Five9, Inc. (FIVN) Earnings Call Transcript & Summary

Why don't we go ahead and get started. Thanks, everyone, for being here. For those of you that don't know me. My name is Arjun Bhatia. I am the analyst here at William Blair, who covers Five9. For a full list of disclosures, you can go to williamblair.com. And with that, I'm happy to introduce the Five9 team. Thank you guys for being here. We have Mike Burkland, CEO; Andy Dignan, President; and Bryan Lee, who is CFO. Thank you, guys, for taking the time and being here.

Maybe if we can start just at a high level, and if you can give us a little bit of a background into where Five9 fits into the CCaaS market, what part of the market do you go after if you're going to segment the CCaaS space and where Five9 sweet spot is and then we can go from there?

Yes. Sounds good, Arjun. So by way of background, we're a $1 billion company, top line is growing in the mid-teens. And a lot of our growth over the years has come from our march up market. We're going after really the largest of the large enterprise part of the market. If you look at our revenue mix, over 50% of our recurring revenue actually comes from customers that are generating $1 million or more in ARR to Five9. So -- and again, if you look back in time, I've been here 17 years with the company, either CEO or Chairman. And when we went public in '14, I think we had 3 customers over $1 million, and now it's making up more than half of our revenues stream -- our recurring revenue. And so we're playing in that part of the market. For sure, that's growing faster than the rest of our business. That is our strategic focus. We're also growing internationally. But we're also seeing, the fastest-growing part of our business is our AI. And it now makes up 9% of our revenue mix, our subscription revenue mix for enterprise, and that is growing at 32% year-over-year. So if you think about our overall business growing in the mid-teens, our AI business growing in the 30s, and it's very additive. There's a lot of narrative out there around the AI impact on contact center. We'll talk about that, I'm sure, a lot. But look, we see this in our customer base as an additive opportunity and expansion of our TAM, and it's proving out that way.

Yes. We'll definitely get to AI. I think a lot to talk about there. Before we do, maybe if we can touch on where we are in the last evolution of contact center because there's a big kind of on-prem installed base in this space, right? I think more so than other software markets, you have a lot of Cisco via on-prem. So -- and over time, those customers have been moving to cloud, and you've certainly seen the benefit of that. But where are we in that migration? Are customers still moving? And I'm sure AI has a role to play there as well.

Yes. Great question, and thank you for asking that, Arjun, because it's so important for everybody to understand that. Look, we're in a $24 billion core CCaaS market, AI being an expansion above that and additional TAM. But if you look at that core CCaaS market, it is it's growing very nicely for us. We're replacing these end-of-life -- in many cases, end-of-life on-prem solutions from these legacy providers where as an industry, as a -- globally, it's about 40% in the cloud, 60% still on-premise. So it is a massive opportunity. If you think about it, it's -- in core CCaaS, it's still a 3-horse race in terms of who we compete with 2 other cloud vendors. We're all continuing to penetrate that core market and replace those legacy solutions. And I think a really good way for everybody to think about this is, look, no matter what your philosophy is on this, AI bear thesis and AI replacing human agents, again, if you look at the Gartner survey that just came out to a bunch of large brands, they're looking to deflect and over time replace humans with AI to the extent of about 5% of the human agents, not 80% like the Klarna original story when it came out. So again, this core CCaaS market is massive, it's very attractive. And if it -- if that $24 billion goes down by 5% or 10% because of AI, great, we're still -- it's still a massive market. We're $1 billion. Our 2 competitors are a little over $1 billion, adding up to, call it, $4 billion out of that $24 billion. If that $24 million becomes $20 billion, so be it. It's still a massive, massive opportunity. And AI is additive to that. And I think investors are going to start realizing this, that this is really an exciting time because customer experience is still very, very strategic to every single brand out there. We talk to our customers all the time. I talk to our customers all the time, and they're really excited about leveraging our AI to augment and make their human agents more effective, but also in some cases, drive self-service. And they're doing that with our AI agents, and it's working really well. The ROI on the last earnings call, we talked about a few examples of some of our customers that are leveraging our AI agents. And they all -- you can see, the ARR with Five9 has grown significantly in each of those cases.

