today on the call, we have President and CEO, and Michael Weening and Chief Financial Officer, Cory Sindelar. As a reminder, yesterday, after the market closed, Calix issued a news release, which was furnished on a Form 8-K, along with our stockholder letter, and was also posted in the Investor Relations section of the Calix website.

Today's conference call will be available for webcast replay in the Investor Relations section of our website. Before I turn the call over to Michael for his opening remarks, I want to remind everyone that on this call, we will refer to forward-looking statements, including all statements the company will make about its future financial and operating performance, growth strategy and market outlook, and that actual results may differ materially from those contemplated by these forward-looking statements.

Factors that could cause actual results and trends to differ materially are set forth in the fourth quarter 2025 letter to stockholders and in the annual and quarterly reports filed with the SEC.

Calix assumes no obligation to update any forward-looking statements, which speak only as of their respective dates. Also in this conference call, we will discuss both GAAP and non-GAAP financial measures. A reconciliation of GAAP to non-GAAP measures is included in the fourth quarter 2025 letter to stockholders.

Unless otherwise stated, all financial information referenced in this call will be non-GAAP. With that, Michael, please go ahead.

Thank you, Nancy, and good morning. We closed 2025 with the best performance in the company's history, expanding our strong foundation that we have invested 15 years building with team members, customers and partners.

Record revenue, our sixth quarter of consecutive revenue growth while guiding higher in first quarter. Record gross margin, our eighth quarter of consecutive margin improvement expanding RPOs driven by excitement around our third-generation platform and continued operational discipline yielding our 11th consecutive quarter of 8-figure free cash flow and ending the year with record cash.

These results underscore the strength of our platform model as we contribute to the success of our broadband experience provider customers. We also launched the third generation of our Calix platform in December with more than 300 customers already migrated as of today, and we are working to complete all customer migrations by the end of first quarter.

This marks a significant milestone for Calix and our customers as Calix's agent workforce integrates into everything that we do, while our partnership with Google Cloud allows our platform to be deployed to any customer in the world whether Calix hosted or as a private instance for a large customer. Last, we continued to improve the culture that supports the success of our customers every single day. As evidenced by our 44 culture and 20 innovation awards, adding 16 awards in fourth quarter alone.

Our team is particularly proud of these awards as success starts with people. The Q4 results, our strong culture and the successful launch of the third generation of our platform that adds capacity and capability for existing customers through agent workforce cloud. The opportunity in the MG market with Smart Life and the ability to address new global markets and large customers with private clouds as our team entering 2026 very confident in our ability to continue our track record of enabling customers to simplify operations and go to market, innovate across residential, business, MDU and municipal segments, which enables them to grow.

For their members, investors and the communities they serve at a faster and faster pace. As we exit 2025, it also marks the end of the early adopter phase of the market disruption and with demand visibility at an all-time high, our entrance into a sustained growth phase for 2026 and beyond. Cory, over you to walk through the specifics of our Q4 performance.

Thank you, Michael. In the fourth quarter of 2025, Calix delivered record revenue of $272 million, marking a sequential increase of 3% and a robust 32% year-over-year growth. This capped off a milestone year, which we are adding $1 billion in annual revenue, reflecting 20% growth over 2024.

Our results were driven by continued strong demand for our platform among broadband experience provider customers, who are leveraging our appliance-based platform, cloud and managed services to attract new subscribers, minimize churn, raise NPS scores and ARPU and expand their footprints. The addition of 25 new customers this quarter further demonstrates the broad-based adoption of our solutions.

Remaining performance obligation reached a record $385 million, up 9% sequentially and 18% year-over-year. Current RPOs were also a record at $152 million, representing an 8% sequential increase and a 26% rise from the same period last year. These robust metrics underscore the visibility we have into the ongoing strength of our business model as we see more BXP customers adopt our platform, cloud and managed services, add incremental offerings and win new subscribers.

The accelerating interest in Agent Workforce Cloud is another clear indicator that the BXP led disruption is past the elbow and we have left the early adopter phase and are now in a sustained growth phase. To that end, we have aligned our publicly disclosed financial metrics to reflect the growth and value of our model.

We have split out our clients' revenue and gross margin from our recurring software and services revenue and gross margin. Simultaneously, we have ceased providing metrics such as platform adoption and customer size that were proxies for our progress during the early adopter phase.

The combination of our customers' ongoing subscriber growth with our platform and the strength of appliances deployment resulted in another record non-GAAP gross margin of 58%, representing our eighth consecutive quarter of margin improvement.

The continued expansion is primarily due to the adoption of our platform by new broadband service providers and the success of our VSP customers. While gross margin may fluctuate quarter-to-quarter depending on customer and product mix as well as memory costs, we remain confident in our ability to drive further gross margin growth as our platform, cloud and managed services scale.

Regarding memory costs, we will remain proactive and as we did during COVID, we will partner closely with our customers to ensure continuity of supply and address any cost increases resulting from higher memory pricing.

Our balance sheet remains strong. DSO at the end of the fourth quarter was an industry-leading 35 days. Inventory turns were 3, reflecting investments to address robust demand. We generated record free cash flow for the quarter of $40 million.

We have now produced positive quarterly free cash flow for over 5 years including 11 consecutive quarters with 8-figure amounts. We ended the year with record cash and investments of $388 million, an increase of $48 million sequentially and $91 million year-over-year. This strong cash position reflects our ongoing profitability and disciplined operational execution.

Also during the fourth quarter, we deployed $17 million to purchase 300,000 shares of our common stock. As we have discussed, we have a disciplined capital allocation process and we remain a disciplined buyer of our own SOC. That said, as visibility increases, our internal valuation models have moved up. and this is illustrated by our fourfold increase in share buybacks from the third quarter to the fourth quarter.

Furthermore, our Board of Directors has authorized an increase of $125 million in our stock repurchase plan. Given the robust demand environment and the strong pace at which our customers are adopting our model, we expect to continue delivering sequential revenue growth, including in the first quarter, which has traditionally experienced slower growth due to seasonal trends.

Our revenue guidance for the first quarter of 2026 is between $275 million and $281 million representing a 2% increase at the midpoint over the prior quarter. This outlook reflects our confidence in the multiyear growth opportunity ahead.

As more service providers recognize the value of transforming into a broadband experienced provider. Regarding progress on BEAT, we now have a clearer view of the size and timing of the program. The available size of the opportunity for Calix is between $1 billion and $1.5 billion.

While we have already seen orders from BEAT recipients, we expect deliveries of appliances later this year and meaningfully ramping into next year and beyond, providing a tailwind to our growth.

Strategically, the BEAT awards also give insight to the practical differences between fiber-to-the-premise technology and low earth orbiting satellites. The vast majority of funds, some 85% went to fiber-based deployments but only 5% of funds went to low earth forming satellites. This speaks to pure physics. This speaks to physics pure and simple.

Fiber has the highest bit rate pairing capacity a multi-decade life cycle and the lowest cost operating costs. In short, it is financially practical to get fiber to the premises you will. For those sites that are too far away, alternate technologies such as fixed wireless or lower orbiting satellites are a good solution.

As such, one should quickly realize to the extent that competition exists, it exists between fixed wireless and satellite, not between fiber and anything else. For the first quarter of 2026, we expect non-GAAP gross margin to remain strong with some near-term impact due to customer mix and from overlapping cloud costs as we transition to our third-generation platform.

I would also note that while we are making this investment in running dual cloud, the transition to the third-generation platform is on track and progressing well. Regarding non-GAAP operating expenses, we expect a sequential increase in the first quarter of 2026, and primarily related to accelerating the development of AI functionality and capabilities across our platform, cloud managed services.

Importantly, we expect to return to our target financial model for operating expenses by the end of 2026, positioning us for sustained long-term profitability and growth. We are entering 2026 with strong visibility and confident in our growth trajectory.

We are excited to host our Investor Day at the New York Stock Exchange on February 24, where we will share more about our strategy and long-term growth opportunities. We look forward to seeing you there. Michael, thank you.

Thanks, Cory. At a time when the pace of change is accelerating at a rate that has never been seen before, we entered 2026 ready to make that pace of change our advantage and, in turn, an advantage for our customers.

We have successfully launched agent workforce and demand visibility is at an all-time high. We have executed with operational rigor to ensure that we have the financial strength to continue to invest and grow by winning new customers and helping our existing customers differentiate and dominate in the markets they serve.

Last, our team culture is ready for the opportunity that we have invested and worked so hard towards our better, better, never best cultural mantra will ensure that our internal teams make the most of AI to improve how we operate while our platform delivers incredible results for our customers. The next step in Calix' journey is here. I'm excited to lead the team as we speed our ability to transform the broadband industry and enable the success of our customers and partners. I would like to close by thanking our team, customers, partners and shareholders whose passion, grit and trust have brought us to this exciting next stage in the Calix journey. Nancy, let's open the call for questions.

