Harmonic Inc. (HLIT) Earnings Call Transcript & Summary

Thanks, everyone. I'm Walter Jankovic, CFO of Harmonic. Welcome to Harmonic's 2024 Virtual Analyst Day. We appreciate you joining us today. With me are Nimrod Ben-Natan, Harmonic's President and CEO; Asaf Matatyaou, SVP, Broadband Products; Dan Gledhill, SVP, Broadband Fiber; and Gil Rudge, SVP, Video Products and Solutions. Before we begin, I'd like to review our agenda for today. First, we'll have Nimrod kick off our session with a brief introduction, a review of our overall strategy and where we see the opportunities ahead of us. Then we will have Asaf and Dan, do a deeper dive into our Broadband segment, where they will discuss current market trends, our exciting growth strategy and differentiated technology in this segment. Following this, I will provide our long-term target financial model for Broadband. Gil will then walk you through our Video segment, where he will discuss our strategy, which is focused on continued growth in SaaS and is designed to address current market conditions. I'll finish up the presentation by providing our long-term target financial model for Video, and then we'll open it up for your questions. Before we continue, I must mention that today's presentation will include projections and other forward-looking statements about our business and financial performance. Actual results may differ materially from our current expectations. I refer you to our most recent Forms 10-K, 10-Q and 8-K, as filed with the SEC for more information on risk factors. Please review the cautionary language here on the slide and note that unless otherwise indicated, the financial metrics we provide you on this call have been determined on a non-GAAP basis. I know many of you with us today are already very familiar with Harmonic. But for anyone new joining us, the slide presented here provides a few key points related to our company. As you can see, we operate through 2 distinct business segments, Broadband and Video. Today, we will present those separately in order to provide you with more insights into each of these segments. We will talk about what we have already accomplished and perhaps more importantly, our forward plans and expectations for these businesses. With that, I would now like to turn it over to Nimrod to kick off today's event.

Thank you, Walter, and thank you all for joining us today. Well, it's my third day on the job, but my journey with Harmonic is a bit longer. I've been leading the Harmonic Broadband business since 2012 and led the organic growth in to broadband market and pioneering the virtual CMTS and distributed access architecture. Prior to that, I've been leading the product and strategy for Harmonic products. So I'm very familiar with our Video business, technology and customers. I am excited about the growth opportunity, both in Broadband and Video streaming. And today, we will share with you more details on our technology differentiation, market opportunity and projections. I truly believe in solving challenging problems which create value to our customers. Typically, this requires disruption of technology or architecture and out-innovating. I also believe in being nimble and focused on execution. I also want to share with you today a couple of organizational updates as I step into the new role. Thanks, Walter, for agreeing to take on additional responsibility. In addition to being our Chief Financial Officer, Walter will take responsibility for our operation, IT and cybersecurity. In addition, we're conducting a search for a new general manager for our Video business. Combined, this will greatly help me to stay focused on customers, products, strategy and investors. I will start with a brief overview of the current state of broadband, highlighting the significant development and challenges in the industry. We'll start with the recent emerging Internet traffic trends, then we'll examine the competitive landscape for service providers and how these providers are evolving their networks to stay ahead in the race to meet growing traffic gains. Lastly, I will review the addressable market, which supports our growth expectations. While we no longer see the same 30% to 40% annual increase we've seen during COVID, we do see more like 15%, both in the downstream and upstream, and we know upstream is more challenging for cable operators. At the same time, we see an interesting impact of special events such as live sports that are causing significant spikes in bandwidth demand. A good example for that was the NFL playoff game, which was exclusive on Peacock late last year, which, by the way, was based on Harmonic's technology. The game had peak concurrency of more than 16 million users, which created an increase of about 30% of peak Internet usage for many of our North American customers. Another interesting example is what happens during the first day of a new game release, where all gamers are trying to download at the same time, the new game and start playing. The upcoming Call of Duty Black Ops 6, which is coming in October, is expected to be a 300 gigabyte file. And you can expect how long it's going to take to download it if you don't have a really fast Internet service. We now see intensified competition for cable, both with fixed wireless and fiber with limited cable subscribers growth or even modest loss of subscribers. We're also seeing cable getting very creative at bundling services on top of their broadband service, specifically with mobile and more recently, the way they rebundle streaming video services. Given the competitive landscape, cable operators must evolve their network to keep up with customer demand both on upstream and downstream as well as take advantage of network modernization, which will help reducing power, space and operating costs, while improving the service reliability. As we know, majority of cable networks are still running on the so-called lower split, which is limiting their upstream speed they can offer with analog optics and aging CMTS platforms. Fiber is clearly the desired end game, but very capital intensive to upgrade and takes many years to execute. The good news is that there are multiple network upgrade options for cable operators to consider with DOCSIS 3.1+ or boosted which can help them leverage the full potential of the DOCSIS 3.1 and achieving fiber speeds. Or with DOCSIS 4.0 for those who would like to achieve a more symmetric multi-gigabit service which, by the way, can enable even higher speeds than 10-gig PON fiber, and you will see this in action shortly. And of course, fiber optionality for fiber islands, fiber on demand, precision PON edging out away from the network or greenfields and of course, the BEAD program. We're also seeing interesting opportunities with edge compute, which can run at the very edge of the network, milliseconds away from subscribers by leveraging the compute infrastructure that is being deployed as part of the network modernization. Our recent announcement of the new Beacon speed maximizer is taking advantage of this capability. In summary, status quo is really not an option. Given the competition and state of the network, the best path forward is by modernizing the network with virtual broadband platform and distributed DOCSIS architecture. As we look at the addressable market, we see 2 major components: DOCSIS and fiber. In DOCSIS, we see migration away from the legacy CCAP into the more modern virtual CMTS and DAA architecture, which will grow the next couple of years as more and more customers are expanding into the 3.1+ capabilities or DOCSIS 4.0. And on the fiber, we see 2 key components: cable operators going to fiber, whether it's the edge out or fiber islands as we discuss and then we'll expand on that. And we're also seeing opportunity in the telco Tier 2 and 3. Combined, the fiber market that we're looking at is only 30% of the total addressable market of fiber. So this is really the portion of the market that we are going after. When you combine both the DOCSIS and the fiber together, we see about a $2.5 billion market we're going after by 2027. And the cloud services that you see is what represents our value-add capabilities to enrich subscribers' quality of experience and reduce operating costs. The Beacon that I mentioned earlier is part of that. After 3 years of leading the category of virtual CMTS and distributed access architecture last year in 2023, for the first time, we were recognized by Dell'Oro as the market share leader for the entire cable broadband category. While we lead the market, this is still only a milestone and we see significant growth ahead across all metrics of connected devices, number of modems and new customers. As we look at our strategy, the foundation is the installed base and momentum that we have in the market today, leading into our market share position. We expect those customers to expand, and we expect to diversify with new customers that we're going to win. And on top of that, we plan on layering on fiber capabilities by enabling commercially effective transition to fiber focusing on cloud services to enrich subscribers' quality of experience and explore edge compute applications. And on top of that, diversify into the fiber telco market, the Tier 2s and 3s, where we believe we can differentiate with our new fiber products, which Dan will expand on. Before I turn it over to Asaf, Dan and Gil to expand on our Broadband and Video business segments, I'd like to share a short video of about 2 minutes highlighting recent exciting and unique product announcement and presentations we made both in broadband and video. Let's play the video. [Presentation]

Hope you enjoyed it, and now I will turn it over to Asaf.