Can we -- so -- okay, let's switch gears now to AI because this is going to be the, I think, important topic, To start off, why in CCaaS, why is AI such a bigger maybe value prop than other markets? Because I hear it in other spaces. But I think in CCaaS in particular, it's much -- the potential is much greater. And is it the labor dynamics in the CCaaS space, is it the ROI? Maybe talk about that a little bit.

It is, Arjun. I think that's a great question, too, because look, AI is going to be deployed across all kinds of use cases across enterprises. But if you think about contact center and customer support, customer service, which is essentially what most contact centers do; it's a reactive motion in many cases, right? And it's a perfect fit for AI to do its job. But again, let's be really, really careful not to over-rotate on that. As I said, 5%, 10% of interactions over time, over years are likely to be handled by AI agents. And what's great about this is that look, as AI advances, it allows us to really provide those applications on top of whatever LLM or other proprietary models that our customers have access to. So it really helps us deliver those AI solutions so much faster and so much more effectively with real ROI. And we're very focused on the tangible ROI. We're very -- we talked about being customer obsessed, and that is our focus. Our focus is helping our customers deliver great customer experience, self-service when necessary and when it's a good use case. But again, the contact center is a very, very good place for AI. But it's not going to replace 80% of the agents -- human agents. So I think people are starting to realize that. And I think, look, our customers know it, our employees know it. I think the Gartners of the world know it, but I think the investor community has completely over-rotated and is -- when that herd mentality flips back...

Well, I think there were a few anecdotes in market, mostly from Klarna, I guess, that have quickly reversed back. which has been a pretty interesting reversal...

I'd also add that the bar in the contact center space is extremely high for self-service and automation, right, because of that labor. You look at mobile apps, you look at self-service, you look at the classic IVR voice bots that were out there for the last decade, the bar is really high. And I think that's why you look at players like us and some of the main CCaaS players out there. We can actually deliver on this because there's -- we have a lot of experience doing that.

Yes. And the industry has been, again, over the last several decades, has always attempted to drive more self-service. And this is just one more phase of that.

And what is your sense of -- I mean this is going to be a hard question maybe, but what is your sense of what the future of contact center looks like? Is it going to be the lowest kind of 30% of inquiries that are handled by AI and the most complicated 70% are handled by humans? Because there is some level of unique humans in the mix still and that exception handling. And we'll get to that part because that technology, I think, is pretty important, where Five9 comes into play. But talk about like what is the right way to think about it.

Yes, I think the right way to think about it, as I said earlier, is look, we've talked to all of our customers. Gartners talked to a lot of their customers. And I think the consensus is, again, somewhere between 5% and 15% over the next few years of interactions will be likely handled by AI. But look, there's no substitute for the human touch. We all know that. As good as AI is and as great as it's going to get, there's still a human touch that's necessary when it comes to customer service. And I would also say that we've got opportunities beyond customer service that we're starting to expand our platform into. We acquired a company called Acqueon, which is an outbound platform, an outbound omnichannel platform. That's also a pretty good use case for AI, but it's also a use case where humans are very involved in.

Yes. And maybe in terms of Five9's capabilities from an AI front, what is in market today in terms of the offerings that you have? Is it full -- is it agent assist, is it full -- kind of full interaction with the customer? Talk through a few...