Darryl?

[Operator Instructions] Our first questions come from the line of Samik Chatterjee with JPMorgan.

This is Joe Cardoso on for Samik Chatterjee. Maybe for the first one, I think last quarter, you talked about a revenue growth outlook for '26, tracking to the low end of the 10% to 15% range ex bed. One, curious if that range is still fair to think about for modeling assumptions here?

And then second, just in regards to the better visibility now that you have around the bead program, any update on how you're thinking about the contribution for 2026. And how we should think about that layering into the financials here as we progress through 2016 and into '27 and then I have a follow-up.

Yes. I think with our high visibility and understanding where feed is coming in, I think we'll be somewhere in that range of 10% to 15%. I don't think we'll be at the low end.

And we're confident demand visibility is high. And the other part is for an update. That's one of the things we intend on walking through, right, Cory, at the Investor Day at length because a lot of our investors are asking questions with regards to how we achieve those growth rates. So we're very confident.

Got it. I appreciate the color there. And then maybe just a follow-up in a similar vein, I think you've been referencing new market expansion, including the international opportunities. Any updated thoughts there? Like as we think about the new markets and Calix going out there and trying to cultivate these opportunities, like how should we think about that as part of the 2026 growth story?

And I think last quarter, too, you referenced that, that's not necessarily in the numbers, but how tangible is that in terms of contribution for 2026?

Or should we think about this being longer in the tooth as you try to like kind of do the block and tackling in terms of opening up these opportunities for Calix. Just curious big picture there, like how we should think about that coming into the Calix story.

It's a great question. So the first part of it, as we -- when we had this call last quarter, we were also coming up to the launch of our platform, and therefore, we wanted to ensure that it was out. So from a confidence point of view, for us to actually understand how we'd potentially open up new markets, we had to get over the hump of actually getting the platform out because that's the grand enabler, and that means we have to move our customers over.

So the great thing is that in December, we started those transitions. And as I said on the call, we have over 300 customers already over. I believe we're going to load another 100 over the next week. And by the end of the quarter is our goal to actually have all of our existing customer base onto the new platform.

So that, in essence, is kind of the first gate and we're through that and very confident with regards to the trajectory on that transition. So that's the first part, which is great, which is why as we look out to we can feel confident with regards to the expansion in international markets and the ability to go after large customers with dedicated private clouds.

What is the revenue in 2026? We actually haven't hasn't really built that. We have not added that into the numbers. primarily because, as you know, as you go after large customers or open new markets, you have a lot of investment work to do and that you have to build it out.

That being said, if you look at my LinkedIn, you'll notice that yesterday, I announced that I will be keynoting at Mobile World Congress in Barcelona in March. This is the first time we've really showed up at that event, which is the largest communications events in the world, I think it's like 140,000 people attended. They take over Barcelona and I've personally been there 10 times when I was at Microsoft and Bell. But for us to be keynoting should be a very good indicator that everybody goes and talks to them about what is your story and when they heard what we're doing with regards to artificial intelligence and how far ahead we are, which is enabled by our platform, they were quite surprised and gave us an opportunity.

Again, I would encourage you to look at who's speaking. We're in the architects of AI section, which is [indiscernible] go look at the speakers in there and you'll see that the CEO of Qualcomm amongst others, that these are the leaders in AI. And so -- and we're right in the middle of that. And as we go and accelerate through '26, while we haven't layered that into the revenue numbers, we will be getting a lot of attention because what we've done is groundbreaking and we're way ahead of the competition.

So there is no one doing what we're doing in the segment. The last part I'll say on that is the value of artificial intelligence is yet to be seen. So when I go back to 5 years ago, I was in a course at Wharton, and I was talking to their AI and data scientist leader, and he said, as I look at artificial intelligence over the long term, the real value in that is how we drive business outcomes, not necessarily the AI engines.

And so as we're working through our platform, the value is how we actually work with agent workforce and provide value to our customers. And that is the ultimate value of what we're doing with the platform as it unlocks at the end of the quarter for all of our customers and future prospects.

Our next questions come from the line of Ryan Koontz with the Needham & Company.

Mike, if you could expand a little bit on -- you just spoke to your traction upmarket with Tier 1s. Can you maybe unpack where your entry points there are upmarket where you see the most exciting opportunities with the new platform?

Well, so what I stated was that we're starting that pursuit. And so the entry points into Tier 1, there's really 4 vectors to it. So the value of our platform is that we can -- when we talk to a large customer, we can code broadly with our sales organization around what are the problems that you currently have.

And so the vectors are you have to find your beachhead into an account. So the first vector is you talk to their enterprise organization around what they're doing in business, so small business and that represents a significant opportunity.

MDU as an example, is a significant issue inside every single large customer and small customer frankly, which is why we built that ground baking product to help them succeed. And that opens up a significant market opportunity, both in our existing base and in Tier 1s. So that represents a beachhead. The second beachhead is what we're doing around networking. That's much more traditional based upon what we built with AXOS and how we won Verizon 7 years ago, and that really comes down to the insights that we provide from an automated platform.

And obviously, when you put artificial intelligence on it and because we understand end-to-end from a networking point of view, what our customers do, not only from an access but also from a subscriber management, all the other component parts we can automate that at a level that no one else can, which drives significant value. And as evidenced by the reason why we got sole-sourced at Verizon all those years ago is because we represented an 80% reduction in operating costs.

And the way they really acquired that operating cost benefit as they had to build a lot of their own clouds. We now have that inside our cloud with our AI engines through Operations Cloud and those capabilities. So that's the second way in. The third and most important one is actually demonstrating a model for how you provide subscriber experience.

So how do you win that consumer and residential market. and the value we offer there is evidenced by all of our existing customers who can achieve NPS is as high as 94, they can offer incremental services. But more importantly, they can move at a really fast pace with new product launches as evidenced by Bright speed, who is able to get to market in 2 days, where I was talking to a Tier 1 the other day, they launched a small business product.

It took them 2 years and $20 million. we could have launched that small business product with our platform for $50,000 or less, no integration costs, no IT work and they could have been in market in 30 days. And so that represents a significant opportunity.

And if you look at the challenge inside large customers, the challenge that they have is how everyone's talking about network automation and cost cutting, Well, the real growth path to growth is how do I add subscribers, how do I grow revenue per subscriber? And how do I reduce churn?

And that's something we do all day long. And when you layer in agent workforce, as we build out our Agentic Army, we're going to automate those capabilities to help them drive it a significant fast rate. And then the last one is an opportunity for us to speak to them around Agentic at a higher level. If you look at the platform that we've built, and we all go in-depth in this in at the Investor Day, so I'd encourage everybody to come Investor Day, and we walk you through our broad architecture with regards to AI.

Our architecture is not constrained to our cloud and what we do. We can actually take our knowledge and orchestration layer and apply it to the entire business and use Agentic framework that could provide knowledge across the company and help them drive efficiencies way beyond our existing business. And so those are the 4 beachheads that we have with the customer. Those are the conversations that I am personally having, yes, with the sales teams and the product team as we talk to our customers.

And that's why we will be at Mobile World Congress for the first time. I haven't been there for 5 years. We will be there speaking to the large Tier 1s because frankly, what we've done is unmatched. No 1 is doing what we're doing. And so that's the opportunity ahead. But as I said, coming to the Investor Day, and we will go through that in depth.

That's great. We'll be there and hope to see you over in Spain, too. And Cory, just in terms of gross margins and memory costs, you've kind of mentioned partnering with customers and passing through some of these transitionary costs. Can you maybe unpack that a little bit for us about your thoughts.

So Ryan in the first quarter, we're really not having any of those costs. Remember, the advantage of our supply chain team is we got ahead of this. And so we're doing really well in that regard. So no immediate near-term impact. And it's one of those things that as we progress through the year, we'll partner with our customers to deal with what we see coming down the pipe.

Our next questions come from the line of Michael Genovese with Rosenblatt Securities.

So I wanted to talk about some of the new disclosure numbers that we have in the shareholder letter, systems and software versus appliance. I guess, to begin with, I mean, it seems like as a percentage of revenue, Software and Systems has gone down over the past year. which is somewhat surprising since we're not in a bad like build more footprint type of environment. So can you just help clarify the reason for that?

Sure. As we have said all along, our software is tied to subscriber growth, and that happens in a consistent and upward trajectory all of the time where you're going to get volatility is going to be on the appliance line, where it can grow and shrink it in a given period. And so what you saw this year is obviously reacceleration of the appliances from 2024 while throughout that entire period of time, the software and services number continues up into the right and grows every day.

