Good morning, everyone. Thank you for joining us here at the Morgan Stanley TMT conference. My name is Jamie Reynolds, and I'm here on behalf of Elizabeth Porter. Today I'm very pleased to have with us RingCentral CEO and founder Vlad Shmunis, President and COO Kira Makagon, and CFO Vaibhav Agarwal. Before we begin, some important disclosures. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please direct them to your Morgan Stanley sales representative. With that, let's get started. Vlad, maybe to kick things off, you know, as you frame RingCentral 3.0 and agentic voice AI, what do you want RingCentral to best be known for over the next 3- 5 years? What does winning look like?
What are 2-3 milestones over the next 12-24 months that would validate this platform shift?
Great. Firstly, thank you for having us.
Yeah.
Good to be here. It's exactly what you say. We'd like to be known as an agentic voice AI leader. We think we're in the most natural position to get there and to keep that mantle. There are a few reasons, and I'm sure we'll go into it. At the highest level, we are the gatekeeper. As consumers are contacting their business providers, they're calling or texting them, and we are the first line of defense. We're the ones picking up the phone or fielding the text message. We are rapidly turning into an AI company, not just AI first. We are already AI first, but AI.
10% of our revenues is already coming from customers that use at least one of our paid AI products, which we have a full portfolio of at this point. We expect the 10% to increase as percentage of the base over time. It already has doubled year-over-year with our last disclosure. We also think that this creates a wonderful wedge and a new motion for us to further accelerate on-prem to cloud migrations. There's still a lot of left to do there, as well as start using it with other communications systems even outside of our direct base.
Got it. Just to dig into that a little bit more and specifically on the agentic voice AI, you know, how would you frame RingCentral's durable moat, and what are the most credible threats when we think about large platform vendors maybe embedding AI as a feature, you know, or some of the fast-moving startups that we see in the space?
Yeah. Natural moat is the fact that we're running, one of the world's, largest and, certainly most reliable and highest quality, B2B voice networks. That's our moat. There are other providers, large and small, who are doing AI and doing wonderful work there. Vast majority of them are not in this gatekeeper position. They don't have... What we do is, billions of calls and tens of billions of, you know, minutes of traffic coming through their platforms. you know, we have over half a million customers. We have, you know, over 8 million end users. These are, you know, very large bases that we're dealing with. They're growing.
Again, just to reiterate, when a consumer, and most of us are also consumers, but when a consumer calls up a business, you know, there is no startup in the world that's in a position to field that call like we do. Even some of the larger companies and some of the better-known names, they may have, you know, systems of record. They may have some, you know, application silo, the vertical as or whatever. That's all wonderful. When that first call comes in, that's the first opportunity to deploy AI, and we are right there with the full up AI portfolio. By the way, $250 million in annual R&D spend to fortify and extend and grow the portfolio.
Got it. You touched on this a little bit in one of the earlier answers, but as we think about the core UCaaS business, you know, how would you characterize the remaining growth runway ahead? You know, what'll matter most from here? Is it cloud migrations, share gains, or expanding wallet share with existing customers?
Some of the above. You know, growth is growth. I can tell you we're differentiating less and less between what you would think of core UCaaS or CCaaS and AI because AI is everywhere. AI is not an add-on. It's not a nice to have. We would be rapidly on our way of getting going out of business, but for AI. With AI, we're here to stay. We are already largely an AI company. I think this realization is sort of beginning to take root that, you know, AI is truly the very best tailwind we have ever had as a company.
Got it. That's super helpful. Kira, to get you into the discussion, you recently announced a OpenAI partnership with the GPT-5.2 integration. I guess, what changes in the product roadmap over the next 12 months as a result of this, and what are the clearest customer-facing differentiation that this enables?
This partnership is really about sort of think of it putting one of the top-performing LLM models together with the best in what Vlad just described, a platform that for handling real-time voice interactions. This is, this together, this gives us our agentic voice AI platform capabilities. Which we work with OpenAI, and we are fundamentally also model agnostic. OpenAI, their 5.2 frontier model is high performance. It enables us when we handle voice conversations, augmented by AI to have low latency, have high accuracy and a cost performance module component of that. That's super important as that ultimately translates into how we perform on margin. Having said that, we are model agnostic. We work with other models as well.
