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

Good afternoon, everybody. Thanks for joining us. So for this session, I have Five9, the President, Dan Burkland and CFO, Barry Zwarenstein. So gents, thanks for joining us. How are you doing today?

Doing just fine. Thank you very much.

Wonderful. Thanks, Samad.

All right. Great. Appreciate it. So happy to have you here with us. I'm sorry, Barry, I know there's safe harbor. I forgot. That's why I'm not a CFO.

So hello, everybody. Today, we're going to make forward-looking statements about events and trends that could affect the industry, the company and its operations. Obviously, the actual results may be materially different. I refer you to our SEC filings, 10-K and 10-Q under the caption Risk Factors for things that could cause such a variance. Thank you.

Great. So thanks, Barry, for that, I appreciate it. I thought it would be helpful, we're just coming off of earnings season. The company had a really good 1Q and it was better than both what your expectations were, I think better than investor expectations as well. Could you maybe start off with what drove that solid 1Q? And maybe what were some of the underlying factors that I think largely align with your overall confidence as well?

Absolutely. So yes, we were very happy with the top and the bottom line and the cash flow from Q1. In terms of the revenue, when you think about Five9, the way it compartmentalized into 2 major buckets, roughly equal in terms of the drivers for annual revenue growth. The one bucket is the new logos, these are the new bookings that go live and then generate revenue. And the other one is in installed base as that expands. In Q1, we had -- we outperformed on the revenue side, beating the consensus by about 5%. And the big driver was, in fact, on the new logo side, where we had record Q1, record actually all time seat turn-ups beyond our expectations. And on the other side, the installed base side, it was pretty much in line with our expectations on puts and takes.

Great. And then I think you had mentioned that seasonality has bounced around a little bit in 4Q. And then, again, on 1Q. Can you just did you see from a seasonality perspective versus a normal 1Q? And how should we think about maybe seasonality going forward?

Absolutely. So if you go back in history, very typically, we have H1-H2 split 47-53, 48-52, that type of thing. And the third quarter was stronger than the second quarter and the fourth quarter than the third quarter. And this past year was no exception. However, we do track 17 different verticals. And there were some changes, most notably in our first and third biggest verticals, namely health care and consumer and retail, which, along with our second biggest financial services, account for about half of our total revenue. In the health care and consumer, those in 2021 grew sequentially by 24%. This past year, earnings grew by about 12% for a host of different factors, including just mainly the macro. That, in Q1, abated and didn't go up as much and it didn't go down as much. And in fact, it was moderately better than we might have thought. But on the other hand, in Q1, on our financial services, which is you've all heard about, we did have some headwinds. It was weaker than we thought. Normally it grows sequentially in the -- excuse me, year-over-year in the high single digits, it only grew basically flat. These are companies that are making consumer loans. They're making mortgage loans, debt consolidation, that type of thing. So taking away from all of this, the growth is still growing overall. The other 14 grew pretty much in line with what we normally have, but just a slower growth as the activity into the contact center was not as strong as it was before, and people can -- don't have as much need for the seats. One parting comment on this, if you don't mind, Samad, and that is our logo retention has been excellent. And when, inevitably, the economy comes back, we will see the benefit of the other side of that same coin that we're now going through on the installed base side.

I think that gross retention point on the logos is really important. This is a very sticky product. So Barry, one of the things you mentioned was new logos being better in 1Q. Dan, you are the man that makes that happen. Now there's a lot of people underneath you, but you drive that car. And so why don't you talk us through what you saw there? What were the things that contributed to that strong new logo performance?

Yes. Thanks, Samad. Q1 is always a bit more challenging than Q4, as an example. So sequentially, it tends to be more challenging because Q4, like most organizations we're pulling forward as much as we can and trying to book as much as we can to finish the year strong. So Q1s tend to be somewhat lighter. We were fortunate this Q1 to see a couple of things. One is we mentioned a series of large deals as we normally do, that were significantly higher from a bookings perspective. We also saw an increase in the pipeline that was for the high end of the market, what we call, strategic accounts and large enterprise, it was double from what it was the year prior. And that's being driven by 2 major trends. One is the companies that are on legacy, on-prem systems, platforms that have been there for many, many years, still 80% of the market, it's estimated, are still on those legacy platforms. Most of them have been end of life. They're not getting any more development into them, and customers recognize they've got to get off of those and onto a platform that's in the cloud. That's one driver. And the second driver is the AI and automation that I'm sure we're going to talk about today is driving that interest to get to the cloud more quickly, so that they can start to take advantage of some of these new innovative AI and automation solutions. So those are the 2 main drivers. Enterprises, especially at the high end, realize it may take them a couple of years to go through a decision process and then migrate on to the cloud before they can even get their hands on the AI stuff. So we're seeing that and feeling an uplift in the overall pipeline as a result.