It's the reason why we say our margins are going to continue to expand over time. because that's unrelenting in terms of its growth, and that's what you're seeing. Specifically, if you look at the tables or the charts related to software and service and you'll see that there was a bit of a downtick from Q2 to Q3, the 1 anomaly inside of the software line that can create quarter-to-quarter fluctuations, the amount of AXOS licenses. As you know, those are recognized immediately upon signing and they are not ratable.

And in Q2, we had a little bit more AXOS licenses than any other period. in the first quarter or third quarter. Consequently, you see that bump up.

Great. And then related type of question, I guess the other thing that surprised me about those charts is how high your appliance gross margins are and how the software and services gross margins are only a few points higher than the appliances. So the questions that follow from that are how are the appliance margin so high.

But secondly, as you sort of as the clouds mature and you move to one cloud versus hosting on 2 clouds, where -- could you just give us a sense of where those software margin should be headed over time?

Sure. Sure. If I take a look at the software margins, we're going through the transition currently. And so they are temporarily depressed due to the dual cloud cost but once we lift that yoke off, those margins will continue. Ultimately, they probably go past 70% and beyond.

We ultimately don't know what the upper limit is to the software and services margin because it depends on success at Tier 1, where the amount of that will likely be just software-only revenue with no hardware or appliances attached good to it. And so the more of that you mix in, it's hard to say where that ultimately adds and tots too, but they should clearly move beyond 70%.

Great. But appliance -- on the high appliance margins, just how you accomplish that?

But we accomplished it because of the fact that if you look at the $2 billion that we've invested in our platform, A big part of that is that we built 2 operating systems, which are fully abstracted from the silicon and give us significant flexibility with regards to components.

The reason why a customer can turn up a system like Wrightspeed, who has older back-office systems that are very complex from they brought those over from Lumin initially. The reason why they can actually do what no one else in the industry can do is because of our platform. You -- because everything is abstracted, the complexity sits inside the software, not at a hardware layer, which allows us to actually transition at a very fast pace.

It also allows us to simplify our SKU count. If you go back to where Calix was when I first started 10 years ago, we had almost 4,000 SKUs, and that means that you have all of this component complexity if you can actually build out an abstracted operating system from the systems and the appliances, what you gain is massive scale with regards to buying single components and putting them across all of your systems, which means that you get higher margins.

And so that is ultimately our silver bullet is that we have built truly abstracted in the network. We're the only ones who have done it. And on the other side with regards to our premises, it's the same thing. And that's why we have a low SKU count maximize inventory and simplify significantly.

And I'll add a couple of things to that. I mean clearly, the appliances show the differentiated value in our model. And there's a couple of other added. We've got a very tight fit to the market requirements. That allows us to obviously reduce SKU count but also allow us to reduce the power consumption that's an add on the access side.

And on the premises side, we've got some clever design that allow us to cover a number of use cases, further shrinking SKU count. When you shrink that SKU count, not only does that create benefits for Calix in terms of the amount of SKUs that we have to manufacture right? It reduces the risk to E&O, excess and obsolete inventory produces the overhead.

So our cost could be lower. But that lower SKU count also translates to a benefit to the service provider in terms of them managing inventory, the number of SKUs that are on the trucks, the amount of their components in terms of the spares depot.

So all of that simplification that we're driving through our platform translates to a differentiated value on behalf of the service provider as well. So I would add those points to what Michael said.

Our next questions come from the line of George Notter with Wolfe Research.

For the disclosures on the software side of the business, I think it's terrific. I'm just wondering, when I look at software and services together Wondering how much of that is the services piece. If I go back to like 2022, you guys used to break out services. You think back then about $10 million or $11 million per quarter was in the services line. was a services-only line.

I assume it's still at that run rate or maybe a bit bigger given the growth in the business. I'm just curious on what how much is software and how much is services. I've got another question as well.

Yes, George, we're not going to go into further breaking that down. But you can use that as a proxy and...

Service is a small component of our business. We're a software company and a cloud company and we're not a services company.

Okay. And then also on the software piece, I guess I'm wondering how much of that software is recurring versus perpetual. Any sense for what that might look like?

It will fluctuate from quarter-to-quarter, George, like I said, with AS. But the majority -- large, large majority of it is recurring.

Okay. Great. And then I also was just curious on kind of where you are with the agent workforce Cloud. Obviously, you're rolling it out, you're putting the platform out there, which is terrific. I guess I'm just curious on what that looks like in terms of the timing of the revenue ramp, how you guys are monetizing it -- anything you can say there would be great.

Yes, we're -- that's something we'll cover at the Analyst Day. And in dwell actually have demos and those elements, we'll be able to show those things. .

Again, the most important part of it was we had to actually get the platform out, which enables it to happen. That's in progress right now. And therefore, what you're going to see is a rapid. What happens is that once the platform is out Think of it as like building a house, right?

The plumbing is done, the walls are in, and now we can actually do what agents are, which is the cool stuff. Put in the windows payment paint the house all those different elements, which is add a ton of agents. And so you're going to see a very active ramp of our capabilities. And that means that the customer value will start to roll in as we go through Q2 onwards. And so let me give you a simple example. I was speaking to a customer who actually signed an AI contract with us just after Christmas. They went all in.

And one of the things -- the reason why he said that was he is a very good customer who believes that we will deliver quickly, and he actually, we showed him when he's going to be transitioning over and how that rolls out for him. And he said the greatest problem that he has is that he had never bought Engagement Cloud, which is our Marketing Cloud because of the fact that we did not have the capacity and capability to do it.

His team members were not sophisticated enough to do some of the micro segmentation that it provides. And he said, now I'm super confident because agents and as I shared at Connections, where I walked through, and so I'd encourage anyone on the call who hasn't actually watched my connections keynote, you can go look at the marketing element and walk through what is our view of how an Agentic workflow works segmenting down to the individual of on parsing out what is their social media preferences so where do you place the ad allowing for very low-cost ads but more importantly, understanding all the white space through agents and what the upsell cross-sell opportunity is.

By having that in an agenetic framework in agent workflow, he can now take somebody who's not sophisticated have to actually do that on their own because they're not the data scientists, et cetera. And in his market, he can now aggressively go after those customers in a very different way.

And so he's all in. And so I think from the end of the quarter, as we roll off, you're going to see our resources also roll out of the plumbing element of our project and those resources free up to go hard at building agents and expanding it quickly.

And the agent part, frankly, when you talk to our product team, building an agent is easy. It's a bunch of Python code. The magic is in understanding the workflow and if you look at our 3 clouds, Operations Cloud, Engagement Cloud, which is marketing and service cloud, all those are is hard-coded workflows and so we actually know exactly what agents to build and how to assemble them in an orchestration layer to deliver the value.

And you're going to see that in spades at a pace that no 1 can match starting the end of the quarter as we finish up the migration. Again, come to Investor Day, we'll cover that in depth.

You're going to see it first in the RPO number.

Our next question has come from the line of Tim Savageaux with Northland Capital Markets.

I wanted to come back to the discussion and appreciate the quantification of the opportunity here which I assume, I don't know, stretching over, what, 3 to 5 years, and they can be pretty significant on an annual basis, and you mentioned a bigger ramp into '27.

And I guess my question is, as you look at that opportunity, I guess is it apparent to this point -- would that be fully incremental to your current run rate business? Would it add to it -- how should we think about layering in what could be, I don't know, a couple of hundred million dollars a year relative to your current run rate business and would you expect another step function into a higher growth rate in '27 something maybe in the 20s relative to your 10% to 15% target.

Thanks, Tim, for the question. I would say if you want to start talking about 2027, please come to the Analyst Day. That's where we might provide some color on that. As it relates to additive revenue, you got to remember that there's a limited number of resources that the industry has to go do this work. .

Crews are not just sitting around waiting for bead. They're going to go to do projects. So some portion of that would come at the cost of other work being done because these free were locations that they would be otherwise built, and now they'll go do some.

And so there's some of that. So ultimately, for this to be additive, you're going to have to increase your capacity and rates at which you're going to go deploy. But the important part is this opens up more of the premises revenue right? As they go out and build these networks, you're going to then start hanging new subscribers off of them.

And so that's the aspect that kind of adds to the acceleration of revenue, which we're more excited about. In addition to the bed revenue alone.

Our next question has come from the line of Scott Searle with Roth Capital Partners.

