Pegasystems Inc. (PEGA) Earnings Call Transcript & Summary

All right. Thanks, everybody, for being here today for day 1 of KeyBanc's Emerging Technology Summit. I'm Steve Enders on the research team. Here for the session, we have Ken Stillwell from Pegasystems. Ken, thank you for being here.

Absolutely. Thanks, Steve. Good to be here.

Good. Maybe just to start off now, I want to get a bit of sense of the demand environment. 2 years into the pandemic. Things are starting to go back to normal. I guess looking back, how would you kind of characterize the impacts of the pandemic on the business and demand environment? And as we start to see kind of reemerging, how are you kind of viewing demand going forward from here?

So I think specifically around the biggest driver for us is really the digital transformation team. So I think a couple of things have helped to keep that, I would say, steady and stronger through the pandemic, which is if you think about the 2 primary use cases, it's remote user access, people not being together, people -- either customers coming to a digital storefront or in the operations area, people maybe not actually being in the same location, maybe working remote; and automation, maybe not having enough people, I mean -- so I think most of our clients are struggling just staffing up processing centers, et cetera. So I think those 2 things naturally have been obviously strong through the pandemic. And as we get out of the pandemic, I don't think either one of them are changing, right? I think the consumers are -- although we're all happy to get back to live interactions, the reality is a lot of our transactions will be in a virtual or kind of a simulated channel. And you don't have large financial institutions, for example, saying let's go build an 1,800-person processing center in Manila. Like it's a different world. So I definitely think that they're looking for the technology to help enable that.

Got it. I mean do you view it as kind of a structural change that we're seeing in the demand environment from -- driven by COVID?

So I think it was going to happen anyway. I think COVID probably pulled forward the change by a few years. I don't think it pulled forward like temporarily the spend. I just think this is going to probably happen 5, 10 years from now anyway without COVID. I think now it's almost kind of -- we can't even imagine a world where we move back to doing some of the things that we did before.

Okay. That's good to hear. I mean you're sitting at the crossroads of a few different areas around technology. You mentioned one of them with automation. You're in CRM, low-code development. How do you kind of view your place in the broader enterprise software landscape? And how has this evolved over the years?

So we were always -- we always prided ourselves on having what we call a no-code environment, which is you could execute certain application design environments or applications without writing code, having to drop in to write Java to be able to execute it. So that said, we never really thought of -- we never thought of the word low code because in our mind, we wanted no code. We didn't want to have to actually write anything. So when the low-code space emerged, it's really kind of -- or at least people started to talk about it as a space, we associate most to the low-code area. But we don't go to market with low code. Like some companies actually do go to market with low code as they have a platform. And you -- we go to market around the use cases. We have vertical use cases. We think BPM or digital process automation, digital transformation, and we think CRM because clients buy us for those use cases, so it's important for us to connect to the segments of the market. However, low code is the underlying way that we deliver the application. So I would say, first, we're low code, but we go to market not around just a low-code platform. We go around the use cases.

I mean kind of like I think we -- ServiceNow in the ITSM space. So I think they really started from a low-code base and have built solutions kind of adjacent from there. How do you kind of view the use case you're going after? What could be your ServiceNow-like solution that ends up getting kind of mass market adoption and could be more defined going forward?

Sure. So ServiceNow is -- and it's -- obviously, we would -- we respect ServiceNow a lot. So comparing ourselves to them is to our advantage. But ServiceNow, I think, their product or their solution is very similar to us, right? It's a workflow platform. It has -- we're executing -- you're trying to automate and execute volume transactions, and you're trying to make the development of that application try to be not development, right? So we're -- I would say we're very much like them. They are not -- we don't see them as much in terms of the space where we sell, the buyers that we sell to versus the buyers that they sell to. So when you think about what our common use cases are, one is around customer service. So that customer service could be actually like a digital contact center or could actually be managing exceptions or things of that nature, dispute exceptions. It depends on the vertical you're in, what were they -- what were each vertical uses. But there's a common like I take something in from a customer and they need to execute something through a series of steps. There's also things like onboarding, customer onboarding. It could be a loan origination and insurance underwriting. It could also be bringing on a new participant in a cell phone plan. So that's another theme. And another theme is what we call one-to-one customer engagement, which is take in as much information about kind of potentially an anonymous-type user that comes to your website and trying to use that information that you might have to predict what is the most likely thing that they're going to respond to, offer them a credit card, offer them a lower rate, send them to a wealth manager, whatever that example might be. So even though we're not -- ITSM is like a theme area that is very, very specific. I would say ours are more around the problems that we solve that can go horizontal across verticals. Onboarding issues and event management, I don't mean like event management like Everbridge, kind of event management meaning an issue or an incident and then decisions or AI in that. And that -- you put a circle around that and say anything where you're trying to execute volume transactions through a series of steps, and you're trying to get human interaction as much out of that as you can, that's kind of -- that's our sweet spot.