Thanks, Nimrod. I love those videos. They showcase how we bring the vision of the future that Nimrod just spoke about into reality, and how Harmonic continues to push the industry forward. We have a tradition that we bring first-in-industry products to trade shows. I was recently on the floor, just last month in ANGA in Cologne, Germany. We were incredibly busy with people spilling out of the booth and customers were excited about the 3 products that we just announced. Customers love being hands-on for things they've never experienced before. Let me highlight a few points in these products. With the Pearl, we showed the highest density of remote OLT that supports both XGS and GPON. Next, with the Beacon intelligent speed maximizer, folks at that booth saw 4x the speeds to subscriber. Now they can squeeze every bit out of an impaired plant. It was a real highlight of the show with many wow responses that this is game-changing for them, saving them truck rolls and labor. And lastly, with Unified 4.0, they saw firsthand the record-setting speeds that CableLabs blogged about. That's 9.4 gigs to a single subscriber. One operator even looked at the back of the cable modem asking, where is the fiber? Whether you've got a plant for DOCSIS, fiber or both, we've got you covered. We're way ahead and continue pushing forward to keep our high share with current and future customers. Now it's not debatable that broadband is essential. It's considered a utility. We expect that it's always working and then it continues to grow and change with faster speeds and newer applications. Let's review the key values to winning in broadband. We already covered the first 2 points that are table stakes, high speed and reliability. Operators also need solutions that grow and expand with subscribers' needs and they want to depend on suppliers that have the reputation and track record to deliver products and services for mission-critical broadband. Environmental impact and the recurring OpEx needs to be sustainable, it can't grow at the same rate as the speeds that they offer. Lastly, the future is dynamic and being future-ready to adapt quickly keeps them competitive. I was talking to an analyst recently, and he asked who would risk a broadband business if you don't have all these ingredients? Their broadband service is way too important to risk. Our proven track record and our rich feature set put us in that essential position. Now let's look at our products and services that are the winning combination in broadband. The concentric circles represent our cOS core solution with our platform as the foundation. The applications running on the platform are around that foundation and our central cloud services enhance the experience and increase subscriber satisfaction. Our cOS core connects to an extensive set of devices that enable Harmonic's ability to expand and diversify to other market segments, which Nimrod recently discussed. Now let's focus on our platform first. It's the foundation that we've been investing in for more than a decade. The value to our customers starts with the optionality, supporting both DOCSIS and fiber broadband simultaneously, connecting to any device and any location. So whether an operator starts with DOCSIS and grows with fiber services or wants to start deploying only fiber today, the solution adapts to any shifts in broadband infrastructure. The infrastructure real estate that cOS runs on is incredibly valuable to operators with compute being as close to the subscriber as possible, enhancing their subscriber service experience. The value to Harmonic is a compute footprint that can expand with revenue-generating opportunities for edge compute applications. The platform also feeds real-time and high-volume data to feed our cloud services advanced analytics. The cOS platform is future-ready for value-added applications such as service chaining, which can drive even more license revenue going forward. In summary, it's the foundation for expanding into new verticals that grow our TAM. One proof point is our recent charter announcement that they are deploying Harmonic's platform in a distributed access architecture, creating a flexible and sustainable foundation for market-leading connectivity services. We will now focus on our applications. It's important that we delight our customers with our growing set of applications, so they can continue coming back for more. We started our journey with the virtual CMTS with its first deployments in 2016. Now it's the market share leading CMTS in the entire industry. We expanded our broadband reach with fiber applications, including virtual OLT and virtual BNG. Leveraging our edge compute, we enriched subscriber services and increased subscriber satisfaction. The Beacon intelligence speed maximizer adapts to plant conditions, delivering the fastest speeds to each subscriber. And L4S reduces lag and packet loss. In simple terms, happier subscribers reduce subscriber churn. Another way to think of this -- another way to keep subscribers happy is having the software resiliency needed to increase service uptime. While the software we deliver is easy to install and upgrade as well as our software velocity keeps pace with the operator needs, the software itself was built with our expertise and technology we've pioneered for years. The value to the business is that it's sticky. Customers love it, and it's difficult to replace. An example of our applications in action is with Comcast, where they use our virtual BNG and remote OLT technology connected to the virtual CMTS, delivering multi-gig symmetrical regardless of the type of wire connecting their customers. Now we turn our focus to enhancing the operator experience and subscriber satisfaction with our subscription-based cloud services. The real-time and high-volume data that we spoke of earlier feed into network and service insights, driving proactive resolution and intelligence AI-based recommendations. The deployment in operational automation reduces manual intervention to deploy rapidly, reducing human error with simplified maintenance. The same data-driving insights also drives our monitoring service, which is a single pane of glass for all applications, combining DOCSIS and fiber services for unified and simplified operations. Our customers enjoy increased subscriber satisfaction, reducing churn while reducing OpEx. Subscription and cloud services is one of the segments that create revenue growth and we are best positioned to leverage our platform and applications. Feedback from our customers include KBRO's CTO, who said, "The platform provides real-time analytics and insights for proactive network intelligence to ensure an outstanding broadband experience for their subscribers." Finally, let's look at the devices that connect the platform, the applications and cloud services to the actual subscribers. The value to our business is the expansion and diversification of the market segments across an expansive set of broadband footprints. Optionality is a theme for us, and our devices fit in any location, indoor outdoor as well as delivering fiber and DOCSIS services. Each has amazing performance and the lowest power consumption, whether it's Unified 4.0 devices or our nodes or our shelves and modules that support any [ type of ] PON technology. We are also proud of our broad patent portfolio. With our devices, we want to highlight the over 40 patents that deliver reduced power consumption as well as notifying and reducing network power interruptions with vBias, Buoy and Last Gasp technologies. The value to our business is that our device portfolio is unique and market share leading, in what would otherwise be a commodity market. An example of the versatility of our devices is with Claro Perú, where they combine their past VCMTS deployments with Reef and Ripple together with the Fin, to deliver both DOCSIS and fiber services. Let's summarize why we win with our transformative software that brings the future today. cOS is market share leading and has the only mature and field-proven solution that delivers symmetric multi-gigabit broadband in an environmentally sustainable way that is future ready. Coupled with our broad portfolio of devices, our business opportunities will continue to expand and diversify. We'd also like to shine a spotlight on the importance of and positive environmental impact enabled by our cOS solution. Vodafone and Comcast have separately published energy savings driven by virtualized cloud-based technologies and network evolution that paved the way forward. For example, Comcast is reporting that it has reduced electricity to deliver each byte of data across its entire network by 40% since 2019. If we look at Harmonic's platform when compared with legacy products, we save our customers an enormous amount of space. That means they don't need to build new buildings or in fact, they reduce their facility needs while increasing their speeds. It also translates into significant recurring electricity savings. This is another example of how we win and how we push the industry forward by leveraging modern and sustainable technologies. This brings us to our concluding slides that list the numerous ways that cOS is a growth engine into the future, delivering the modern broadband infrastructure that has been held back for over 2 decades by legacy equipment that served our industry well, yet it's holding it back at this transformative moment. Let us focus on a few highlights to distinguish between legacy and modern solutions. As we've shown today, the Harmonic cOS solution checks all the boxes that we listed earlier as the key values to winning in broadband. The future is today with cOS and it only gets brighter as we look ahead. And with that, I'd like to hand it over to Dan, who is going to focus on our broadband fiber business.