Nice to see the RPOs continue to hit record highs and continue to post good year-over-year growth numbers. Maybe to follow up on George's question, and I suspect you might be referring me to the Analyst Day, but as we start to get full commercialization across the installed base of the third gen platform and Agentic workforce and Smart Life, I'm wondering if you could talk a little bit to the time to monetization of when you're expecting to see that start to ramp? .

It would appear kind of given the features and functionality of the platform that we might start to see some of that contribution starting to accelerate in the second half of 26. And I wondered if you could also kind of put that in the context of historically, we've talked about going from $1 to subscriber to $10 per subscriber. Kind of how you're thinking about that?

And do we start to see software and services in late '26 and '27, starting to inflect above RPO growth, particularly as RPOs, I believe, are just reflecting minimum contract revenue levels and not as we start to see incremental monthly subscriber revenues on top of that.

So a bunch of things rolled into there? Basically, when do you think we start to see the monetization of the third gen platform and kind of the inflection point of when that sort of hit more critical mass?

Scott, I think you're right. I think second half, you start to see that, and it will be reflected in RPOs. I also think that as we stated, we're entering a sustained growth period because we have this significant monetizable base that's already in place.

And one of the things that has been constraining later adopters is the fact that they don't have the capacity and capability to actually deploy the technology and win in their markets.

And so if I go back to the example that I just provided in the previous one, when I talked about it was a customer who did not have the capacity in the form of -- I don't have a data scientist, I don't know if that's some of those capabilities or the capability in the form of -- if I think about marketing, you can do broad -- most service providers regardless of size, small ones, 2 large ones to very, what I would call, unsophisticated marketing.

It is brand-based or it is price and speed. And I don't really customize my offerings to the needs of the individual. What we're introducing with Agent Workforce is the ability to go through and do significant micro segmentation of a customer base and get down to a segmentation of one.

So looking at a household and understanding all the behaviors and the needs of that household and then at a very low cost hitting a button and providing a custom campaign into the social media that is relevant for them. So as opposed to spending $10 and blasting it out on to everyone, I can spend $0.05 and buy an ad on Instagram from 8:09 p.m. which is a social media channel of preference with an ad that actually hits the sweet spot of their white space.

So these are -- represent significant opportunities in the past for barriers because you needed to be a large company with a data scientist team and all the other things.

And frankly, even if you were a large company, the truth in this industry is that all of these companies have tons of data scientists, and they generally serve the C-suite, they do not have the capability to do that either because you can't actually do mass segmentation down to a small level to serve the sales and marketing team, we're trying to close deals with customers when new subscribers and drive upsell and cross-sell.

Best example I can give is that when I was an executive running a $900 million business at a large Tier 1 telco, we had '27 data warehouses, and I had to go like somebody 1,000 scientists and inside of my $200 million P&L, I had to go build out a $10 million data warehouse for my team because I couldn't get the data that I wanted. And so this -- our ability to unlock the data in their business and automate it so that they can do the most important thing in their business, which is not cost.

It is when new subscribers and grow revenue per sub while reducing churn is our magic. And so this is a huge enabler for us as we go, just like you said, in the latter part of '16 and through '27, more importantly or as importantly, at the same time, opening up new markets. And so when I'm at Mobile Congress and other and in these events, those are the conversations we're going to have because everybody is talking about how to cut costs, the magic is how to make money.

And this is what we're really good at. We're going to help them make lots of money. And in the past, we had to convince them to do the actions based upon best practices. Now we can say, here's what the data says across 1,100 customers, 1,200 customers Here's what the here's the train agents to make it happen, plus the bucket, cracks the bucket and it works.

And so this is a seismic shift in what's possible in our industry, and we are at the forefront of it, which is why we've been quiet about it since November of '23 as we've been building it all out. and now where guns to go.

That's very helpful. And if I could just from a follow-up. Given the third-generation platform and now that you could start to address larger customers with private clouds.

You also have an evolutionary path with the hybrid architecture, no stranded assets for those customers. I'm wondering 2 things. How is that conversation and dialogue going now with the Tier 2s and the Tier 1s in terms of what you're able to offer now with the new platform?

And what's the timing of when we should expect to see some incremental customer contribution and/or monetization of those Tier 1s, Tier 2s and potentially international customers?

With the Tier 2 customers, that's been happening because the Tier 2 for the most part, is a super regional. Like if you think about Tier 2 is that they're generally across 5 to 20 states, they're not global in nature and really a lot of their needs.

They still have -- they also have some of the constraints that you see in the smaller customers in that their ability to invest do I invest my dollars and building more fiber? Or am I going to be building out all these data science teams and all the other things.

So I think those conversations are happening very actively -- for larger customers, as we said, we talked about the Tier 1s that we've already actively engaged with, and that's going to take a step up as we basically unveil what we're doing at Mobile Congress to the world and to the Tier 1s in partnership with Google.

So the other part of this is that we've transitioned ourselves away from AWS and everything that we're doing is on Google Cloud. And Google is an incredible partner. Not only are they interested in our go-to-market strategy because of the fact that we offer workloads on their Google Cloud that they can talk to their customers about our insights and capabilities, specifically our business insight on how to run a business represents a market shift and an ability for them to differentiate against their competition when it comes to cloud.

And so we will be at Mobile Congress in the Google booth, downing and partnering with them, talking to large customers. That's a the revenue stream on -- sorry, time line is that large enterprise accounts have been doing this all my life, and those are 18 to 24 month sales cycles.

So you're talking we're going to be going into a lot of pause through '26. We're going to be demonstrating it and educating them because that's the other part. The large customer mindset is I'll build it myself and generally very siloed. So they'll take an individual use case in the build a silo and then another area of the business, we'll take a use case of the problem that they haven't built their own silo, whereas we represent a significant shift in mindset and that we step into a large Tier 1, which here is our genetic library that's fully trained, pointed out your data and it goes, oh, by the way, here's our agent tool tin on how you build out your own agents and build those into the orchestration layer.

So it's going to require a lot of education. So it's a latter part of 2016 and definitely growth trajectories for 2027 on top of the earlier questions with regards to the tailwinds on BEAD.

Our next questions come from the line of Christian Schwab with Craig Hallum.

Most of my questions have been answered. But just a quick follow-up on the feed. Given your historical strength in the smaller of the regional telco companies. Would you assume that you would get 50% market share over a time frame? In the TAM that you guys outlined for bed is 50% a good starting point or now that you've probably gotten more color exactly who has money that it could potentially be bigger than that?

Christian, thank you for the question. As you know, we do very well in that segment, and I suspect we'll continue to do very well as it relates to the...

By the way, let me expand out on the leading, right? So BEAD is the infrastructure that goes into pass a home pass is not a home one. So we have incredible technology to drive the lowest operating expense to run a network and to automate it.

And the third generation of our platform takes what we can do for network automation to the next level. Everybody else is -- no 1 has an abstracted operating system, no 1 has built subscriber management into the platform, all the different capabilities that are critical to run a headless network.

And we can run a headless lights out network top to bottom and significantly eliminate power issues because you're taking 3 or 4 boxes collapsing them down to 1 box. And at the same time, give the highest level of reliability because the data path is inside the operating system, not service chain across multiple boxes.

All those things come together to run great networks is what Calix does. But then the other part, and this is the most important part is that as be rolls out, they still have to go and win the subscriber. That just puts in place that I can get the network in place, but I have to go win the subscriber.

And so regardless of whether -- what our market share is for bead, whether or not we win the network or not, what our value proposition, which is going to get significantly stronger, especially in these early stages for our regional customers, is the capability to win new subscribers.

And with this capability of allowing our customer success organization to transition from same to core ashes a service per Cory, here are the things that you need to do to win more subscribers, grow revenue per subscriber and reduce churn and then beg quarry to listen because we're not the ones who actually come up with this advice.

It's best practices based upon looking at 1,200 customers. And then with that the advice we give Cory based upon seeing everybody successes and failures. And then begging Cory to do something and Cory's saying, I'm stubborn, and I don't want to do it, we can now transition to our success organization saying to Cory, here is the best practices.

Here are the yields it can provide -- by the way, press the bucking, this is what the agent workflow will actually do for you, so you don't have to do the work for yourself. And so this mark this, in the end, is the promise of agents -- this is the promise of what you can do with deep insight with regards to how to run a business and then apply a genetic on top of it.

No one cares about the LLM, the LLM is commodity. And we can use all the LLM we can pick whoever is best for whatever the use case is, but the value here is the business insight and then saying to Cory push the button. And this is the marked transition.

I talked about crossing the chasm forever. We have crossed the chasm.

We're on the other side because we now have the capability to, in essence, help them automate their business and take the capacity and capability constraints off the table which is a market opportunity for us to accelerate growth because in the end, if you want to talk about software and cloud, our goal is to help our customers win subscribers and grow revenue.