Yes, I'd like to think of it kind of in 3 phases. If you think about the front end of an interaction with a customer, AI agents, as we call them today, but we used to call them IVAs. And this is the self-service part of interactions. So they can be voice bits or chat bots. We have both. We acquired Inference, the leading IVA provider, 4-plus years ago. It got us a huge lead in AI, and we've built on top of that. So think of it as kind of the first phase of self-service on the front end of interaction. And then you think about Agent Assist, which we also have solutions there from -- for everything from transcribing interactions to populating CRM systems with that, to helping the agent have information at their fingertips through Agent Assist. And then you think about kind of post-interaction workflow automation, process automation. We acquired a company called Whendu a few years back, and that is also powering our genic AI. So if you think about that spectrum from self-service to helping agents to be more effective in the contact center human agents, to after interaction automation; this is where Agentic AI starts to become really exciting. And our AI agents were actually -- going to be announcing this very soon, our [ Agentic CX ] platform. And this is a situation where AI agents can work with other AI agents, whether they're our AI agents or whether it's [ Agent Force ], for example. And we're partnering very, very tightly with Salesforce and up ServiceNow and other platform players to basically make sure that this is a team sport. We've always partnered well with Salesforce. We have 1,200 joint accounts. They're excited about our partnership. They want to leverage the fact that we're in the interaction stream, right? And if you think about Agent Force, it's going to mature, it's going to become an option for several of our joint customers. And look, at the end of the day, it's going to be some of their AI and some of our AI working together. But they've got to be plugged into our platform. AI is only as good as the data, the contextual data it has access to. And to get for Agent Force and other third-party AI to do its job, they've got to be plugged into our platform. We use transcript stream and voice stream to provide that through APIs, and we monetize that. So this is an exciting time for us.

And I think a big differentiator for us is if you go back 7 years ago when we got into the AI space, we made the decision to build our applications on the software side of AI and leverage the underlying engines, right? We talked about that the fact that we can use multiple engines under the hood. And so if you look at how rapidly the space is changing, you had initially, there's GNI, now you're getting into Agentic. And you're going -- to Mike's point, there's the velocity of announcements of features and functionality we can deliver because we built our AI around the fact that we knew this space would evolve. And so the velocity of innovation, I think, is going to be very high for us over our competitors.

And you're a little bit agnostic to what's underneath actually is what you're saying?

Yes.

And maybe not to jump the announcement a little bit, but it sounds like you're going to have Agentic capabilities in each of the 3 buckets that you laid out and then they will ideally work together. So maybe, Andy, you might have a good perspective here, too, because when you're going to market with customers and you're talking to them about Five9 delivering their AI capabilities, within the 3 buckets that Mike laid out, would you lead with a particular one? Where are you seeing adoption first from customers that are looking to bring AI through Five9 into the contact center?

Usually, it starts with the voice bot and chat bot or the -- our voice and digital AI agents. Most customers historically didn't -- the cost and complexity to go do a voice spot, whether it was the classic nuance part of the world, was tough for most customers to do. So leaning in immediately with market-leading voice bot and chat bot is kind of that first step. In the agent assistant or in some places, better known as copilot, that is essentially the human and voice bots and chat bots working together. Those are -- that's actually our fastest-growing product. Our #1 is our AI agent, our IVA that came from the Inference acquisition, fastly followed by Agent Assist. And so that's just a good proof point of you're in -- there are a lot of use cases for voice bots and chat bot. But ultimately, then when there's escalations needed, a very seamless handoff to the agent to then allow AI with copilot Agent Assist to be able to help that interaction is a differentiator for us.

And we're continuing to innovate with additional AI products on top of this. We have a product called AI Insights, which is really for the overall business to get insight into the interaction flows in and out of the contact center, for example, and then helping them through this automation essentially. We just basically turn it on, let it run for a couple of weeks, and then our customers have insight into exactly the mix of interactions, the sentiment of them, the capability -- the automation opportunities because it's use case-by-use case, that's basically what the clustering is. It's a really powerful, just diagnostic front-end solution as we go in and work with our customer base to figure out where to go with AI for second and third. And our AI Blueprint Program is a great initiative that helps our installed base customers, how we help them basically define that blueprint, that road map for them as to where to go for a second. And we use AI Insights, for example, to get that insight to help define that road map. So we're using our own AI technologies to actually help our customers deploy the other AIs.