Thanks, Asaf. I'm lucky enough to spend most of my time talking to Harmonic's customers and prospect accounts about how we can solve their most challenging problems. I'd like to share with you today their feedback and why they select and trust Harmonic with their broadband solution. We'll frame the conversation by covering the markets we serve, talk about the unique value that we provide to our customers and walk through the most common use case for our solution and technology. By the end, I hope to explain and illustrate how Harmonic is fundamentally changing the way operators can design, construct, deploy and operate their broadband ecosystems to fundamentally change the underlying economics of fiber-to-the-home. Talking about the markets we serve. We start with cable operators. Today, Harmonic has over 185,000 RPDs deployed for DOCSIS. Each of these locations can coexist with a remote OLT, the infrastructure component that powers fiber-to-the-home. In fact, each RPD can represent an opportunity for colocation of many remote OLTs. Looking at the telco side of the market, we see an opportunity bringing a distributed access architecture to agile, Tier 2 and Tier 3 telcos who look to expand their network and can leverage and quickly deploy our unique innovative solution. Collectively, these represent a fantastic near- and long-term revenue opportunity for us. Near-term revenue will be driven by cable companies who are modernizing their networks, and taking each of those 185,000 RPDs to fiber with the addition of our remote OLTs. Telcos will ramp much like distributed access did for cable when we initially introduced the cOS solution. Now let's talk about why operators select Harmonic. The story starts with Open ONU. In fiber, the ONU is the device that is located in the customer's home. It's frequently known as an ONU, optical network unit, or ONT. These have traditionally been required components of OLT vendors' ecosystems. They were vendor-locked. Harmonic changes that paradigm. We allow operators to select best-in-class third-party ONUs and ONTs that work with our network infrastructure. Breaking the restrictions has disrupted traditional OLT vendors and allowed a new negotiation leverage for operators seeking to manage their costs. We are reducing and changing the way traditional OLT vendors have overearned in this segment. Our cOS solution also simplifies and streamlines the way an operator views next-generation fiber technologies. There are a number of standards for fiber-to-the-home. Utilizing the cOS umbrella, we simplify that by providing a single pane of glass for provisioning, management and operation of fiber-to-the-home regardless of what standard may be utilized as the underlying technology. For operators who have existing networks of multiple technologies, it simplifies their day-to-day business. And for operators who are unsure or uncertain of the next-generation technology, they can have confidence that whatever the future holds, they'll get access to those devices, utilizing their existing cOS core. Lastly, we talked about providing a solution that's flexible for any network topology. Think about distributed access and what we've done for the cable space. We are allowing operators to deploy shelves, nodes and low-density OLT pluggables from a single virtualized core with streamlined operations for all of the devices under one pane of glass. Let's talk about how an operator can utilize these technologies in practice. The solution that we're about to walk through goes by many names in the industry: fiber optionality, precision PON, fiber on demand, fiber island, or more creatively, one operator has named this the unicorn node. It's great to see operators taking our solution and adopting it as their own. The story starts with our virtual core serving both DOCSIS and PON. We've talked about the way that it provides an operator the flexibility and leverage to manage their existing infrastructure, manage their existing buildings, switches, racks as efficiently as possible. Those buildings and infrastructures are connected to an outside plant, reflected in this diagram by the ripple node provided by Harmonic. That ripple node for DOCSIS providers serves residential and business customers generally within 2 kilometers of the node. This is the nature of DOCSIS. The node must be relatively close to the end subscriber. Thanks to the ripple node supporting the addition of 10-gig PON through module additions like our recently announced Pearl, we can take that existing node location and expand the service area that the operator can reach. Now with Pearl, operators can deliver fiber-to-the-home broadband at a radius of up to 60 kilometers from each ripple node. Think about what that means in terms of building a network efficiently, monetizing and utilizing existing locations to extend your reach and serve previously hard-to-get-to areas. Note that we've not disrupted the DOCSIS service. The DOCSIS service is sustained and still operating from this ripple node, the OLT module has been located in the second slot so that operators can do both simultaneously. Operators very frequently look to extend their network even further. The ripple node provides a unique capability to daisy chain nodes, to get to hard-to-reach rural places without constructing a cabinet or point of presence on the ground. Thanks to our unique technology, we're able to not only serve the 60-kilometer radius but actually go even further. Think about what this means in the context of Harmonic having 185,000 RPDs deployed today. We think about a map with our existing node locations serving DOCSIS, augmenting each of those existing RPDs with fiber gives us a coverage for new and existing customers, unparalleled to anyone else in the industry. This nets a unique capability for our operators and partners to refine and rethink the way they operate their broadband networks. We are very successful and grateful for our capability of expanding into global markets with large partners like Claro and Millicom, who have utilized the technology I've described to efficiently and affordably deploy fiber-to-the-home in new locations. Again, combining all of the capabilities of our technology, they're rethinking the way they design, manage and deploy their fiber-to-the-home solutions. And with that, I would like to hand it off to Walter to review our financials.

Thank you, Dan. Now I'd like to present our 2026 target financial model for Broadband. Based on what Nimrod, Asaf and Dan have just outlined, including expected market growth, our strong leadership position, current customer wins and our expectations of continued commercial success, we are targeting Broadband revenue of $800 million in 2026, that reflects a 27% 3-year compound annual growth rate and is well supported by both a tops-down and bottoms-up view of the market. As presented today, we expect this growth will be driven by accelerating DOCSIS network migrations and fiber expansion with both cable operators and telcos. Based on expanding to a broader base of customers and a higher mix of cOS, our virtualized platform, we're targeting non-GAAP broadband gross margin of 49% for 2026 which would be an increase of 220 basis points from our 2023 actuals. We're being cautious in projecting higher than 49% gross margins considering the level of new network migrations that may result in a higher mix of devices, nodes early on. Based on this anticipated revenue growth and margin expansion, coupled with significant operating leverage on our OpEx, we expect to achieve $224 million in adjusted EBITDA or 28% adjusted EBITDA margin in Broadband for 2026. The 3-year compound annual growth rate on the adjusted EBITDA is expected to be 47%, reflecting the strong operating leverage in our model. Now I'd like to turn it over to Gil to present our Video business.