And when they do that, we get a portion of it. And so unlike everybody else, that's what our goal is, help them win by driving money.

Thank you. We have reached the end of our question-and-answer session. And with that, I'd like to turn the call back over to Nancy Fazioli, for closing remarks.

Thank you, Darryl. Calix will participate in several investor events during the first quarter, most importantly, hosting our Investor Day at the New York Stock Exchange on February 24, as Cory and Michael referenced. Please register to join us.

Information about these events, including dates in times and publicly available webcast will be posted on the calendar page of the Investor Relations section of calix.com. Once again, thank you to everyone on this call and webcast for your interest in Calix, for joining us.

This concludes our conference call. Have a good day.

"

"

"

"

" JPMorgan

" Roth Capital Partners

" Wolfe Research

" Craig Hallum

" Northland Capital Markets

" Needham & Company

Greetings, everyone, and welcome to the Calix Third Quarter 2025 Earnings Conference Call. [Operator Instructions] As a reminder, this conference is being recorded.

It is now my pleasure to introduce your host, Nancy Fazioli, Vice President of Investor Relations. Thank you, Nancy. Please go ahead.

Thank you, Latanya, and good morning, everyone. Thank you for joining our third quarter 2025 earnings call. Today on the call, we have President and CEO, Michael Weening; and Chief Financial Officer, Cory Sindelar.

As a reminder, yesterday after the market closed, Calix issued a news release, which was furnished on a Form 8-K, along with our stockholder letter and was also posted in the Investor Relations section of the Calix website. Today's conference call will be available for webcast replay in the Investor Relations section of our website.

Before I turn the call over to Michael for his opening remarks, I want to remind everyone that on this call, we will refer to forward-looking statements, including all statements the company will make about its future financial and operating performance, growth strategy and market outlook, and that actual results may differ materially from those contemplated by these forward-looking statements.

Factors that could cause our actual results and trends to differ materially are set forth in the third quarter 2025 letter to stockholders and in the annual and quarterly reports filed with the SEC. Calix assumes no obligation to update any forward-looking statements, which speak only as of their respective dates. Also on this conference call, we will discuss both GAAP and non-GAAP financial measures. A reconciliation of GAAP to non-GAAP measures is included in the third quarter 2025 letter to stockholders. Unless otherwise stated, all financial information referenced in this call will be non-GAAP. With that, Michael, please go ahead.

Thank you, Nancy, and good morning. I just returned from the incredible experience of Calix Connections, where we celebrated customer-driven innovation on our unique platform and the success that the platform drives for our customers by enabling their teams to win new subscribers, grow revenue per subscriber and reduce churn.

To stand in front of that crowd and celebrate NPS scores as high as 94 is a testament to the partnership and trust we have built with our customers and they with their subscribers and the communities they serve. At Connections, we officially launched the Calix Agent Workforce, our end-to-end integration of Agentic AI into everything that we do via our third-generation platform that will launch in partnership with Google this quarter. This marks the next stage in our company's ongoing evolution to help our customers simplify operations and go to market and innovate with our platform, enabling them to grow for their members, investors and the communities they serve. The ability of our customers to dominate their markets when armed with our unique and highly differentiating platform and managed services model was on full display in our results.

In the third quarter, the Calix team achieved record revenue in our fifth quarter of sequential growth while guiding higher in fourth quarter. We set another gross margin record, our seventh consecutive quarter of margin improvement. We also had 20 new customers choose the Calix platform to dominate the markets they serve and RPOs grew sequentially.

At the same time, the team maintained a rigorous focus on operational performance and the balance sheet with OpEx investments returning to the target financial model while yielding our 10th consecutive quarter of 8-figure free cash flow, ending the quarter with record cash.

It was another great quarter of performance by our customers and our teams who support them as we launched the culmination of our vision where those with the Calix platform, managed services and access to data are best placed to succeed through the power of Agentic AI. Corey, over to you to walk through the specifics of Q3.

Thank you, Michael. During the third quarter of 2025, we delivered record revenue of $265 million, reflecting sequential quarterly growth of 10%. Our overperformance relative to our guidance reflected continued robust broad-based deployments from our BXP customers as they added new subscribers and footprint expansion as they continue to choose Calix for network upgrades, new builds and competitive displacements.

RPOs grew 2% sequentially to a record $355 million and increased 20% year-over-year. Our current RPOs were $141 million, up 5% sequentially and up 28% year-over-year. This metric is a strong indicator of the strength we are seeing from our platform cloud and managed services model. The combination of our BXP customers winning new subscribers and strength of Access Edge deployments led to another record of non-GAAP gross margin of 57.7%, representing a 90 basis point sequential quarterly increase.

Our balance sheet metrics were strong. DSO was 30, inventory turns were 3.8 and free cash flow was $27 million. We have produced quarterly free cash flow for over 5 years, including 10 straight quarters with amounts in the 8 digits. We ended the third quarter with record cash and investments of $340 million, an increase of $41 million sequentially.

Moving to guidance. Given the broad-based demand picture and the rates at which our customers are deploying products, we believe we can continue to grow revenue sequentially even with the significant overperformance achieved in the third quarter of 2025.

Our revenue outlook for the fourth quarter is between $267 million and $273 million, which at the midpoint would represent a 2% sequential increase in revenue and reflect revenue growth of 20% for the fiscal year as compared to 2024.

Our non-GAAP gross margin guidance for the fourth quarter at the midpoint would represent a slight increase from the third quarter and reflects our expectations regarding customer and product mix. This guidance for the fourth quarter means our gross margin improvement for fiscal '25 will exceed the higher end of our target financial model of 100 basis points to 200 basis points. And regarding non-GAAP operating expenses, we expect OpEx will increase sequentially primarily related to investments in Connections and to accelerate the development of AI agents and functionality to our platform into the first half of 2026.

That said, we expect to be back within our target financial model by the end of 2026. Michael, back to you.

Thanks, Corey. This is an exciting time for Calix and our customers. The pace of change that AI is injecting into the market is like nothing we have ever seen in human history. As I shared at Connections on stage and with the 350 general managers and CEOs who attended our leadership track where Netflix took 10 years to get to 100 million subscribers, OpenAI took 2 months. The pace of change is mind-boggling and for many, it's overwhelming. For Calix, that pace of change is our advantage and an advantage for the customers who leverage our platform and managed services.

We've invested 15 years and $2 billion to put in place the foundational building blocks in the form of Access Edge, Experience Edge and Calix Cloud to make the most of the Agentic AI opportunity that is ahead. Our migration to our third-generation platform in partnership with Google allows us to support the success of our existing customers while expanding internationally into new geographies with local sovereign data centers and support large Tier 1 customers with a dedicated instance of our entire platform. We are already well into the fourth quarter. And while Connections is complete, thousands of attendees will be educated on the AI-enabled opportunity ahead as Connections on Demand, our virtual program launched yesterday. This is in addition to the on-demand replays on calix.com.

At the same time, our customer success team is revamping how they support customers with the Calix agent workforce to speed transformation and expand our impact on our customers' ability to add new subscribers, grow revenue and reduce churn.

In addition, our internal enterprise teams are focused on improving how Calix operates and improves with artificial intelligence. With the insights that our product teams have gained through 2 years of learning as they build AI capabilities into Calix Cloud, our enterprise teams are uniquely advantaged to aggressively embrace AI in everything that we do internally to improve our operational performance and help us scale at the fastest pace possible.

As I shared in the Fast Company article, the 4 AI questions that every CEO needs to ask to succeed. Over the last 6 months, we have seen employees create 725 exploratory agents with 40 of those agents being selected by our AI steering committee to be scaled across our internal enterprise to drive operational gains across Calix.

The next step in Calix's journey is here, and I'm excited to lead the team as we speed our ability to transform the broadband industry and enable the success of our customers and partners. I'd like to close by thanking our team, customers, partners and shareholders whose passion, grit, trust and partnership have brought us to this exciting next stage in the Calix journey.

Nancy, let's open the call for questions.

Thank you, operator. We're ready to take questions.

[Operator Instructions] Our first question comes from Samik Chatterjee with JPMorgan.

This is Joe Cardoso on for Samik. Maybe perhaps for my first one, very strong revenue performance this quarter. I believe this marks 3 in a row now with results tracking $10 million to $20 million ahead of the high end of your range. Just wondering if you can help contextualize what has been driving this outperformance relative to your expectations at the start of the quarter. And as we look ahead to next quarter, how are you thinking about the sustainability of those drivers and potentially implications to your outlook? And then I have a follow-up.