Okay. No, that's very helpful context. I mean I know you do play in the marketing space, at least a little bit around some of these kind of solutions you're talking about here. I think we hear a lot around IDFA policies from Apple and Google cookie policies changing. How do you kind of view the impact of what that can mean for your marketing business and what that could mean if people are using more first-party data to -- at least going forward?

Yes. So the interesting -- that's a really, really great point. Probably the first time someone's asked that question. So there's -- as many of you may know, there's a lot of changes that are happening in terms of how you get access and who controls the access to first-party data. So really increases the need for a solution like Pega because you don't have necessarily all the native information that you might need within an endpoint solution. So you really need to have some type of an orchestration tool that knows the rules of what the client wants, the information that might be resonant or not resonant within the countries and the geographies and then who has access to that data. And then also, what's really important is the rules engine or the algorithm or the kind of the spider diagram about how you probability adjust and think about what outcome do you want. So you're doing something like someone comes to a website and say, "This is Ken Stillwell." I know it's Ken Stillwell because I noticed his IP address, but what I don't know is I don't know these other elements that may go through and -- with my iPhone, which might be the access point. So trying to figure out a way to capture all that data, and in real time, I mean, literally like thousands of decisions per second, we run through our clients to be able to, before the screen renders, make a decision on what you want to put in front of the client. And that's incredibly important in the consumer world. It's not so much important in things that might have more batch transactions like I want to change my election on my 401(k) plan. That could be something that you put in the information and updates the next day. When someone's coming to a website specifically in a retail or a heavy consumer environment, they're going to be about patient for maybe 3 to 5 seconds. And if they don't get something relevant, they're on to another place.

Sure. Okay. No, that's helpful. Maybe switching gears a little bit. You've been going through a big subscription transition for the past few years here. I guess where do we kind of stand today? And what are kind of the remaining steps that you're thinking about to get to your kind of final target stage, at least in terms of subscriptions?

Sure. So it all starts with the change to the go-to-market, right? Because once you actually change the way you sell or change the way you position with your clients, naturally the other steps in the transition are things that just kind of happen through the execution of that model. So the first thing we did was we had to change the way that we sold. We were selling -- we sold -- somewhere in the neighborhood of 2/3 of our business was perpetual licenses before we started the subscription transition. And we went through a classic model where we essentially unwound all the perpetual licenses, sold everything on a recurring basis and did that relatively quick but had to go through like kind of a 5-year transition. First, go-to-market, we did that very quick. And I think we did a pretty good job there. The second one is, it just takes time to get the actual results of that through because you're going to hit get the revenue trough similar to other companies that have went through it. We're certainly not the first or probably even the 50th one that went through that. So you know it takes -- you hit that trough of growth rate. And then as you kind of exit it, the growth rate starts to converge with your ARR to your revenue growth rate. We are at the point now where those have converged. And we have about a year or so left where the next phase, the last phase, is where your free cash flow or your billing start to normalize. So now you get back to kind of normalized operating margins. So we really -- we went from a company that was generating $150 million of cash flow a year on $700 million or $800 million of revenue, and we went to a company that was burning $100 million a year even though our growth rate has accelerated because we're right in the trough. Now we're back to where we're going to be generating close to $100 million a year again, and that's still with 18 months left in the cloud transition. So I'm -- I think we did a great job of the go-to-market transition. I think the revenue and the financials are playing out as we would have expected. And I think the only thing with cash flow is we just need to make sure that we're getting the operating leverage in the recurring business so that we not only get back to our pretransition margins but higher than that. So that's kind of -- that's in front of us, so we need to work on it.

You mentioned in there that there's a change in go-to-market that happens as part of this transition. And I think there has been a bit of an evolution in the go-to-market over the past few years. I guess what's kind of been the biggest change that we've seen then? And how should we think about kind of the sales productivity going forward since you have had a big push on the hiring front there?