Thanks a lot, Walter. Let's dive into a discussion on the Video business, and there's some good news here as we look forward. The question we'll be answering is, what is Harmonic's strategy to elevate the Video business, looking at profitability together with growth opportunities. We have 2 engines driving that. The first engine is our appliance business. We've updated our strategy to focus on most profitable products in our most profitable geographies, leading to increased EBITDA. The second engine is our cloud business that we expect will yield $100 million of recurring revenue by 2026. This business has 2 high-value growth opportunities, one with live sports streaming and the second with ad tech for streaming. And with the rightsizing we've performed, this is expected to provide sustained profitability, and that profitability is starting already in the second half of this year. To set the stage, let's first take a look at the market dynamics and Harmonic's positioning. At 40,000 foot view, you can see we have 2 main segments in the video market. On the left side, you can see the broadcast market. They include the ABCs, Fox and NBCs of the world. Together with the pay TV operators like DIRECTV, DISH, Comcast and Charter. On the right-hand side, we have the streaming market with subscription VOD like Netflix, Hulu and Prime Video. TV everywhere is the streaming of live linear TV like Warner Bros. Discovery, and again, NBC and Fox. And then we have the pure live streaming, including live sports streaming like Peacock and Apple. Each market has unique characteristics. In broadcast, we have a unified signal that reaches all users. Every subscriber sees the same video at the same time. And most of these channels are 24 x 7. And there are also strict standards with zero tolerance to on-air issues, if you will, broadcast standards. The streaming market is different. Streams are unique per viewer aligned for personalization and targeted ads. Viewing is on demand. The signals are only streamed when the user clicks to consume them, and this market is constantly growing and evolving. When we look at the infrastructure for broadcast is primarily fixed, hardware and software running in private facilities with a typical refresh cycle of 5 to 7 years. In streaming on the other hand, the infrastructure is highly dependent on the number of viewers accessing the service. So the infrastructure needs to be dynamic. And that draws on the private and public cloud to take advantage of all the scalability and elasticity. Interestingly, technologies and boundaries between the 2 are starting to blur. Broadcasters are moving to take advantage of the cloud, they're opting for hybrid workflows with some equipment on-premise, some in the cloud. And we're even seeing some primary workflows like channel origination moving 100% into the cloud. And the reverse is also true. We're seeing streamers moving some part of their workflows on-prem to save costs but also looking to adopt the more strict broadcast standards in their streaming platforms. You cannot have a live event or sports or news and have it fail. The implications are just too severe, especially if it's an exclusive event. And this is where we get excited. This is where we have a unique position. Actually, we have years of expertise on the broadcast side, and we're bringing that to the streaming world with the best video tech in the market. So what is the total addressable market that we can address with our portfolio? We see the TAM in 2023 to be around $2 billion. The appliance business is mainly CapEx driven by the number of channels sold. This segment has recurring SLA to support the products till the next refresh cycle of their installed base. The market is mature with much more established competitors and [ the abilities of dullard ] equipment of software and hardware out there that need ongoing support and services, which is also a nice part of our revenue. And this segment is all about efficiency and cost reduction. They need to do more with less. Switching to the TAM on the cloud side. As you can see, already in 2023, it's larger than the appliance TAM. And this is where the growth opportunity exists. This business is OpEx driven. It's based on usage as opposed to fixed cost, and there's a massive growth in platforms and direct-to-consumer offerings. And the viewers are embracing the streaming option more and more. And the more viewers, the more usage and that yields more traffic, more ad inserted and of course, more money generated. Switching to 2026, we can see the trend continuing with the cloud growing, but the overall TAM is remaining at that $2 billion mark. This is mainly driven by broadcast migration to the cloud and growing of the live streaming segment. The fact that Harmonic can address both sides of the pie, including hybrid workflows is the reason for our excitement and is key for the profitability and growth that we'll cover today. Now let's take a look at our first engine, the appliance business. Here, we focused our investment to yield increased profitability. As you've seen in the previous slide, the TAM for this segment is declining. And naturally, our projected revenue is declining, too. However, the trick really here is profitability. Take our position in the market, our portfolio and track record and combine that with the rightsizing of the business that's underway this year, and that yields profitability even with the decline in the revenue. So how do we get there? And what is different from previous years? What we've changed. First of all, we're rightsizing the business as we discussed. Secondly, we're focusing our engineering dollars on our core and most profitable product lines, together with focusing sales on the more profitable geographies. This is all backed by a strong brand and track record with the XOS spectrum product clients that are leading the industry. With this, we expect to maintain our current market share while increasing profitability, and this is a key takeaway for our appliance business. The updated strategy will continue the investment in our core products that are used by the most renowned names in the media industry, as you can see on this slide. For XOS customers, we'll continue the function collapse of the most advanced media processor in the market. This allows for more cost savings for our customers by doing much more than one unit. And we're also expanding the TAM we can reach by adding these new features. For Spectrum, the industry's gold standard of [ playout ] and channel origination, we anticipate a refresh in 2025 and 2026. This refresh was driven by end of support of thousands of Spectrum channels that we have deployed in the field today. Now let's take a look at our second engine, our cloud business, and this is where the growth opportunity is and where we expect to drive $100 million of recurring revenue by 2026. VOS and our cloud business have grown at 59% from 2020 to 2023, and this is where we see the most potential to continue to grow our revenue. Our growing cloud business falls into 3 main buckets, as you can see here. Most of the business today is in live sports streaming and linear and VOD streaming. And as you can see from the logos, we're trusted across the globe with the industry's strongest players from pure streamers to renowned sports providers, telcos, service providers and the largest broadcasters and pay-TV operators in the world. They all use and trust our technology for their business. Now that we understand what's behind the existing business, let's take a look at our growth opportunities that get us so excited. The first is the continued growth of live sports streaming. Fans are really shifting more and more to streaming and sports are becoming the most valuable asset in television. The viewers get better experience, streaming can easily reach every corner of the globe, and it's not restricted at all. And that experience includes much more innovation from [ joining ] progress to highlight, multi-views, 4K, HDR and more. And this is why the viewership is increasing and with that an increase in advertisement and ability to better address the fan base. In addition, we've seen the recent announcements that more and more sport streaming events are moving exclusively to streaming, and this is where the opportunity is. So why is Harmonic and VOS solution best positioned to capitalize on this massive growth? First and foremost, exceptional quality. Take an example these tweets, and the second tweet from a sports reporter stating, this has been the best I've ever seen for soccer streaming ever. And this is all powered by VOS360 streaming platform. With the cost of sports rights and the amount of money tied to each live event, there's zero tolerance for issues. This is paramount to the content owners. You cannot miss Messi's goal in the World Cup or miss a touchdown in the NFL playoffs, or the last shot in game 3 of the NBA finals. This is where we shine, bringing our broadcast standards and expertise to the streaming world. We have a seasoned 24x7 global DevOps team that practically monitors these feeds and the tech running them. And this solution has proven at scale, and at the largest scale in live sports streaming in North America ever. In 2021, we hit a milestone peak of 6.3 concurrent viewers with the Super Bowl, which overlaps with the Olympics. Fast forward to '22, we reached a high of 10 million concurrent viewers, when we streamed the FIFA Soccer World Cup across the globe. In January this year, we reached an unbelievable milestone of over 16 million concurrent viewers in the larger streaming event in North America, as Nimrod mentioned in his opening comments. This was all on our VOS platform and all with 0 issues and broadcast quality. From a technology perspective, VOS is fully integrated with the leading ad providers and CMS vendors, and we have an array of features dedicated for live sports, including higher video quality for a specific event, watermarking to protect from piracy and, of course, server-side ad insertion, allowing to monetize these premium events. And we're ready for the future. At the latest NAB show, we demonstrated partnership for advanced workflows using AI. Producing a Tier 1 sports event with a time line and highlights requires quite a lot of manual labor. With AI, we can enable the same experience for lower tier sports with no manpower required at all. In addition, we continue our investment in advanced video codecs to allow for better quality for less bandwidth. We're excited that our first implementation of VVC will go live in the upcoming Olympics in Paris. Where we've around 40% bit rate savings versus HEVC. Now let's switch to the second major growth opportunity, and that's with ad tech for streaming. The portion of dollar spend in targeted ads is just increasing as you can see in this chart, and streaming allows for targeted ads with monetization based on the actual viewership. This is critical for sports and live events that we just discussed. And we're adding new ad formats to expand this even further and yield a higher price per ad, and this is where we have some great innovation. In a sporting event, we all know that during a timeout or the end of the quarter, we'll be hit by commercials. So we get up, we grab another bite to eat or refresh our drink. So this is why we invested in our new in-stream advertisement technology and this tech creates ad insertion opportunities while staying in the live action. This captures the viewers without leaving the game and yielding a far higher CPM per ad inserted. As you can see in this clip, we're in the middle of the live action of a basketball game. And while there's a lull in the action, the player is walking to the free throw line, we insert a double [ buck ] as you can see here. Effectively, we've just created a new ad opportunity that didn't exist before. And because this is streaming and performed server side, it can be targeted and monetized per the actual viewer. Allowing multiple different versions, yielding yet again a higher CPM. And all of this is driven by VOS platform with the same reliability and strict broadcast standards that we discussed before. And the solution can handle the scale of even the most demanding sporting events as it's built for scale and reliability from the ground up. Looking at all that innovation and the growth opportunities, they lay the path for us reaching the $100 million of recurring revenue by 2026. The bridge that you can see here on the left takes our 2023 revenue and expands it in 3 main categories: Well, first, we have some natural expansion from our existing customers, but then add on top of that the targeted ad insertion with the new stream advertisement that we just discussed. The second is new Tier 1 customers that are moving from the traditional on-prem appliance-based workflow to the cloud or hybrid models of on-prem and cloud. And the third is the live sports streaming that we have modestly forecasted growing by 15% of the existing -- 50% of the existing run rate. So there could be upside in this segment as more sports streaming rights are purchased. With that, I'd like to thank you all and handing back to Walter to walk us through the financials of the business.