Yes. Actually, that's a good kind of maybe segue into customer readiness because when I talk to customers, AI is 100% on their road map all the time, right, of -- they want to deploy at some point. But to your point, even earlier in terms of what the market experts and Gartner are saying about. It's going to happen over time. So where are we in terms of customers having their own sort of data estate in order, having change management, all the process and maybe kind of blocking and tackling that we sometimes overlook? Is it there yet? Is it we're knocking down the blockers? How would you characterize that aspect of it?

Yes. I mean I think we talked about the AI fog middle of last year, and a lot of that was -- came down to two different things. One, customers, CEOs that all said, "Hey, you need to go figure out how you deploy AI within the company." But most companies, first off, they didn't have the expertise internally, right? And obviously, they're getting hit by vendors, probably 10 vendors a day saying they can do X, Y and Z. And a lot of them, secondly, their own data wasn't in a place to really adopt AI. And so since then, most companies that we deal with now, pretty much when we go down the path of an RFP, they have a head of AI or they have an AI sort of committee within their company. And they're ready, right? And so customer readiness is there much further than it was. And then the second piece is, we took a step back and we asked -- we talked to our customers, and they essentially said, "Hey, look, we want you to focus on what you can deliver now versus the hype." And this is where we invested in hiring more AI expertise within our customer success teams, our TAMs, our professional services, our sellers. Enabling them, we believe we have the strongest CX sales force in the industry. We also think now we can have the #1 AI expertise in the industry. And so we took that and then did things like AI Blueprint, to what Mike talked about, to make it very tangible for our customers. It's not just a case study, right? This is what this use case that we're seeing in your business, how it can deliver the ROI for you. And be able to have that demonstrated ROI as well as every customer that wants to have references, right, from within verticals, and so that was really our focus. And we've seen that sort of fog lift essentially because customers are more ready. And then certainly, we're meeting them where they need to be met.

Yes. And what role do you think partners can play in this in your go-to-market broadly, but especially in kind of addressing this?

Yes. If you look at partners, I mean, we call it our balanced route-to-market strategy, right? And if you look at it, there's the -- we still have that $24 billion CCaaS opportunity. A lot of that is coming from Cisco on-prem, Avaya, Classically, there were resellers, right, in that space that's sold than that. They're still part of that mix. And so we've doubled down on resellers. Then there's a classic referral kind of partners. And then we obviously have our ISV and tech partners like our ecosystem marketplaces as well as Salesforce. And then you have the service providers like BT and AT&T. So we're doubling down across the board. At the end of the day, having a strong ecosystem is a key part of this. But you get into the SIs, we look at -- we just announced that -- in February, Deloitte was our digital AI partner of the year. They're usually in these opportunities, helping customers define their digital transformation, CX transformation and big time in that sort of the data changes. And so allowing them to have access to APIs to be able to differentiate what they do and provide those services, that's how you get them to be able to build a practice on Five9. And again, we see a lot of them leaning into that. And so they're right on the front end, making sure that they can -- that when they think about delivering the customers AI strategy, they know exactly what we bring to the table and they have the capabilities to go deliver that as well.

Okay. Very good. Maybe last topic on AI and then we can -- there's some other topics I wanted to talk about as well. But at pricing, have you figured out how to price each of your capabilities? Do you think that's an evolving sort of mechanism or -- and how does it compare to your core seat-based pricing model?

Yes. Yes, I'll start. Andy, feel free to chime in. But look, all of our AI SKUs are pretty much consumption-based pricing, but those are pre-committed consumption bundles. So you can think of them -- what -- customers do like predictability. We are -- look, we're in some early opportunities where we'll, on a case-by-case basis, let prospects turn on AI and kind of pay as they go for a period of time. But at some point, they like to know how much they're going to spend on products, software products, and we obviously like the visibility as well. So -- but they're based on consumption bundles. So sometimes it's data. Sometimes it's other throughput. Sometimes it's capacity based across the AI set of products that we have. So -- and again, I think it's right down the fairway in terms of how most AI solutions are being priced these days.