What we do is we find fit-to-purpose best model for the given problem at hand to solve. There is no future for RingCentral without AI. Vlad just elaborated on that. That means that our roadmap for the next 24 months or the next 5 years, it's AI first, and AI first means that continue close relationships with LLM model providers, continue to evolve that and continue to provide more value to our customers in the form of our products.
Got it. When we think about the three As portfolio between AIR, AVA, and ACE, you know, what's the intended customer journey? You know, adoption of these solutions as point products or an integrated suite with a clear land and expand motion. I guess, what's the most common entry point you're seeing today, and what's the next best cross-sell?
Yeah, great question. AIR stands for AI Receptionist, AVA for AI Virtual Assistant, and ACE for AI Conversation Expert or conversational expert. They work together. That's number one. At the top, it's the journey of the customer is they come in, they call, the call gets handled by AIR, potentially for more involved conversation. It gets passed into a human being. This is where AVA come in to assist during the call. The analysis of these calls, including quality reviews, proactive management with management who looks at calls being handled. The feedback loop, very important, that feedback loop back into both AIR and AVA. What you have is not just parts, but the sum of the parts.
Flywheel effect, where, they add up to more than each individual part, and they're the value to our customers. We have a number of customers now today showing real proof points in real business outcomes using AIR, AVA, and ACE together, such that, they're showing better... First of all, they can handle all incoming calls, no missed calls. That was there. They can answer questions, routine, perform routine tasks. That translates ultimately into better, customer, management and ultimately revenue. Everybody cares about revenue. To monitor the quality of that revenue so they can do better, this is where ACE comes in. With that, they get a better performing business overall.
Got it. Turning to RingCX specifically, and Vlad, feel free to jump in as well. You know, where are you winning today? What product gaps are you prioritizing next? How does RingCX fit in alongside your partnership with NICE?
Yeah. At a high level. I'll let Kira answer the second part of the question. Maybe I'll take the first and the third. At a high level, for a number of years, we've been proving the case that there is a very large and, you know, fruitful opportunity with businesses who prefer to have UCaaS and CCaaS from the same provider. By the way, this day and age, it's UCaaS and CCaaS both tied with the same AI. That's very importantly. If you think about it, you take any business, if they have a dedicated contact center, well, that's fine, but they probably have more employees who are not, you know, contact center agents.
Why not have a system, which, or why not go to a portfolio which can serve both regular rank and file as well as dedicated contact center agents, which, by the way, is exactly how on-prem leaders have been selling for decades. You know, your Ciscos, your Avayas, your Mitels, they all have, you know, on-prem versions of UC and CC working together. Okay? We continue seeing that use case. Now that we have our own tech with RingCX, it's obviously a lot easier for us to control the roadmap, to be able to, you know, commit and commit and deliver on customer requests. And, you know, of course, economics is better. You know, we don't have to rev share.
You know, price point is also can be very aggressive. What it does to the NICE relationship is, it's still complementary because we're not positioning RingCX as a standalone, you know, very high-end contact center, where NICE inContact is, you know, is a leader, one of the basically two leaders, you know, at the world scale. There is a following. We have a base. The base is stable. It's growing slightly, you know. Our future is more with CX, in the mid-market and below. In high-end enterprise, it's wonderful to have that product. Feature set, Kira, you can address.
The feature set is that it handles from simple cases to more complex cases. RingCX scales very well across from small to very large customers. We now have a full portfolio. We have RingCX, we have WFM for workforce management, and we have all of the AI. There's really today, no contact center without AI. AI permit dates every aspect of our RingCX product. The future is really tied into more AI-enabled routing, more AI-enabled workforce management, more quality control, and more proactive assistance, such that really imagine the future where RingCX is really a proactive product, not a reactive product. Kind of let's leave it at that. Working together with both across the entire portfolio.
The fact that RingEX and RingCX work together is super important because in any organization today, you've got contact centers, but you've also got back-office people.
Got it. With the traction that you're seeing with RingCX, like how does that split out between, you know, your existing customers and net new customers?
It's about 50/50. That's what it's been, and that's what we're seeing. In terms of reporting kind of where we're selling RingCX. I would add to that our large deals today are what we measure, we call something $1 million TCV deals. Over half of them have RingCX attached to them, and those companies also adopt all our AI portfolio as well.