Thank you, Gil. Now I'd like to present our 2026 target financial model for Video. Based on what Gil just discussed and our expectations for both appliance and SaaS within Video, we're anticipating $215 million in revenue for Video in 2026. That reflects a negative 1% 3-year compound annual growth rate based on a negative 12% 3-year compound annual growth rate for appliances, which is in line with the expected TAM reductions that Gil mentioned. This is offset with a 25% 3-year compounded annual growth rate for the SaaS business. During Gil's presentation, he mentioned the potential refresh cycle in appliances during 2025 and 2026. Although that poses some potential upside opportunity for Video, the exact timing on refresh cycles is difficult to predict, and for purposes of our target model, we are choosing to be conservative based on what we believe to be a more sustainable revenue base in appliances. We're targeting non-GAAP Video gross margin of 64%, an expansion of 310 basis points from our 2023 actuals based on our expectations for SaaS revenue growth, as this becomes a larger portion of our business. As mentioned earlier in the presentation, we've embarked on a restructuring program, which is expected to significantly improve our Video profitability. This program is expected to deliver $28 million in cost savings in FY '25 and beyond. Based on our restructuring actions and the growth drivers discussed earlier, we're expecting to achieve $30 million in adjusted EBITDA or 14% adjusted EBITDA margin in Video for 2026. As mentioned in our presentation, our key focus in Video is driving sustained long-term profitability for this business. This completes our formal presentation today, and we would like to now open the lines for questions.

[Operator Instructions] Our first question comes from Ryan Koontz with Needham.

With regards to the fiber opportunity, how do you compare that on a revenue per subscriber basis compared to DOCSIS for Harmonic?

Ryan. So I'll start and I'll let Dan to expand on that. Largely speaking, it's about the same. Perhaps the number of widgets or ports you may need will be different. But as you know, the split ratio is different. You can put a node for cable that will serve 500 subscribers and the OLT for fiber may split to 128, 64 or 32, so you may need higher density and higher count of ports. So at the high level, we would say it's largely the same. Dan, you want to add anything on top?

Yes. I think the one data point I would add for you, Ryan, is that the capacity of the remote OLT node is several times higher in terms of subscriber count. So a ripple might support in excess of 2,000 subscribers in fiber-to-the-home mode. So we see a volume opportunity to serve more subscribers from each location.

Got it. That's super helpful. And on your open ONU strategy, I certainly understand that. What kind of costs are you incurring to kind of qualify all these partners? Certainly, the Tier 1s are going to kind of dictate to you, I am sure who they want to use for the rightsize opportunity. But apart from that, where it's not a volume-driven business, are you kind of carrying those costs to qualify the ONU suppliers?

Yes. So I'll start by -- first of all, the fiber market is not as open and interoperable as the DOCSIS cable market where you have CableLabs. So identifying this along with the kind of bookend and the proprietary that some of our competitors are serving. We decided to embark on that. And indeed, it's an investment that we needed to make not just in equipment, but also automating that and being able to repeat that with each and every release, we get out the door and we get from the ONU. Dan, do you want to add more on how we manage that internally and with customers?

Yes. I think you hit on it, Nimrod, it's through automated test processes and the structure around it. There are some industry standards to also facilitate specifically CableLabs and BBF.247, are helping. But it's predominantly driven by our ability to automate and leverage our experience integrating a wide range of third-party devices from our DOCSIS history that's making it sustainable on the fiber-to-the-home side.

Got it. And so I assume those partners are primarily driven by your customer feedback on who to work with there?