Thanks, Joe. Appreciate it. The overperformance is driven by a couple of things. One is just the broad-based demand that we're seeing across our customer base. But it's also a function of some of the competitive expansion of our footprint. We're seeing some of these customers, I guess, surprisingly, not only do cap and grow, but some rip and replace as well. And so when you combine that with the broad base from our existing customer base, that's driven the overperformance.

Yes. Let me contextualize a bit of that. So having just come off with connections, what's happening is that our customers are winning more. That's what's happening. And so when our customers win and they add subscribers, then we win because of the fact is that, for example, if they build out a network and then they win a subscriber, that's incremental revenue for us and opportunity to continue to expand.

So because of the fact that we're uniquely positioned in the fact that we have almost 1,200 customers on our platform, and there's not one single driver, all of them are doing better. That means that our revenue goes up, which is also why we guided higher in fourth quarter because we feel very comfortable with the robust demand that we're seeing as Cory identified in the -- or as we identified in the letter.

Got it. I appreciate the color there. And then maybe for my next one, we obviously saw the announcement coming out of Connections last week, lots of innovations being unlocked there, particularly in the backdrop of AI. However, if we kind of take a step back here, just curious how you're thinking about the implications of this innovation cycle for Calix and maybe more specifically on the investment side of things, particularly, it just sounds anecdotally like there's greater appetite to scale more aggressively. I mean, even if I think about you guys mentioning like international regions, like it sounds a lot more aggressive than what you guys have talked about historically. So just curious how you're thinking about the implications to the business model as it relates to more on the investment side, just given kind of this innovation cycle where we're still in the very early innings of. Appreciate the question.

Yes. So from an investment point of view is that we have a target financial model. We continue to make the investments. But the broader one is that we've been investing for 15 years. So we're $2 billion into this platform. And if I think about those investments, as we articulated in the letter, those create the foundations upon which we can grow. So there are -- first of all, in the existing markets, there's significant growth opportunity ahead because of the fact that we're uniquely positioned.

If you think about one of our existing customers, and I talked about this on stage, an existing customer who has the Access Edge in place, which is a consolidated network, which has incredible intelligence in it. Then you have everything that we do on Experience Edge, which includes 2 components: intelligence at the edge in every single WiFi system that exists on a customer premise, but then all of the solutions that sit on top of it, MDU, Smart Town, smart business for small businesses and smart home, all of those come together into the cloud, and we've now invested and put the AI layer on top of it, which means that those customers who have those 3 component foundational elements are best positioned to, first of all, take advantage of AI because it farms all the data, puts it into an AI engine and automates significant components of their business and allows them to scale at a significantly more rapid rate than in the past. And the best example that I can give of that is that in the past, our customer success teams would talk to a customer who has those things in place, make suggestions. But then in the end, it was the decision for the customer as to whether or not they had the capacity and capability to take that advice and turn it into action to grow their business.

Now our customer success organization is going to sit down with the customer and say, we've looked across all your business. We have all the data. We know what the insights are. We know how well you can perform. By the way, here are the AI workflows that -- go press the button and it will do it. You don't have to add capacity. You don't have to hire a bunch of new people. This is going to augment your existing capabilities and provide you a massive jump in capacity. And so when I think about the growth opportunity in our existing markets, it is significant because of that capability to transition from advice giver and optionality to press the button and we'll do it for you. That's the first step.

On the expansions into international and where we go into large customers, this has always been planned. It's been a timing opportunity, and those represent significant growth opportunities in revenue and margin. The reason why we're talking about those now is because of the fact that we've been very quiet on it because we have always been a very conservative leadership team who actually talks about what we can see very carefully. and what we see right there. And with the third generation of the platform turning on this quarter in a few weeks, we now actually see where that expansion goes. And if you took the time to watch the Connections event over Monday and Tuesday.

On Tuesday, we had the Head of Google Telecom up on stage with me talking about how this partnership could flourish. And so it represents a significant opportunity because our platform would sit on top of a Google Cloud regardless of geography and regardless of size of customer, which provides us a significant scale opportunity, frankly, without having to make significant investment other than sales and marketing because it's a cloud. And if you have nothing running on the cloud, you don't pay for anything. So as it scales, we have a great opportunity not only for revenue but also margin.

And Joe, I'll add on to the model question to that when you look at the AI investments that we're doing from an OpEx perspective, in the quarter, you saw us hit record revenue and at that time, get ourselves back into the model. But we also see an opportunity here to accelerate some of the development work that we wanted to do from the latter part of '26 into the first half. So you saw in the OpEx guidance that we took up the level of OpEx. It's for those investments in R&D related to AI functionality, but we will be back in model by the time we exit 2026.

Yes. And to give you practically what Corey was talking about is that if you look at our cloud, again, I talked about this on stage, Calix Cloud, which is Operations Cloud, Service Cloud and Marketing -- or Engagement Cloud, which is the marketing capabilities and engaging with customers, all that Calix Cloud is, is a mapping of workflows on how to run a broadband business.

This is why when it comes to artificial intelligence, we're so uniquely positioned because we are going to -- now that we have the infrastructure in place upon which to build AI, which is the data layer, the knowledge layer, trust and orchestration, which then leads to trusted action. We built all that out. That was a $100 million investment since November of 2023.

Now that, that's in place, the easy part and the candy, the ROI is right there. So we're going to accelerate building out those agents. And if you talk to our CPO, he basically states, building an agent is easy. All the hard work was all the other work that we did. So we did all the hard work. Now we're going to do the easy work, and we're going to monetize this like crazy because of the fact that we can take a workflow in Operations Cloud, again, that I showed, I laid these out at Connections.

We can take an operational workflow inside Operations Cloud and every single step is essentially an agentic agent because it's a task-based entity. We can build all those agents and pop them out and then say to our customers, here's your workflow for anomaly management. Here's your workflow. And while all the other industry people are going to be building out these complex custom workflows, we're actually going to take all the insights that we've built over the last 6 years in Calix Cloud and plow through all these Agentic workflows, which means that by -- as we come out to the latter half of 2026, the AI army that we're going to have turned up inside our platform is going to be unmatched because we know how to run broadband companies. We have all the workflows. We have all the data, we have all the insights, and we have the trust of our customers to partner on building all that out.

So to not speed through this at an incredible pace, would be foolhardy. And in closing, I'll say the last part of it is everybody is talking about AI and around where to invest. One of the interesting things I've been hearing a lot about lately on investment platforms and other areas is all the tangentials affiliated with it. So you have the big guys who are building out their LLMs, which are, in essence, commodities.

The ones who are actually going to provide the greatest value are the ones who understand the business. Well, that's what Calix Cloud is, a complete map of how to run a broadband company, and we are going to AI the whole thing top to bottom over the coming months.

The next question comes from Scott Searle with Roth Capital Partners.

Great job on the quarter and outlook. Mike, maybe just to dive in on the small customer front, it was broad-based. It was strong. I'm wondering if you could provide some other commentary in terms of if there were any pull-ins, the sustainability of that going into 2026 and kind of how you're thinking about that visibility building at this point into 2026 and the sustainability of double-digit growth.

Thank you, Scott. That's a great question. And I want to actually kill this right from the get-go for everybody who has said that we are -- we have a broad-based demand. There is not a single customer that had significantly more influence on anything that we generated in this quarter or in next quarter. It is broad-based demand.

The strength of our business model is that while things can go up by segment and down by segment, it is actually up to 1,200 customers that allow us to actually drive demand. So with regards to visibility, and I'll turn it over to Cory to go through some specifics. But because we're so closely partnered with our customers, we see -- and the reason why we guided higher into Q4 and see the strength in the business is because of the fact that our customers are winning.

It can't be affiliated with, and if anyone writes that, I'm going to be going to kind of go mental on it, is it cannot be written around a single customer because it's around our customers winning more. And as they win more, we win more and our investors win more. It's that simple broad demand across the base.

Yes, Scott. So I think the visibility we have into the demand profile, as Michael outlined, gives us that confidence that the sequential growth is durable. Now we had a large step up here in the third quarter. So I think the sequential increases will may be more muted as we move into 2026. So in terms of our target financial model of 10% to 15%, I think we'll be at the lower end of that range ex BEAD, right? And I'm sure I'll get a BEAD question. So I'll Preempt that. I'll let that question be erased.

Okay. I'll pass on the BEAD question, Cory. But just to clarify, the sequential growth, so we'll see sequential growth from the fourth quarter into the March quarter. And then as my follow-up, Gen 3, right, we talked a lot about AI and Agentic opportunities there. If we look at the current quarter, international was down due to one customer, I think, in the European theater. But the Gen 3 platform is supposed to deliver private cloud capabilities, sovereign data center capabilities. When do we start to see that accelerate?