Super helpful. Shifting to the go-to-market side, you know, you successfully built a very large channel and GSP ecosystem that now represents, I think, over 10% of ARR, is growing faster than the company overall. I guess with half of these GSPs enabled to sell the expanded product portfolio, you know, how is the revenue mix between partner-led and direct business evolving? As you scale AI product adoption across the GSP base, what are the key enablers or requirements to accelerate the velocity of the attach rates of AI solutions through the carrier channel?
With the carrier channel, we're very proud of the network that we've built and the relationships that we've built. I think almost all of them at this point are taking on our AI products, and some of it we've disclosed publicly, AT&T, Telus, Vodafone previously. We're not disclosing just yet exactly what the breakout. The uptake is great, we always with the new products, and we've always first take it to our direct channels and directly control our channel partners where you know what the numbers are. It's $100 million now in ARR at the end of last quarter.
We anticipate, as we continue to mature, AI, presence of the, of our AI products within the GSP, channels, we'll start disclosing those numbers, but not just yet.
Got it. Vaibhav, you know, would you be able to speak to how the unit economics differ between, you know, direct and GSP-led deals?
Yeah. From a GSP perspective, look, it's a, it's a unique motion. As Kira mentioned, it's highly differentiated, and what it gives us is, puts like hundreds of thousands of boots on the ground for the product to be able to reach the customers. I think from a unit economics, it's a highly profitable motion. We've publicly disclosed that that motion is growing in double digits from a growth perspective and from a time to payback, which is a measure of unit economics. It's under an 18-month payback period for us. Highly profitable motion.
Got it. On AI monetization, can you walk us through the approach to pricing, whether it be minute bundles per conversation, per seat, or some sort of hybrid? How should investors think about unit economics as usage scales?
Yeah. Great question. In our AI pricing, we are very deliberate, and we are pricing products based on the value that's being delivered and the customer use cases that are being addressed. AIR, as an example that Kira talked about, you know, is at the top of the funnel. It addresses phone calls as they are coming in or interactions in terms of, you know, when you make a call, AIR picks up the phone and the agent is able to take actions such as answer basic questions, you know, book appointments, etc. It's priced on a usage-based model. The revenue directly scales with the number of minutes that the customers are using on the product. That's an example of how, you know, of usage-based pricing with AI.
Overall, Vlad talked about the early traction that we are seeing in AI. We've reached $100 million of ARR. You know, 10% of the base is now using at least 1 of the AI products. Overall, you know, AI, is additive for us. It's an addition to the core products that customers are buying and therefore is expected to be accretive. I think from a, from a customer standpoint, what we are seeing in that cohort of customers that are using 1 AI, you know, there's an uplift in both ARPUs and net retention rates. That helps kind of with margins. From a gross margin standpoint, you know, as these products scale and as we get past the initial infrastructure spend, I mean, these products are all expected to be accretive to gross margins.
I think there are three key takeaways. One, AI is additive, adoption is growing, and the attach is growing. From a customer standpoint, the wallet share will increase, and from a.
Infrastructure growth margin standpoint, you know, as these products scale, it'll be accretive to growth margins and these will all result in, these will be levers for long-term margin expansion.
Great. Kira, going back to you. Are there verticals where voice AI is consistently seeing faster adoption and maybe clearer ROI? You know, thinking about some of the go-to verticals you've spoken to, like healthcare and financial services. I guess, you know, how far are you willing to go towards building out, you know, vertical-specific capabilities versus maybe a broader, horizontal approach?
You said it. We've traditionally done really well in golden verticals: healthcare, financial services, SLED, retail. Why have we done well? Because those are the businesses that require high intensity of B2B interactions and then B2B2C that we support on our platform. They are mission-critical, they need to be reliable, and they have a lot of compliance requirements, as especially healthcare, financial services, SLED, et c. Our products are largely horizontal, but we do have go-to-market motions that are vertical specific in healthcare and financial services and SLED. We'll continue to strengthen those, and we'll continue to add capabilities to the product that make us specifically attractive to specific verticals such as financial services, healthcare, et c.
especially with AI, that enables us to do quite a bit of contextual work within the product that is vertical-specific.