Yes. We certainly prioritize customer requests. We also try to prioritize that by what we see in the market in terms of new technology, type of silicon and so forth.

Got it. And maybe one last question on the Broadband side. Great to hear about your kind of strategic direction towards edge cloud as well as kind of cloud-hosted subscription services. Do you have any other models out there? And I know you guys have this kind of intelligent bandwidth products that just came out. But any other sorts of examples of that you guys have in mind for kind of layer on applications for cOS?

So a couple of things, and I'll let Asaf to expand on that from the platform point of view. We are exploring other directions. I can give you one other example. We touched on the complexity of scaling live events such as the Super Bowl. And this is actually putting a lot of stress on CDN networks. So one idea you can think of is that you cash and replicate live sports event at the very edge of the network, which would be the cOS platform. There could be other applications such as I touched on the downloading of new games released. So this is like one file that has to get to a lot of users in a very short amount of time. So these are ideas. There could be other ideas related to AI. We are open-minded about that, looking at opportunities. And Asaf, maybe you want to add from the platform point of view, kind of what else?

Yes. Thanks, Nimrod. So we look at it from an opportunity point of view, looking at big data versus fast data and things that have the value, the subscriber bringing it close [ as the ] underlying foundation that enables us to bring other applications. We call it application bundles. And the example of Beacon, for example, running on our platform is simply just another application. So when we look at these other opportunities, we know that the underlying platform gives it high performance and high availability. And that way, we can evaluate and install and take advantage of those opportunities on our existing installed base. And I think as we look at the customer experience and subscriber enrichment, we will discover new opportunities that bring that fast data closest to the customer, to the subscriber.

Got it. Yes. It sounds like a real interesting opportunity as it evolves. I assume that in addition, that cOS has sorts of APIs that your largest customers could even develop their own apps for that? Is that accurate?

Yes. Absolutely. As Asaf mentioned, we have the capability we call application bundles where they can bring their own application and load it on the platform.

Okay. Great. And just one last one. With regards to the ad insertion, I know you guys have had some announcements in the last few months about this kind of in-stream advertising. Is that something that you're uniquely able to do because you're the streaming platform? Or do other ad insertion -- or are you competing there with other ad insertion technologies that can essentially do the same thing, but there's APIs that other ad platforms could plug in the sports streaming?

Gil, do you want to take that?

Yes, that's a great question, Ryan. So there are some folks that are doing the main differentiation. So I think I'll answer it 2 ways. First of all, we can do it if we're the video pipeline, but we can also do it if the customer has an existing video pipeline, and we can layer that on top. The real differentiation that we have here with this technology is really the scale and reliability because when you have a live sports event, it scales at a massive rate, the beginning of event and you have to scale all the infrastructure. And then [ have a joint name ] and add it seamlessly with the reliability that we're talking about in the presentation. That really is where what we provide shines and is slightly different. So again, it can work with whoever has Harmonic as a video pipeline, but it can also work if the customer has a different pipeline in place.

Got it. So it sounds like we should think of this as another product or a module for the streaming business? And the next kind of follow-up on that is, can this also be applicable to the appliance business, to the linear business? Is there an ability to do so to speak, an in-stream for appliances? Or is appliances always broadcast only?

So appliances are primarily broadcast only, if you will. But there are opportunities actually in the -- with some of the cable providers and that's some of the opportunities that we're seeing now where they have their existing QAM streams, and they have their actual [ node PG ] streams within that platform. And they've started -- they need to require ad parity between the legacy QAM and the OPT. And we're deploying some of those solutions today with the same technology. Once that's deployed, we can now -- are adding targeted ads in the [ ad infrastructure ] and this new in-stream ad technology. So basically, there are multiple opportunities except for the standard streaming applications. So it is something we're really, really excited about because of multiple opportunities to exploit that technology.

Got it. Great. And maybe, Walter, one last one for you. On the gross margins on cable, sounds like you're expecting a stronger mix of hardware there in terms of the gross margin targets. What are the kind of puts and takes there that we should think about gross margins going forward? Do you -- is that -- do you feel like that's a baseline or a conservative estimate or you're expecting stronger share in hardware generally? That would be helpful.

Certainly, Ryan. I think with regards to our gross margin projections in the target model, they are based on taking a more cautious approach. If you think about what's occurred, there's the whole wave of migration that's taking place, as Nimrod mentioned earlier in the presentation in terms of customers, and so as we see 2026, which is less than a couple of years away, we expect to continue to see early deployment migrations with new customers. And so that will warrant a higher mix of devices. And so based on that, we are being responsible in terms of our view of what that mix will look like. We're still in the early stages of deployments. When we look at those waves of deployments, when we look at the buildup of not only the largest customers, but the wave of customers coming behind that, we definitely want to make sure that we're capturing footprint and that we're looking at the appropriate mix in terms of what that will look like. Is that the long-term view of the business? No. The expectation is as we continue to get through those early deployments that the mix of the business will be more favorable to the margins.

Our next question comes from Simon Leopold of Raymond James.

A couple of things I wanted to explore here. One of which is what are you thinking about the possibility that the industry consolidates or integrates FDX and ESD technology into a unified device? I'm hearing that, that's being discussed. And wondering if -- first of all, what do you think of that possibility? And two, if it happens, does that lead to a push of your opportunities out of this year and pushes them down the line into '25 and '26. Anything you can offer on sort of this scenario or possibility? And then I've got a couple of quick follow-ups.

So a couple of things. We think that any technology that will unify the 2 flavors of DOCSIS 4.0 is very positive news for the industry. Cable has all along been very unified like the DOCSIS generations supported by CableLabs, I think that helped a lot to basically be the leading broadband platform. And admittedly, ESD, FDX created some confusion or some even delays of decisions. So I think that as you mentioned, the unified silicon end to end all the way to the CPE, from the network to the CPE is going to simplify that decision because at that point, you don't really have to decide how to deploy it. It's more about, okay, this technology is available. I'm going to put it into my network, it's backward compatible to DOCSIS 3.1. Even if I have a preference for one versus the other, I can use that, and then I can change that over time. So I think that's great news for the industry, kind of removes the confusion. We clearly have, we believe, an advantage because it means that the technology will have to support FDX and ESD, and we've got both technologies. And in terms of pushout, we don't see that pushing out anything. What we discussed is our DOCSIS 4.0 deployment is progressing and is not dependent on unified. It will transition to unified once available, but there is no dependency on unified to push that forward. For the other DOCSIS 4.0, in fact, we believe that unified will kind of accelerate decision-making and simplify the decisions for the relevant operators.

Great. That's very helpful. The other thing is, I wonder if you can offer some thoughts on the fact that one of your competitors is buying the assets of another now bankrupt competitor. I can't imagine if that changes things much for you, but just if you've got any thoughts on how that might affect the industry as a whole.

Yes. Walter, do you want to start kind of your perspective on that?