And Mike, to follow up on your commentary on Agentic AI in general, as you start to think about that driving the flywheel within your customer base, does this poise you guys for actual acceleration of RPO growth as we get into late 2026 and the ability to really drive that recurring revenue?

Yes, Scott, great question. The way we think about the monetization of the AI agent is not so much a separate charge for it, but an acceleration of our business model, right? So our customers will acquire subs faster. They'll roll out more services more quickly. So that's how we'll monetize it.

So yes, it will increase RPOs because as they go faster and adopt more of the platforms and the solutions, that will ultimately translate into a contract value that will get reflected into RPO. So ultimately, yes, that's how that will happen.

So -- and the key word that you used was actually flywheel. In fact, we talked about that from a good to great point of view on a regular basis is that you have this -- that is exactly what Calix is. When I go back to what I -- answering the previous question, the flywheel that we have is that is 15 years of investment that's been getting -- that has been going into our platform, allowing us to actually become stronger and better and stronger and better because we enable our customers to be stronger and better and stronger and better, which is why I go back to the -- is it broad-based demand? Absolutely because we are the enabler for our customers to actually go at a faster and faster pace to do the 3 things that drive growth, which is, one, how do you actually add subscribers; two, how do you increase revenue per subscriber; and three, how do you eliminate churn? And if you eliminate churn, then obviously, your gross adds go up faster. And so that is the power that we see in our platform. And those core components, access Edge. And so if you've actually done the work, and I said this on stage, where if you consolidated your network with everything that we've done, which is collapsing the B&G access aggregation onto a single system, you have the most powerful network that exists with all the data and the insights to run it autonomously.

Same with on the Experience Edge and then you bring it all into the cloud, throw an AI engine on it, and we can help our customers optimize their business. And as they optimize their businesses, they add those subscribers and become more profitable, we become more profitable.

With regards to international, to your question on the one, again, we keep getting these questions every single quarter. It's just lumpy up and down, right? It's lumpy up and down, but we have broad-based demand across all those customers. In no way, shape or form does it signify the strength or weakness of that customer. It's just -- it's a timing thing. And so we go through this every single quarter is that is there any change? There is not. We have strong demand across small, medium and large customers. And then with the international markets, that's really a later 2026 is where we see that we start to expand into the right markets.

But we're going to get our existing base over because, frankly, our existing base is all primed because they have those foundations in place and a huge opportunity for us to make them wildly successful at a very fast pace. So your statement of flywheel is 100% accurate.

The next question comes from George Notter with Wolfe Research.

Can you guys hear me? I was curious about your monetization strategy on the Agentic AI workforce. And is there -- did you guys consider like charging customers a la carte for this capability and maybe being a bit more aggressive on monetizing rather than just sort of giving it away, I guess, and kind of working towards hoping these guys are going to generate more success in the marketplace. I guess I'm just curious why you guys didn't take a more direct approach on monetizing it.

We are taking a direct approach on monetizing it. It's actually a matter of perspective. So there will be some elements that we charge for and then there's others that we will monetize through the -- by helping them drive gross adds -- so we have a very clear monetization strategy, and we're very comfortable as a leadership team that it's going to yield significant revenue and upside, massive amount of upside in revenue.

Got it. So if I look forward and just think about the growth in the company, I think you guys -- Cory, I think you said 10% to 15% or maybe at the lower end of that range. But [indiscernible] sorry.

Without BEAD.

Right. Okay. And then -- so of that growth, like is most of that coming from same-store sales in a sense, like existing customers growing faster or growing more with you? Or is it new customer wins that drive a big percentage of that 10% to 15%? How do you kind of parse out where the growth comes from?

Was, again, broad-based demand across all the different segments. So if you -- first of all, is that existing customers have a bunch of unmonetized footprint. So for example, if I only have -- we have some customers who are at the right market share levels, who are at 65%, 70% market share. That's kind of your, I would say, almost a theoretical maximum when you're in a competitive market. But we have some customers who have only 20% market share. So that represents a significant uplift in those scenarios. At the same time, we've launched our MDU products. So those customers all have an MDU problem. that's -- and they've driven it pretty aggressively on stage. We actually had an MDU company who has something like 900 apartment buildings under management, and they did their first 2, and we're ecstatic about transitioning to Calix for those scenarios.

So that's a -- there's a great example of a significant unmonetized footprint, both in our existing base because all of our customers, about 1/3 of all tenants in the United States are in MDUs. So there's a massive untapped market that we haven't even started selling into that allows our sales teams to expand significantly. We're very bullish on where MDU goes.

As you noted in the quarter, George, we added 20 new customers. So those are essentially starting customers. I don't know where we are year-to-date, but it's got to be around probably 60-ish, I guess, I don't remember. So we have -- we're winning new customers and expanding there. So those growth numbers file into, and that's what we're specifically looking at with regards to where Cory is talking about. And then we have new market expansion, which we haven't -- we have not incorporated into the 10% to 15% because it's too early.

But we have all the international markets and everything we're going to do there, and we have the Tier 1s through dedicated platform implementations, which are not in those numbers. And then the last part is we don't have the BEAD things, but we got our first BEAD orders this quarter, and that's not in the number either.

Got it. Okay. And then just on BEAD, since you guys brought it up, I mean, how much -- I guess, how incremental could that be in 2026 and 2027? I mean, do you have any sense for magnitude or opportunity or timing there? That would be great.

Yes. So Yes, I'll just give you my thoughts on the BEAD. We are more constructive on BEAD than we were 91 days ago. The turnaround from the states on their preliminary awards was faster than we had expected. As such, a certain percentage of those customers have the ability to plan for a certain amount of the jobs next year. And so while we did receive our first order during the quarter, it's still too early to determine the demand dynamics for next year other than there will be some amount versus last quarter, I thought there would be like none. But to provide a little bit more color, here's what we know. Of the 49 of the 50 states reporting, California still hasn't submitted their awards yet.

The total amount of dollars have shrunk by about 50%, right? So the original program being $42 billion, it's now going to be something like $20 billion. And when you include the matching funds, now we're talking about a $30 billion program. Fiber is still the dominant technology at 65% of locations and 85% of the dollars. Fixed wireless was 12% of locations and 8% of the dollars. Meanwhile, LEO was 21% of locations and 5% of the dollars.

So as we have said, historically, we've done really well with government programs and that we expect to do the same with BEAD.

The next question comes from Christian Schwab with Craig Hallum.

It was kind of crystal clear, the sustainability of double-digit growth that you're seeing. But as we look into next year, should we assume we -- given the outperformance this year, Cory, that we should be at the low end of gross margin expansion year-over-year of 100% to 200% or is that not correct?

You've got that right. Thank you. We haven't talked about that yet. But yes, given the overperformance on gross margin this year, I think the increase next year will be more muted. It will be at the lower end of that 100 basis points to 200 basis points.

So we'll continue to expand it? Yes, our software content grows every single day.

We're going to continue to guide the 100 basis point to 200 basis point, and we're going to grind at it, right?

That's right. But it will likely be at the lower end of that 100 basis point to 200 basis points next year.

Great. And then now that we found this new found enthusiasm for BEAD, if you think about aggregate dollars that could come to you over a multiyear time frame, when do you think that peak spending would be?

Well, we still think it will likely be more of a lens shape deployment curve. And so it will start ramp up to some level in '26, then probably level off for a few years before it tails off. Like we said with all these government programs, they take a lot longer to get started than anybody thinks. The dollars that they pull through tend to be much larger than the program and that these programs then go on for a longer period of time.

So we do think it's still a lens shape. It's still too early on the awards to understand the buying dynamics of how much they buy, how much they buy ahead, what mix of products that they're buying and how much they're actually going to actually start building next year because, again, it's very late in the cycle. But -- so that's all we can say at this point is that there will be some BEAD revenue next year. just don't have a good sizing of what that might be. [indiscernible]

Yes, we'll give you another 90 days.

Well, the other part, I will say -- so again, I had a lot of these conversations with that connections. To Corey's point around this is going to be significantly larger than anticipated, which we have the exact same narrative that we've had where other people 3 years ago were saying BEAD money is around the corner. We have been very clear. So one of the things you should take away from it is the fact that we're actually proactively talking about it now means that the money is about to flow, right? So in 2026, the money will flow. But the bigger part of that, that everyone needs -- and you kind of touched on it, right, is that the BEAD money -- so for example, when they go and they build out a BEAD network and they drive it to what was funded, while they're doing that, and in fact, we had this very specific conversation with a customer, where while they're actually building to the ones that they got funded for, they're actually, in essence, passing a whole bunch of other ones that didn't get funded, and they're going to go drop those customers in. And so all of the TAM affiliated with those is pretty significant. So for example, if they got funded to do 10,000, in many cases, there could be another 5,000 to 15,000 that they pass that they're going to build as they go. And so that's not really built into the BEAD TAM. That's the first thing.