So they wouldn't let me keep working here if I didn't ask about AI. So I'm glad you actually -- you brought me to it. So let's talk about that, right? It's a topic you don't go a day without hearing about it in the broader zeitgeist, Dan. And obviously, it's being discussed. And you guys have been working on AI for years. So why don't we talk about what Five9 is doing there and how you're thinking about the impact, the benefit of AI or the impact to the overall business?

Yes. So -- and this is really helpful. Thanks, Samad, because there's a what I'll call a false narrative around, oh, AI is going to replace the need for human agents in the contact center and take over customer experience. And while we're seeing AI play a role, it's mostly there to help agents be more effective and more efficient at their jobs. So we're applying -- and we've been at this for several years, and we've been reaching to really helping move your technology towards AI, so that you can get more efficiency. And we even made 2 strategic acquisitions in this realm. And we are beneficiaries as the market shifts and starts to implement. The more AI and automation we can bring to customer experience and enter the contact center, the more Five9 benefits. And I'll explain it and try to put this simply. If you think about when -- think about us as we have a capacity for concurrent sessions or ports on a system. And let's just, for simplicity, call it, we have 100 ports or capacity to handle 100 concurrent inbound interactions from my company's consumers, whatever company I happen to be. Consumers inquiring to a business aren't changing. In fact, one could argue as we give them more options, they're actually going up as we deliver chat, e-mail, voice, chat bots self-service IVAs, all these technologies that help. We're going to take those calls in. If you use the traditional basic voice that's handled by a human agent, we queue the calls and feed the agents, those are done at $200 per month, per port or per concurrent session. As customers start to take advantage of the AI and automation and IVA, for instance, intelligent virtual agent, okay, can now listen to the call, transcribe it, look for an answer, come back and speak that answer to the customer and the customer doesn't have to interact with a human agent at all. For that automated agent or virtual agent, as we call it, the cost that they pass is $400 per month, per concurrent session. So if I took the 100 access parts and I found some very basic straightforward use cases that can be handled by an IVA, my revenue would go up because I'd have 90 at the $200, and I'd have the remaining 10 ports at $400. That's just the base level. Then when you look at AI and automation, in general, people say, well, how is it being applied to contact center? And we would argue that it's being applied throughout the journey of that customer interaction. From the very beginning where we answer the call and identify the caller and drive their intent to, okay, whether I've gone through the IVA example that I just gave you and self-serve, without an agent or whether I opt for speaking to an agent. If I speak to an agent, we're also going to listen in on the conversation, transcribe it in real time, understand what's being asked through NLP, fetch the answer and feed it back to the agent, and we call that Agent Assist. I'm assisting that agent to be more effective and more responsive and much more efficient with their time. And that is also an area where we can help agents be more efficient. The third example and then I'll stop there is one more of our modules, which is called Summaries. And this is a technology where we're using ChatGPT to take the transcript of the call, go out and use GPT to summarize it, bring that back to the agent, let the agent, in a matter of 3 or 4 seconds, scan the summary, submit it to the CRM for a permanent record. The reason that's important is in many contact centers after the call is over, the agent spends a minute or 2 wrapping up their call with notes and putting in that information before they take their next call. We can reduce that from a minute or 2 down to a few seconds. We've just allowed them to do more with less. So we're seeing a lot of efficiency savers and those types of efficiency savers have been introduced for the last couple of decades, since I've been doing this. And we're constantly trying to help customers do more with less. Most of them say it will help them grow their agent count slower than they would have otherwise. But nobody, very rarely do we have a customer that says, "I'm going to reduce the number of agents because of the automation". It helps, and it will make a dent. But if you think about the different automation or self-service alternatives that we give to customers, many brands already push the easy stuff out to their website. I'll go to our website. You need to change your password. We all use the website and have one e-mail to us. Or you need to make a hotel reservation or an airline reservation, you're going to do that online already. So the contact centers have already pushed out a lot of these transactions to self-service vehicles. This is just 1 more vehicle. So that's where we see things.

That's very helpful. And Barry, as a follow-up, we get a lot of questions on what -- how should we think about the margins of this, right, especially as it grows in the mix? How does it compare to maybe your traditional subscription margins? And how should we think about that going forward as it scales?