So you're like you're buying a pack of consumption, it expires at some point, and it sort of use it or lose it up until...

Yes. And if you go over that, you also pay the [ overage ].

Yes. Perfect. Okay. And then do you -- sorry, in your early customers that are adopting, do you have a sense of how -- where are customers adding it, to your point, is that it's additive to their core seats, and it's an uplift to the overall ACV?

Yes. So on the last earnings call, I talked about 3 examples of many across our customer base. And what most of them are doing is look, they're building a business case based on labor arbitrage. So they're going and getting budget based on AI, allowing them to have less humans in their contact center. But when the rubber meets the road, they actually deploy our AI agents and they typically just don't grow their agents like they were planning to grow. So instead of increasing their agents -- again, these are -- a lot of these are growth businesses. Normally, they would grow their human agent count in good times and especially in good macro environments, which we're not in. So there are essentially these three examples that I talked about. Many of the -- I think you could think about it as mostly keeping their agent count relatively flat, maybe a slight decline, but they're redeploying these humans. They're also -- even if their call volume is reduced slightly because of AI doing its job in containing self-service interactions, they're redeploying some of those humans, they're allowing those humans to have longer, more meaningful, more strategic conversations, more upsell conversations with those customers as opposed to trying to get them off the phone as fast as possible. So that's where the reality is. And again, those three examples, I think our ARR increases. With each of those, we're anywhere from 37% ARR increase to [ 5 9 ] over the last couple of years as they've gone through this journey with us to 100% growth in ARR. So it's working really, really well. And again, part of it is we're delivering ROI in these businesses. And these are some of the largest brands in the world, as you know.

Yes. Okay. Well, I'm sure we could keep talking about this forever. But maybe for the benefit of everyone here, if you -- so you mentioned at the start, Mike, that there's kind of 3 major CCaaS players in the space that are kind of going after the opportunities. Where do you kind of think about Five9 fitting in competitively? And what are your kind of core differentiators versus the other two that you would point out?

So I would just say, look, at a high level, we're the only of the 3 that is a pure cloud company, a SaaS company from day 1, right? Those are -- both of our competitors have legacy parts of their business. Sometimes they're getting -- migrating their legacy customer base into their cloud solutions. We don't have a legacy installed base where winning, quite frankly, against them in some of the largest opportunities like the $50 million ARR plus financial services company we won. And look, it's because of reliability. Our name is Five9 for a reason; scalability, it's because of our AI leadership. And I would say, very importantly, it's our people, our expertise and our customer obsession. We are very well known. And our reputation compared to our direct competitors is night and day. Our customers know we care about their success on our platform. And we are going to deploy the people necessary to help them make that happen. And we're just different in that regard. And they know that, they learn that through the sales cycle. We spent a lot of time with these megas especially through that long sales cycle, and they get to know us and they get to know what our culture is all about and how obsessed we are to help them drive that success.

Yes, these are complex. These very large enterprises are complex and the megas multiyear deployments, right, moving off of on-prem oftentimes multiple on-prem systems within the customer. And we're just well known in the industry that we have perfected the migration off of prem to cloud. And I think that's a big reason why we're winning.

And maybe, Andy, for you, we've had some sort of go-to-market adjustments and changes over the last year or so. And part of it has been balancing how do you go after big deals versus your kind of core. So maybe talk about where you are in those changes and how you are managing that balance at this point.