The second thing is that once they get the networks in, that's really just to go and build the connection to the subscriber. And going back to George's questions with regards to how do we actually monetize and where are we going? The most important thing for us is to add subscribers because if we were just back to old Calix where we were just a network company, then we wouldn't be so fixated on how do you win the subscriber. But as we talk about in our model, the monetization of $1 to $10 a month in software margins, which are massive, is where we're focused because that's the long-term growth and what actually allows us to continue significant margin expansion. And so as we -- they go and they build that network, BEAD gets them there. And then we go help them actually how do I convert and get to 65% or 70% market share on the BEAD money that I put into it to actually build the network while at the same time, expanding with the customers who are not funded on BEAD, but I'm passing them anyways, I'm going to drop a fiber in.

So in those scenarios, we're talking about it now because in 2026, it's finally there. We don't talk about things -- it goes back to the other question is why it's bullish on AI now because it's right there, and we don't talk about things that we can't actually touch and see. So that's why we're talking about BEAD now, we can touch and see it.

The next question comes from Tim Savageaux with Northland Capital Markets.

Congrats on the results. Just a couple of quick ones here. On the BEAD front, I think historically, we used to use maybe 10% of the award value or network cost to kind of get a sense of what the access infrastructure opportunity might be. I wonder if that's still a good number or given your [indiscernible]

We're using more we're using more 5% to 10% depending on the -- it's more 5% to 10% is where the number is. But then what you have to also do is extrapolate the incremental opportunity when you win subscribers, which is not part of that, to your point on it being the access network.

You got to think that these locations are harder to get to. There will be put more money into. [indiscernible] More construction dollars, Tim. So 10 is too high. You're going to have to go less than that.

Yes.

Fair enough. Speaking of the $1 to $10 per month, any update on the timing for a more detailed breakout of the appliance business versus the software and platform business? I know you were end of next year was what you were discussing.

Yes. I mean that's still the stated goal by the end of '26.

Got it. So no change there. And if we look at the guidance for Q4 for sequential growth, the various quarters this year, you've had maybe small customers driving things in Q2, large and medium. Anything to call out in terms of customer segment movement into Q4 in terms of driving that sequential growth?

Yes. Broad-based demand across all segments.

Yes, nothing to call out, Tim.

Yes. But no -- well, no, I'm going to call that out. That is something to call out, right? Broad-based demand across all segments. the value in our business model and the fact that we have almost 1,200 customers is the fact that we -- and we don't have a 10% customer, correct? Right? And we have no 10% customers, which means that from an investment point of view, because of the fact that you see the segments going up and down, that's the value of our business because we have a diversified revenue stream. And so one might be up one quarter and down the next and up the next quarter, it's actually allowing us to even out the whole thing broadly. And when you have broad-based demand, which we do, that allows us also to plan the business very well.

We will take our last question from Ryan Koontz with Needham & Company.

Maybe a couple of topics, if I could. Maybe philosophically, relative to your 10% to 15% growth model and thinking about your SAM and your current customer relationships and how do you think about growth limiters in the industry that can allow you to outperform that, whether it's relative to the BEAD process, supply of fiber, supply of labor, supply of components, which doesn't seem to be that big a deal anymore. Like how do you philosophically think about risks and upside relative to things that are out of your control?

So from a risk point of view, at this point in time, we've actually been talking a lot about that as to where do we see the risks. we really don't see them with regards to limiters. So there's BEAD, the BEAD stuff, the reason why we're actually saying, hey, now we see it coming out is because of the fact that, obviously, because it's a governmental process, and as we said right from the get-go is that these things are a bit complex and takes longer to get here. But when it does get here, it starts flowing and it flows for longer.

So there's some timing components affiliated with that, right? So that's the first one. And then from a systems and components point of view, we don't see any issues other than there's some memory prices going up, but we balance that out with some other areas, right? So but these things all balance out because of the fact that we have a broad-based business. There's nothing specifically from a leadership point of view that we're looking at as a risk. I will talk to upside, but are there any other risks that you see, Cory?

No, I think as it relates to the practical limiters on the business as you go and build new networks, it's permitting and labor. So that puts a natural governor on just how fast they can go.

Well, but by the way, in Washington, that's probably one of the biggest topics that everybody is talking about, which is how do you actually speed permitting and access and those different components. So there's so much focus on driving that, right? I would also think that the midterm is coming up next November that people want to get things done as they go into an election cycle.

So that's definitely -- I would say that's going to speed the process to some extent. And so I would say that there's nothing out of the ordinary that Cory and I are talking about and the leadership team is talking about that is a significant risk. In fact, the exact opposite, we feel very confident with regards to the broad-based demand, the running of the business, the talent that we have inside the company and our relationship with customers. So we see that as everything is green.

Now with regards to upside, I'll go back to what I said around artificial intelligence. So this is where we're uniquely positioned because of the fact that we have deep insights on how to run a broadband company. We've mapped it all into our cloud. We have intelligence at the -- in the network, and we have significant intelligence on the prem. And we bring all that together in the brain, which is the cloud, which is now AI capable, and we're uniquely positioned to help our customers do more. So -- and I go back to the question I got asked about the monetization of AI, we are absolutely going to monetize AI in a significant way by helping our customers monetize their business at a significantly faster pace. And you have to understand how sales cycles work, which is my background, spent 30 years running enterprise sales organizations. And so would you rather go and pursue a customer and get, I don't know, 5% or 10% price uplift on a single cloud? Or would you actually help that customer add 25% more subscribers, which is -- takes your revenue per subscriber from 0 to significantly higher in the $1 to $10 range per month.

My view is I want the $1 to $10 as high as I can as fast as I can instead of grabbing a 5% or 10% kicker. -- makes no sense. And so -- and plus, this will also help us drive -- if you look at our clouds, we have roughly 1,000 people on our clouds or customers on the clouds. And then it drops off. We have 3 clouds, it drops off. And therefore, how do we get all of our customers to every single customer having 3 clouds. That's what AI can help us drive to, and that increases monetization.

And so much rather have 1,000 customers on 3 clouds than a large – a 1,000 of them at x clouds in that way. So we think there's huge opportunity. And frankly, I've been here 9.5 years. This is what we work towards. We tested neural networks for the first time back in 2017. We saw the opportunity, but we couldn't actually meet the privacy and security requirements that our customers needed to use artificial intelligence in a trusted way.

In the last 2 years building all of that. And now we're literally going to have our success teams go into a customer and say, "Hey, Cory, remember how I was telling you that you really should do this and you said, I don't have the team members or the capacity to do it because I'm busy doing other things”. Now with AI, we're going they say, "Hey, Cory, here's what you should do, press the button and the AI engine will do it”. And that -- when we talk about crossing the chasm, that is going to allow us to cross the chasm at a rapidly faster rate, which is why we are so bullish on AI.

That's great stuff. One more quick follow-up, if I could. Nice to see the rebound in the small customer cohort best in like 10 quarters. But you're looking upmarket, the medium and large cohorts have grown over 50% year-to-date, clearly, a great indicator there. How much of that gain upmarket is competitive displacement versus your customers investing more and being more successful with your help?

Do you mean in small or in large?

Medium and large, medium and large.

And those are competitive displacements. large percentage of that is us actually becoming the go-to-market engine for these customers, right? So where they were with someone else and that's who they were using just for what I would call dumb WiFi, and now they're looking at what we do across our clouds and everything we're doing on the Experience Edge and saying, I want to radically improve my go-to-market strategy, and Calix is the best one to do that.

So absolutely, we're displacing in the Fed because they had a previous partner, and we're now helping them drive their business. And one could argue that from a competitive displacement point of view, if we're with a service provider and they're competing with someone else who's not using our product and they crush them as they do, that's us actually displacing a competitor in a different way with helping our customers win and crush their competition who's not using us, is actually us helping us -- that competitive displacement in a different way. And that's absolutely 100% how I think about it every single day.

We have reached the end of the question-and-answer session. And now I'd like to turn the call over to Nancy Fazioli for closing remarks.

Thank you, Latanya. Calix will participate in several investor events during the fourth quarter. Information about these events, including dates and times and publicly available webcast will be posted on the Events and Presentations page of the Investor Relations section of calix.com.

Once again, thank you to everyone on this call and webcast for your interest in Calix and for joining us. This concludes our conference call. Have a good day.

Thank you. You may disconnect your lines at this time, and have a great day.