So I think the biggest change that we had in our go-to-market is when you sell a perpetual license, you tend to have -- I mean I'm generalizing here, but you tend to try to sell out a client early on, get the client to make a big commitment. They get kind of committed to that. They pay naturally. As you buy more, they get bigger discounts. But the kind of the whole shelfware kind of thing that's been -- so you try to get a client to overbuy. Problem when they overbuy is they -- typically, you have to give them deeper discounts to get the overbuy, but you do get the cash upfront. In the model of subscription, you can't give away -- you shouldn't give away those discounts to be able to get the extra commitment because the reality is if the client doesn't use it, they're going to shut it off and then they get the discounts that you wouldn't have otherwise given them. It's a little different than the perpetual model. So I think that, that change of thinking about the unit economics and how we think about discounting and how we think about the trade-off for commitment from the client and duration against price discounts and then also how we work with customer success and really have an upsell or expansion motion with our clients after they deploy because in the subscription model, you really want clients to start fast and help with adoption and then the future buying will happen naturally. But in the perpetual model, you really want to convince them to buy a lot early so you can get the big commitment upfront. So that is a mind shift for our sales team. And that was probably the biggest kind of transition in our go-to-market.

Okay. And you have put a big investment into place in terms of the number of salespeople you've brought in and building up the channel opportunity. How are you kind of viewing that going forward? And how does the channel investments you've made kind of augment the potential customers you can bring in?

So we've historically tried to sell direct to our clients with, I would say, no help from our partners. In fact, historically, we kind of like kind of pushed our partners out of the way and said, "We'll sell the deal. And when we're good, we'll let you know, and you can go and do the implementation." And I think we've really -- when Hayden Stafford came in a couple of years ago, he did a great job of really kind of maybe opening our eyes to the relationship that the global system integrators wanted to have with companies like Pega to say, "Listen, let us help you sell. Let us be involved in the relationship on the front end." Because the reality is Accenture has better relationships with our clients than what we do, right? They're in there as a trusted partner. So why wouldn't we leverage them. They need to sell Pega because they have Pega-certified people that need to be billable. So there's really a mutual objective here. And so in order to do that, though, is we need to give them some incentives so that they can put business development resources on the front end. So that's where we kind of carve out a small amount of kind of commission, for lack of a better term, to the GSIs to be able to be more looking for leads, helping us get warm leads and helping us even close the deal. So that was a huge first step. The second step now is should we find other similar partners that actually can go a step further, that actually can deploy or sell managed service offerings on our platform to larger pockets of clients. And as our architecture has evolved to more to multi-tenancy, that allows us to offer that on Pega Cloud more than we could before. So we went from really like almost Neanderthal selling to one client perpetual license to selling subscription leveraging our partners and the rest of the ecosystem to now actually saying here is a platform of low code that you can sell these use cases and enabling our partners, Steven. So we're not at that last point yet, but that's the next step.

Okay. No, that's interesting. I guess how are you viewing the ISV community and integrating with the tech partners? I think there was another area where Hayden was really kind of pushing and trying to augment some of that. But how are you kind of viewing that opportunity from here?

So I think there's things we're ready to do with our partners, and there's things we're probably not ready to do. Certainly, we've done ISV models already with companies like First Data where they deploy a Pega application, and they actually do credit card disputes and reissuances for lots of Mastercard or Visa-labeled products through like credit unions and smaller banks. So we've actually run the model even in the -- kind of on the previous architecture. But I think it's very easy to get to the MSP model with either the GSIs or even our clients. Like for instance, if Verizon wants to offer a managed service to, say, all of their corporate customers, they could actually build that service on Pega. So that, to me, I think, is very low-hanging fruit. The next phase of that kind of goes back to what I said earlier, which is can we get to a point where we have a low-code platform that partners actually cannot get in trouble when they deploy any type of application use case that is relevant to the vertical that they're in. And we're not there yet. But I would say with the evolution and what's coming in 2023 with our Project fnx work, which is our next evolution of our Infinity platform, that's where -- that will open up a completely different market, right? That allows us to go down market. It lets us go into noncore verticals. It lets us get much simpler use cases to be able to scale across multiple customers.

I want to ask about Project fnx. But we have about 10 minutes left. Just want to see if there's any questions out there from the audience. So Project fnx, I think we've talked about this for a few years now. What are kind of the big changes that, that -- what are the biggest changes that Project fnx brings? And how does that kind of augment the Pega platform?