Yes. So if you go back going into 2024, we made some major changes initially in how we go after our installed base, some changes in our customer success. If you look at Q1, we announced that we had the largest year-over-year growth in our installed base in the last 3 years. So that was kind of step 1, make sure that we get upsell and cross-sell all of our solutions, obviously, with AI that was really important. So that -- the success of that is there. Then in the midyear last year in our new logo side of the business, we talked about it. We had our sales teams were kind of whale hunting a little bit, right? You start to win bigger and bigger deals, that happens within sales organizations. We didn't have the discipline to sort of make sure that each team was focusing in their lane, right? You had our megas and our large enterprise deals. That's our major part of the business. So we've now a much stronger discipline in terms of the bulk of our resources. Our sales teams are in that bell curve going after the dolphins, as we call them, those $1 million to $5 million deals. And we still have a team that's focused on those megas. There's still a lot of very, very large opportunities out there. We're well positioned and going after them, but much more disciplined in terms of how we do that. And so you're seeing that play out in a more predictable forecasting our business and overall, continue to then invest in things like I talked about, enabling our resources to be the best AI experts out there, being more vertical-focused, right? Historically, CCaaS was always sort of vertically focused, but AI has really pushed you to make sure that as you go in and you approach a health care company or a financial services or retail that you have all of the use cases and the experts to be able to go deliver that.

Okay. Perfect. And then maybe if we can focus on the near term a little bit here because there is certainly a lot of questions about macro and tariff impacts. So I'm curious, how -- what you're seeing in terms of pipeline and how your customers are engaging with you? And then Bryan, I don't want to leave you out. But how would you account for that in guidance? Because that is a big question, I think, right now, especially given visibility into the back...

I'll make a quick comment on tariffs, if I could. And then Bryan, I want to make sure you get some airtime. Look, we talked about it on the last earnings call, we probably over rotated a little bit in the message. We had a couple of deals internationally, 1 in Sweden, 1 in Canada that mentioned directly the tariffs, and we're having trouble with our leadership. This is them telling us during the sales cycle that we're having trouble with our leadership getting their permission to do business with a U.S.-based vendor. There was 2 deals. It wasn't a significant impact. But in our -- again, our approach is to be very transparent with everyone in terms of what's happening with our business. And it was -- this was something that was affecting a lot of companies, and we wanted to make sure that people knew that yes, we've had a couple of cases where it did affect. But Bryan, if you want to talk about...

Yes, absolutely. So from a guidance perspective, if you look at our -- the macro environment in Q1, I would characterize that as being relatively stable. And of course, the uncertainty increased in April. And even though we didn't see material changes in our business, we decided to become slightly more prudent in terms of our guidance, which is why for the annual guidance, we kept unchanged at $1.14 billion or 10% year-over-year growth. And if you kind of break that down between the new logo side of our business versus the installed base, illustratively speaking, if we assume that dollar-based retention rate remains at 107%, which is what we reported in Q1, that would generate about $52 million of incremental revenue for the rest of the year. And we need about $62 million to get to our guidance. So then you're left with that $10 million that needs to come from new logos. And we have a vast majority of that would come from our backlog of new logos that we already won. And we have great visibility into ramp schedule those. And then from the go-gets, new bookings that have to come in, it's a very small portion, and it's really only through June or so because everything else impacts 2026 revenue. So based on those components, we feel comfortable with where we are for the annual guidance.

Okay. Perfect. And Mike, maybe just to clarify on your comment, I would assume what you're maybe trying to say is it hasn't spread since those -- the concern on international doing business with U.S. vendors kind of been relate...

I haven't heard. But Andy, you are close to the sales.

As I said, we maybe over-rotated a little bit in terms -- but yes, just being transparent, we haven't seen that continue to play out kind of...

Okay. All right. Perfect. Well, we are up on time. Andy, Mike, Bryan, thank you so much for joining us. Thanks, everyone, for coming. We are upstairs for Q&A in -- hold on, I will get you to -- we are in Jenny B. So if you have questions for the team, please come upstairs. We'll see you there.

Thanks, everyone.