Yes. So you should think of it favorably. We're still on a little bit of a voyager discovery on some of these, but not on the IVAs. On the IVAs, when we bought the company, we had margins in the 80s. Our current margins in Q1 was 60 for the corporate-wide for everything higher on the subscription. And we're experiencing similar type margins on the IVAs now. The target for the other offerings within the portfolio is to get to something in excess of 70 and probably in excess of 80. And let me just make one further comment, is that we are not investing ourselves in developing, obviously, things like LLMs. We are taking advantage of the commoditization that you all know about in terms of the prices of these Large Language Models coming down. GPT-4 is 1/10 of the cost of GPT-3. So the analogy that people use is we selling the cake and just because the price of flower goes down, it doesn't mean to say we're going to cut the price of the cake as we sell it.

I see that you're thinking about cake already for later. So now we know where your head's at. Maybe, Dan, I want to stay on this for one last question. There's -- this is almost like a tectonic shift, right, where you've been talking about AI for years, suddenly the plates rub together and there's this earthquake of commentary around it. But I think one of the other areas where it may ultimately benefit you is taking advantage of AI requires change. So how do you think that, that benefits Five9 versus maybe some of the incumbents that are in the space?

As far as change to what?

To take advantage of AI, you need to be in the cloud, right? So how does that end up benefiting you guys?

Yes. That's the first prerequisite to really take advantage of this technology as you've got to get to the cloud. And as I mentioned earlier, 80% of the customers are still with these on-prem legacy systems that need to be upgraded. And to think for a moment, while there's chat bots, they can't just take an on-prem legacy platform and upgrade it to a chatbot. What do you do with the other 90% of the transactions that need to be handled with a full platform? If you step back and think about it for a moment, what we deliver is an interaction platform that allows customers to come to it in a variety of ways, whether they're coming in with an inquiry through chat, through e-mail, through a voice call, whatever it is, we then can determine, again, identify who they are and then determine what to do with that inquirer. And the engines that we use can be swapped in and out. But the platform itself has to have the scale, the reliability, the security, protecting the customers' data and the list goes on and on and on about what the capabilities are built into that. That's the true moat. So while the engines might be commoditizing and leapfrogging one another and we continue to swap those out as we see fit to stay with the best of breed on those, the platform itself is what is really difficult to replicate. We've had very few entrants into our market. I've been here at Five9 for 13 years, and it's the same players we've mentioned from the day we went public in 2014 are still the same players that we compete with today. There's been mention of the Microsoft and the Zooms and Twilios of entering this space even as far back as 2017 for Twilio Flex. They're not here yet with the full platform. You've got companies that have come in and stated that they're going to enter this space. There is one that has built a cloud platform. Genesys has been successful. They took in Interactive Intelligence acquisition that had been developed for 3 or 4 years, spent another 3 or 4 years. And here they are, and they're there and they're a viable competitor. But that gives you an idea of how long it takes to build out the full stack and the full platform solution. And the advantage of the platform is such that if I build integrations as an example, in a large health care conglomerate that most of you know that we've signed, they have 40-some-odd integrations with back-office data sources that they use. And it's so important to be on a single platform because we build those integrations once, and they apply to all the different transaction types so we can help them deliver a consistent experience. Regardless of whether they came into the chatbot or whether they came into a voice call or whether they came to an agent, the information that they're going to have read back to them all comes from the same data sources. Imagine if you put a stand-alone solution on the side, you'd have to build all those same 40 integrations with a little chatbot. Nobody is going to do that. It's cost-prohibitive. So there are true advantages to being on a platform solution.

Great. That's helpful. Barry, why don't we switch gears and move off of this topic. I want to maybe dig into large enterprise activity in general. You guys obviously had a good 1Q. Can you just remind the audience how you're thinking about what's embedded in terms of large customers ramping as the year progresses into your guidance? And just I know it's too early to think about '24, but just how you're thinking about in general, that progression in the current environment?

Yes. So we're talking here about -- to go back to my very first comments about the new logos ramping up during the course of the year. There are 3 buckets over here. And I'm going to go from the smallest to the biggest. The smallest is the parcel delivery service, which initially launched in terms of recurring revenue in the fourth quarter of '21. That is going to be done by the end of this year and be back-end loaded. Then bigger than that is the health care company that Dan just referred to that started gently in the fourth quarter of 2022, will continue growing through the course of 2023, back-end loaded again and then be done in the first part of 2024. With respect to the biggest bucket, which is those orders that Dan and the team entered in the second half of 2022 and for the first part of this year, which will go live in the second part, those are going extremely well as well. So that's the visibility we have on the new logo side for 2023. You're right with respect to 2024. I'm not yet ready to talk about it. But I will say that our model of $2.4 billion in 2027, assuming the macro resumes positively next year, remains very much our goal.