Great. Vlad, just as we think about demand right now, you know, what are the signals you're watching in terms of sales cycles, pipelines, conversion rates, renewal conversations, and what are they telling you as we, you know, start out 2026?
We'll let Kira answer that maybe.
Demand is very strong for AI. I would say it that way. Like, I think like we said at the beginning, there is the future of RingCentral is a company that is an AI company. This agentic voice portfolio that we have, agentic voice AI portfolio, is all about AI-first, and every aspect of our product portfolio now has AI. We talk about 10% of our customers using at least one product, but we have a much wider adoption of AI that is actually not monetizable to this, such as our AVA product permeates across our RingEX and RingCX products, such as sample taking notes, et c. We're seeing customers now going from being cautious to being sort of leaning in and saying, "Let me...
You know, help me understand what you have, because I know you've got it. You know, we've heard you got it, or even if we haven't heard you got it, we, you know, we have these needs." The shift in the customer behavior has taken place. We position ourselves today and how we think of ourselves and certainly how we train our sales forces is, it's about transformation. We're enabling transformation. We're a lot more than enabling communication. It's enabling transformation, enabling businesses to become more efficient, to produce better ROI on investment with us even because of the capabilities that we've put into portfolio and this integrated approach to go to market.
Great. With customers leaning in maybe more aggressively, is it fair to assume that these sales cycles are maybe shorter than what you'd seen historically? Or, you know, can you speak to any noticeable improvement in win rates that you've seen as you've rolled out these new products across the portfolio?
I would say that customers behave like a horse. You know, larger customers are gonna take their time. They're gonna test it. They're gonna kick the tires. I wish there was a way for large customers to have AI train them to move faster. Maybe that's, you know, still something that we'll come up with. Of course, like our horse behavior seems to be more or less similar to what we've seen in the past. Now, remember that now they're buying more. The good thing is, they're buying a lot more of our portfolio with a similar behavior and timing patterns and similar ROI.
Just maybe one thing to add is we are, you know, relatively early stages, but still of a new net, new motion for us, which is product-led growth.
Yes.
As, more and more of our customer base is actually using our app.
Yeah.
You know, not traditional phone, you know, that gives us, you know, really good opportunity. You know, some people are asking even, "Okay, well, you know, take me through the journey. You know, okay, you got, you know, I'm a customer, UCaaS or CCaaS, how do I know about AIR, for example?" Well, how you know about AIR is that in the app. In app, it says, "You know what? Here's AIR. Maybe you don't know about it. Maybe go check it out." We're also tracking that. Look at the high level. We do have a bit of an execution machine we've built over the last couple of decades. It's functional. It's getting, you know, more and more efficient. By the way, AI is making.
You know, AI is used very heavily throughout the company. Every function, including GTM. Okay? It's helping us identify opportunities within the base, outside of the base. You know, the ways that we are triaging, which leads go to which sales groups, which can be handled automatically. We're not that far from having, you know, full AI agents, you know, do some business development for us. All of those are, A, happening and also being tracked. Kira is looking at a very, very complicated dashboard, which is why she personally is hard to replace with AI at this point.
Some super great color on how you're using AI internally. Maybe to just unpack that a little bit more, you know, what are some of the biggest changes you're pursuing from that perspective in 2026 compared to sort of what you got in motion in 2025?
We have lots and lots of things in motion, you know. Sort of directionally or holistically, if you think about it, our AI addresses what happens before a human picks up a call, while human picks up a call, if they do, after that call is completed, if it's recorded, it gets post-processed. Results of that get fed right back into the agent. You know, kind of make it real. Let's say you call in, AIR picks up, you ask a question, AIR does not know the answer, okay? It patches you into a human agent. That human agent happens to know the answer. The call gets recorded. ACE picks it up, says, "Here's something new. I can feed it right back into AIR and AVA.
Next go around, I don't need to pick up." Okay? And this is why this whole AI trend is super good for us because, you know, it lends itself to, you know, outcome-based and usage-based pricing, which is, you know, many people are asking for. We figure that we have kind of the general scaffolding together. We can do a lot more at each and every stage before, during, and after. Of course, we're doing quite a bit of work, early work, but still in verticalization. As we, you know, double down on, you know, retail or financial services or healthcare or what have you, there are opportunities for deeper integrations, more, you know, specific workflows, just custom knowledge bases.