Well, I think we'll -- we've seen here is consolidation as a result of what's happening at the high level, which is the migration to VCMTS and DAA, and I think that's the big driving factor in terms of some of the other competitors having to take certain actions. And I think we're in a position where we actually see this as a positive. It's a proof point that we're moving in that direction. I think from a customer standpoint, maybe Nimrod would want to elaborate on this. This is where we see customers needing to make decisions in terms of the direction they go and if anything, having to make those decisions quicker.

Yes. So maybe to add on, Walter. The fact that they made that consolidation move basically -- and how we assume that they will consolidate the virtual core product means that this is a priority for them. This is where they see the future of this broadband access. I think it also eliminates one competitor because they will -- from 2 to 1. So this is one other thing to consider. But we think this is a good indication for the industry transition to this new modern architecture, where we feel we have a unique advantage in terms of our footprint and experience in running these kind of new architectures and this virtual implementation. Asaf, you want to add anything on kind of the change itself and the virtual architecture?

Yes. I think that it also shows that it's not easy to do it. I think it's -- investing in the platform and then virtualization on top of that is not something that is trivial and it takes time to get it right. So I think the investments we've made will continue [ to bring ] forward our unique position, and we'll see how it plays out with that investment in the industry.

Great. And then the follow-up is hopefully pretty easy. In the past, you've broken out the expectations for your fiber to the x revenue within broadband. Just wondering how you're thinking that contributes to that $800 million 2026 target?

Yes, maybe I'll kick it off, Simon. Thanks for the question. In regards to the breakouts, we had previously broken out the fiber from the total number. And what -- when Dan presented, I think it was a very good way to look at what's happening in the fiber space, especially for us. We've got a lot of customers that are hybrid customers. So they're purchasing both DOCSIS as well as fiber. And if you remember the slide that Dan presented with the ripple, you could have that basically facilitating both of the [indiscernible] technology. So it's not something that we can allocate with precision. With that said, when we look back at what we said a couple of years ago in terms of where the fiber business is expected to be, we're still on that path in terms of fiber being that size in 2026. And so when you refer back to the slide that Nimrod presented earlier, he broke out that SAM, the addressable market that we're going after between cable operators as well as Tier 2 and Tier 3 initially. And so think about that in terms of the overlap in our solution in those situations. That's why we haven't broken it out with precision, but we're on track. And maybe the team can talk a little bit more about some of the proof points we've had along the way here on the fiber side.

Yes, I'll just say that we're on that path. And the -- if you go and manually calculate the contribution of fiber, I would say, at the end of '23, it was in the tens of millions of dollars. So it's focusing towards that target. We talked about kind of the 2 key markets in cable and in telco, the telco Tier 2 and 3 is still early on. This is an area of investment in cable, it's an area that we already see success. But even in pure telco customers, we already have several of these customers running on our platform. So this is certainly progressing. Dan, anything to add?

I think you and Walter have covered it well.

Okay. Thanks.

[Operator Instructions] Our next question comes from Steven Frankel with Rosenblatt Securities.

I'd like to take another cut at that 2026 Broadband forecast and maybe give us some insight into how much revenue would you attribute to customers that you already have today versus what's in the pipeline?

Sure. I'll start with that one, Steve. Thanks for the question. So with regards to the customer mix that we have in 2026. When we compare to today, 2024, and we look at the top 2 customers as compared to everybody else, that's a key metric that we're focused on. We see a material shift down in terms of the percent of those top 2 from '24 to '26, highlighting the faster growth in the rest of the customer set as we might -- as their networks migrate and impact that revenue forecast.

But any detail on kind of how much of this is -- a lot of those customers that have kind of been in the waiting room in the last couple of years that are going to be ramping up versus discounting what's in the pipeline to get to that number? That's why I'm trying to understand your visibility on that number.

Yes. So when we built up our model for '26 and the team spent time looking at it from, as I mentioned in the earlier part of the presentation, both the tops-down as well as a bottoms-up. Looking at all the customers that we've won and are just in the early stages of deployment as well as customers that we're targeting in terms of wins. So as you know, we already have 113 customers aligned in the Broadband segment. And so a big part of the growth comes from those customers migrating their networks. In addition, there's elements of additional win and expansion that the team has built in as part of the view in pulling together a bottoms-up picture of the customer mix to achieve that number in 2026.

Okay. And then maybe take one more cut at it. So when we get to 2026 and you look across your customer base, kind of where do we think we are in deployment? Are we 1/3 of the way there? Are we 3/4 of the way deployed by '26? And what inning will we be in 2026?

I think it will vary by customers, those that started early on maybe in a more advanced stage of the rollout. But at that point in time, we may see them starting to shift to more and more fiber depending on which strategy they will take. So we certainly see a growth beyond '26 whether it's fiber, whether it's completing the migration. These projects are multiyear by definition. You could see that kind of over the last couple of years with key customers that have shared the progress that they made. I think recent update from Comcast that they will be 50% of their network by end of '24 was an indication, and they were fairly early in the [ program ].

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

I wanted to follow up on that discussion about the diversification of the business, maybe along a couple of lines. But first, Walter, you mentioned that percentage from the top 2 going down [ and try and pin ] you down a little bit on that, would you see that below 50% of the Broadband business in the '26 time frame? Or can you give us any other metrics?

Well, I think the way to look at it, Tim, is that look at the subscribers across the whole globe as we've kind of highlighted previously, and then you think about those 2 customers, they've got a certain big share of the market from a subscriber standpoint. So that's one metric to focus on. I think we've kind of said it before, they make up about 1/3 of the market. As we look forward, and I don't want to get pinned down on any specific percentage, but I think in terms of the thought process, will they be more than that 1/3? Absolutely. There are significant customers we expect to hold, the share we hold with those customers today. But to my earlier point, we definitely -- when we've modeled this out with the team and looked at our model for '26, we've seen a reduction from where we are this year, and it's material in terms of significance, in terms of dropping down.

Okay. And you mentioned share, and that was kind of where I was going to head next. I think we've got '26 targets and '27 TAMs, but even working with that a little bit, it looks like your current DOCSIS share is something on the order of 70%. It seems like you're making an assumption that, that declines a bit over the next few years and kind of correct me if I'm wrong there in terms of what's implied market share wise in your '26 target. And maybe let's just focus on that in DOCSIS. And then I'll follow up on fiber.

Sure, Tim. So first of all, in terms of looking at market, I think the one thing I'll emphasize is there's the addressable market, the TAM, the tops-down view. But as I mentioned just a minute ago, we also did a bottoms-up view. And so our key assumptions around our '26 numbers based on the top 2 customers, holding share. When we look at the rest of the market, we are being more conservative and having a slightly less share as we look forward, and that's how we've built up our numbers to be on the cautious side in terms of the roll-up of the 2026 number. So that's the comments I'll make around how we use TAM, but we also leverage a bottoms-up view from the team in terms of migrations, expectations, et cetera. So when you look at share percentages, keep that in mind. But I think you've correctly identified that we do have some conservatism in customer base. And what I'm referring here is beyond the top 2 customers as we think through the next few years.