Great. And then, Dan, when we're thinking about large customers, in general, even the ones that you've already booked, but just there's a lot of scuttlebutt about companies having to make decisions now, modernize. There's this urgency. What are you seeing from a competitive landscape? And what are the 1 or 2 things that customers are really talking about that are differentiating Five9?

Yes, thanks. Yes. And there is that interest to move to the cloud. And we love our opportunity, especially as I mentioned, upmarket, the pipeline doubling our validation that we can serve the largest and most complex organizations in the world. It's interesting because customers do have a hard time distinguishing and differentiating purely the products. We all have slight nuances and differences in how we go about bringing the technologies together. But if you look at our chief competitors, NICE, Genesys and Five9, the 3 of us, what we find is what's very critical and important is the team of experts that we apply to put that product in place to allow the customers to extract the most value from it, that cannot be underestimated. We pride ourselves on carrying, by far, the best NPS scores in the industry, almost unbelievable. We hired a Head of Professional Services, who didn't want to believe the numbers. And now he's running our Professional Services team globally and understands why. It costs us a little more upfront, but we put in the expertise and the consultation and the ongoing optimization of the solution. And that last point is very key. A lot of companies will supply a solution, implement it, and then they kind of turn their back and say, hey, here's who you call for break fix. And we don't. We apply a team of resources that live with that customer on an ongoing basis. They have a sales account manager. They also have a technical account manager, an engineer that can go in and inspect, on a quarterly basis, to make the tweaks and adjustments to the system. That's very, very important key differentiator.

Great. We only have a few minutes left so, Barry, I want to come back to you and think about -- we've talked about growth mostly, but you guys have also had a really nice couple of quarters on the margin side. Can you just help us understand what you're doing? Is that just revenue upside that's driving the leverage? Is that also cost control? Just help us understand the recent leverage that we've seen kind of really start to pick up in the model.

Yes, it has been a few things. I'm going to focus on EBITDA margin, making occasional references to growth. On the EBITDA margin is we've been able to moderate 2 major investments that we have been making in the past 2 or 3 years. The one -- they're both very much internationally related, driving our professional services. So we're in all these different countries in the world where the possible delivery companies, et cetera, need them. Countries that we can't even go into, most of the people in this audience sometimes and cloud operations. And both of those now are abating. We've also increased our capacity in the data centers, so that we can accommodate the type of growth that we've been seeing. And now it comes down to further international expansion and product innovation. In terms of operating expenses, we're pretty much where we want to be, in terms of our long-term model, which is for 23% adjusted EBITDA margins in 2027. By that, I mean we, currently, this past quarter at 44%. The midpoint of the model is 47%. So we actually have some leeway. In terms of gross margins, there's 3 or 4 drivers there, but the biggest one is what you alluded to, which is with revenue growing, we have a slow, steady, classical improvement in margins that we leverage the fixed and save fixed cost. We also are going to have the benefit of the higher ARPU that is going to come primarily from these AI and automation solutions. A number of other ones, but I'll spare you the details.

I think maybe just the last one, and this is I think is a question for both of you. Just thinking about capital allocation and within that M&A, you haven't really done a deal in a while. I know it's been more, call it, like tuck-ins. So I'm curious how you think about M&A and then maybe capital allocation more generally, Barry.

Yes. From an M&A perspective, we're always looking at immediate adjacencies and seeing how to extend the platform and portfolio of solutions. We made those acquisitions. We saw inference as the leading IVA company and scooped it up for that very reason because we saw this trend that was -- that is now upon us. We also acquired Whendu, which is a workflow automation solution, which allows us to do a lot of closing the loop and outreach and SMS messaging to round out that piece of the portfolio. And we're constantly looking at other areas where we can do, nothing we can obviously talk about. But yes, we're keeping our eye on lots of things.

Yes. I do not have a whole lot to add to that. We generate consistently every quarter, LTM operating cash flow and free cash flow. Records for both of those as last quarter. We expect that to continue, growing our cash balance. And we'll make a small technology tuck-in from time-to-time, but that's about it.

Great. Well, we'll leave it there for time. Barry, Dan, thank you so much. And Barry, I'll see you in London at the NASDAQ conference.

We look forward to it, Samad. Thank you very much.

Thanks, Samad.