We are in a good position because of the size of the network, is we, you know, have learnings across the network. We have our hands full. Good news is we are spending $250 million a year and, you know, more and more of it is going towards AI. That's, that's a unique asset.
Super helpful. That's probably a good segue into some of the more model-specific questions. Vaibhav, you've guided to meaningfully higher GAAP operating margins for 2026. I think calling for about 430 basis points of increase at the midpoint to around 9%. You've also outlined a longer-term target of 20% GAAP operating margins over the next three to four years. What are the biggest levers to get there in terms of stock-based comp reduction, operating leverage, gross margin improvement, and what are the biggest risks to that trajectory?
Absolutely, yeah. I think there'll be two big drivers of the GAAP operating margin improvements. The first one is just, you know, improvements in margin. You know, there's a lot of operating leverage in the model. You know, our fixed costs don't need to grow in line with revenues. We've consistently now demonstrated a sustainable margin improvement profile over the years. Revenue is consistently outpacing expense growth. That's point number one. Point number two is we are being very disciplined in terms of our spending, whether it's headcount, vendors. You know, we are increasingly using AI within the organization that's causing efficiencies. Across the board, we are very disciplined. That in addition to operating leverage is creating margin improvements. This year we've again guided to non-GAAP operating margin improvements, which I expect will continue.
The second thing is, we also look at operating margins in conjunction with SBC. Reducing SBC has been a top priority for us, and we've now gotten SBC from 20 points about 3 years ago to under 10%, and our long-term target is 3-4 points. Again, SBC remains a key retention tool, you know, for employees. We'll continue to use it as a means of incentivizing employees and aligning their interests with the shareholder interest. We are being very disciplined there as well. We are hiring in lower cost locations. We are disciplined in terms of how much grants we are giving out. We are also shifting meaningful portions of compensation from, you know, stock to cash. People are getting paid. It's the payments are, you know, being done in cash instead of stock.
If you think of the journey from here to 3-4 points, that'll be a meaningful driver of GAAP operating margins. Overall GAAP operating margins from here on will grow faster than non-GAAP operating margins.
You know, you've also guided to $590 million in free cash flow for 2026. I think that represents about 11% growth and maybe a 21% free cash flow margin. You've also driven a really impressive track record over the past few years, getting that free cash flow generation up from around $100 million to the $530 million in fiscal 2025. For 2026, can you just walk us through some of the drivers for the free cash flow guidance and some of the key underlying assumptions?
Yeah. Again, the levers for free cash flow expansion are similar in the sense that operating margin improvements are flowing through into free cash flow. One of the things that we've done is our quality of free cash flow conversion from operating margin has consistently improved. The operating leverage in the business and disciplined spending and working capital improvements is resulting in us approaching about $600 million in free cash flow. You know, Vlad talked about the $250 million of R&D spend, majority of which is going into AI. That's an example of investing the money back in the growth and innovation. We are consistently paying down debt.
We are buying back stock. This quarter, we introduced our first-ever quarterly dividend as a strategic enhancement to our capital allocation profile.
Got it. A lot of flexibility from the capital allocation perspective. I guess, just how should we think about what conditions would cause you to lean harder into one bucket, whether it's organic reinvestment, you know, maybe growing the dividend, accelerated share repurchase, or potentially some sort of, you know, tuck-in acquisition?
Yeah, look, I think the approach that we follow is a balanced one and a disciplined one. The first use of cash always is investing back into the business organically or inorganically. We have done some acquisitions in the past, like we completed CommunityWFM, which was complementary to our product portfolio, kind of made sense from a ROI standpoint. We invested money in acquiring assets that were complementary to our product roadmap. The first use always invests back in the business. Number two, strengthening the balance sheet, paying down debt. We have been on that trajectory. We've been paying down debt, and we've outlined that we'll be our desire is to be investment grade by the end of the year. That's the second use of cash.
From there, we are returning additional capital in the form of buybacks, on which we are being opportunistic. You know, it depends on market conditions and where stock prices are at, et c. We have a $500 million authorization remaining, which we expect to execute on over time. We added the dividend.
Great. We are at time. Vlad, Kira, Vaibhav, thank you so much for joining us, and a thank you to the audience as well.
Thank you.
Thank you.
Thank you.