Great. Okay. And another element of that could be just continued growth in the TAM. I think you were talking about -- I realize this is a while ago, but a $900 million TAM for '25, and I think we're now looking at $1.25 billion or so for '27. So there's some pretty nice growth there. Similarly, on the fiber side, I think similar dynamic, which previous share assumptions were around 20% of the TAM for cable PON, but you're adding kind of the rural telco market into that equation. And so I assume that share assumption kind of ought to be less than that original target as well given the expanded TAM and the rural. Is that a fair way to look at it?

I think it's a fair way to look at it. If you look at where the traction is today in the business, obviously heavily with the cable operators, these hybrid concepts that both Nimrod and Dan alluded to earlier. And in addition to that, as we've mentioned before, we've built up our investment in our go-to-market team focused around those -- initially around those Tier 2 and 3 operators just in terms of speed of being able to win deals in that domain. And our solution set is valid for Tier 1s in that fiber telco space as well. And that's why post '27, we continue to see our addressable market growing because we'll start looking at the Tier 1s as well. So I don't know if Dan or Nimrod would like to elaborate on that further because I think it's a very important point in terms of how we look at the market, and how we see the growth going forward.

Yes. The -- I guess, our initial priority for the telco market is for the Tier 2 and 3 where we believe we have a unique differentiation. It's also a market that we believe can yield faster results relative to the Tier 1 market that takes much longer to win. In terms of the relevance of our technology, it will be as relevant for the Tier 1s. We know how to win Tier 1s, but we know that it takes a bigger investment and longer time.

I would to add to that, just when you think about some of our larger wins and some of the customers we serve, they have hybrid operations where they are both cable and telco operators. So we actually have proof points serving some of the larger operators with fiber technologies as well. So to Walter's point, as our customer acquisition strategy evolves, we'll be able to expand our serviceable market in the future.

Great. And last question for me. And here I'm looking for customer count metrics, and that is around -- can you give us any color on the number of current Tier 2, Tier 3 telco fiber wins? And/or can you give us a sense of the size of the active pipeline from a customer perspective?

So on fiber, Tom, we talked and we shared the metric of total customers, 113, fiber-specific opportunities out of that is a couple of dozens where many are hard cable or hybrid operation. And several, we don't share the exact count, are pure fiber. As we said, this is a relatively new area of investment, and we expect to grow that. Our key internal metric is to match or get as close as we can to our 113, which obviously keeps growing. At this point, we are couple of dozens into that 113.

Okay. And real last question here. Can you describe -- obviously, you're looking a little further out, but can you describe the current state of your Tier 1 engagement on the fiber side? And when would you like to see some sort of field or lab trial or activity? What are your expectations there?

Let's see. We did talk about the fact that Comcast is deploying our fiber solution. They talked about that as well. They use our devices and our virtual BNG, that's...

Outside of cable.

Outside of cable for Tier 1, we don't have any direct focus like the AT&T Verizon is not an area that we're focusing on right now.

[Operator Instructions] Our next question comes from George Notter with Jefferies.

I guess I'm kind of sitting back here and listening to this whole dialogue and I have to admit, it feels like your business historically hasn't been super predictable and listening to how you guys are parsing the different aspects of the business and where the growth is coming from. It sort of implies a level of precision that I'm not sure is really here. And as evidence, I would point to some of the sessions you guys have done in the past and some of the numbers that you guys have outlined and they didn't really pan out. And so I guess I just want to kind of level set here and think a little bit about why are you guys setting these expectations? Why is it important for investors? Like, why does it make sense to really kind of feel like there is a lot of precision and full process when maybe there hasn't been in the past? Sorry for the tough question, but I think it's worth asking.

George, no, I like the question in that it gives us an opportunity to tell you a little bit more exactly about the process we've embarked on as we went through this as a team to build up our view. Certainly, things evolve and certainly, the environment changes on an ongoing basis. I'll just point to the current market conditions at a macro level, definitely impact and shift timing of when people are going to spend on those large projects. In terms of the amount of diligence around looking at the markets, not just from a -- as I mentioned earlier, a tops-down perspective, but a bottoms-up and really evaluating the customers, the customers we have today, the migration plans and looking at all of the key initiatives that we've talked about today give us an ability to kind of bring those 2 elements together, both from a top-down and a bottoms-up perspective to evaluate what are we setting as our target model for '26 based on the migrations that have to occur. And you heard from Nimrod and the team earlier in terms of status quo is not an option. So the migrations and the reason for the migrations that need to occur. And we're starting to see a lot more activity and that gives us confidence in terms of the trajectory of not only the largest customers, but as we go through the entire customer set. So maybe I'll ask Nimrod to perhaps provide a little more color on that because I think that's really a key driver in terms of giving us a level of confidence in building up this target model for Broadband.

Yes. Walter, I think going back in time, when we projected the business, we were at the time when this technology was used by very few early adopter customers. And you can understand with a small sample of customers that revenue can be -- or size of the market can be lumpy with a very small set of customers. I think fast forward to today, we crossed the chasm. There is no single customer, single operator that is not positive on the architecture. If I go back in time, even our competitors were kind of challenging the viability of the architecture, it's not going to work, it's not going to scale. I don't think anyone is doubting that today. So I think that's one key point. I think the other maybe 2 other important factors. We work with customers who, until a while ago, had a very fixed plan to go to fiber. But guess what? Cost of capital and the complexity and time that it takes to go full fiber overbuilt themselves, made them more realistic and kind of look into, hey, what can we do to maintain our subscriber base. And if we're not going to invest in the network, we're going to lose them while we kind of wait to go full fiber. And I guess the last -- well, 2 other factors. I think the competitive environment for broadband have changed 2 years ago or even 18 months ago, fixed wireless was not a competitor. Now we have both fiber and fixed wireless that are putting a lot of pressure on cable. And I think the third point is availability of technology. If back then, it was either going to DOCSIS 4.0 or fiber, I think in the past that we talked about earlier is introducing a simpler, faster upgrade opportunity much cheaper, which really gives you fiber speeds, maybe not on the upstream, but 1 gigabit on the upstream, 1.5 gigabit and up to 8 gigabit on the downstream. So I think when you take all of that together, this is contributing to our update today and the way we look at the next couple of years.

And I'm not showing any further questions at this time. I'd like to turn the call back over to Nimrod for any closing remarks.

Yes. Well, thank you all for joining us today. I hope this session provided you with a better understanding of our vision, how we look at the market, the strategy, the technology differentiation and our unique market position. We certainly have a lot of work ahead, new markets that we're getting into, expanding the business. There is consolidation in the market, kind of changes in the competitive environment. But I think the foundation that we built, the market position that we have today is a great foundation for the future. And we're excited and committed to keep executing on our growth plan. So I want to thank you again, and have a great day.

Ladies and gentlemen, this does conclude today's presentation. You may now disconnect, and have a wonderful day.