So I think the -- if I could summarize Project fnx, I would say, moving from more of a monolithic architecture environment to a cloud-native micro-services multi-tenancy. And that -- and those -- so that -- some companies struggle with that move because they literally have to rearchitect or redevelop the entire product. For us, the interesting part about it was when we first started to build the Pega platform, and I'm going back to, say, like the 2005 kind of time frame, all of our components were actually separate. We actually integrated them together into one platform because that's what the market wanted. Well, guess what, the market doesn't want that now. They actually want micro-services. So for us, all we had to do was essentially kind of unsuperglue those components out so that we could update and kind of manage each of the rooms in the house [indiscernible] back the entire house. And that's -- so that -- in enterprise, that's not as much of an issue for the enterprise use cases that we've had over time. Problem is you go -- if you go into more scale use cases and certainly if you go more into mid-market and [ common ], that's critical, right? You cannot actually have an application that isn't updated or refreshed sometimes daily. So that's kind of the biggest thing. That's what fnx does for us.

Okay. And I guess how do you think about impacting either the model or in terms of the types of customers that you can go after now?

It doesn't impact the model -- our go-to-market. It impacts one potential thing, which is we might see more consumption-based arrangements, right? So similar to a little bit of some of the conversations I've heard from Frank and Mike at Snowflake, like moving -- where you're actually moving to more of a consumer consumption-based model, you are kind of -- you build some level of predictability in the adoption curve. Well, that will be different. When we start to go into more clients, we will have to really bet on the adoption curve and think about how the pricing works in that consumption model. And the reality is if you have multi-tenancy and you have a modern application, you can manage that just by how much capacity you procure in your cloud environment versus if you're a single tenant, you have certain fixed costs that you need to deploy against each tenant. So you really can't have a true consumption model because the economics don't work until clients get to some scale.

Okay. That makes sense. Maybe you talked in the past about Hayden and the changes he's made as part of the -- as part of bringing him as Head of Sales. With him leaving, are there kind of any changes in terms of the go-to-market strategy? Or how are you kind of thinking about what that means from here?

So much, if not most, of what Hayden did is quite valuable. And we won't -- it won't -- we won't change it at all. I would say the only thing is the sales management cadence, the working with partners, the pushing out the way the sales teams -- really just the discipline of how you manage the teams and manage the pipeline, just the best practice. That's all great stuff that we will take his good work and continue on. The one thing that I would say we probably -- probably was a step ahead of where we were ready was pushing go-to-market into new logos down market where our product wasn't quite there yet. So I think we probably got a little bit ahead of ourselves there. We will need that, but we probably don't need it as much in 2022 and '23 to execute our strategy. If we were going to tweak anything, that would probably be the only thing. But everything else is very reusable, and we love where we are.

Okay. That's good to hear. Just in terms of the guide that you gave and the outlook you provided, how are you kind of thinking about what the sales productivity rates that are kind of implied in that? Are there any areas where we can potentially see upside on that front?

So we had -- we improved sales productivity from '20 to '21, I would say, a little. Certainly, we're nowhere where we need -- where we want to be or need to be. I would think that there'll be kind of an order of magnitude similar to that improvement from '20 to '21 from '21 to '22. So we'll be moving in the right direction. I think what will happen is we have a lot of sales capacity in the system. And so I don't think we need to invest a lot more in net new sales capacity to be able to achieve the growth rates that we're on, quite frankly, even accelerate our growth rates. It's really around execution and ramping. So I think we're in pretty good shape. I don't think we're not in a situation where we need to make a big noticeable increase in go-to-market. I think it will be very kind of incremental from this point.

Okay. And I guess with that context, how should we think about the -- where kind of operating margin could go from here after this? If we're talking about rep growth potentially slowing down a little bit and also exiting in the cloud -- or the subscription transition, excuse me, what could that potentially do for the margin profile?

So the way we're thinking about it is we're thinking about something like 500 basis points of margin expansion each of the next 3 years, right? That's kind of the way between 2012 to -- excuse me, 2022 to 2024. Some of that is just because you're coming out of the cloud transition. Some of that's just natural tailwind that you would have. Some of it's because our Pega Cloud gross margin is not at kind of a timeless level yet. So you kind of -- you'll get a little bit there. . And then with some of the sales efficiency improvements, I think we're at like very low single digits because like I said, we're not out of the cloud transition. Next year, we'll be in the high single digits, and then we'll be in the low kind of double digits, and then we'll kind of get up closer to where we're achieving something closer to Rule of 40 once we're at that first year out of the cloud transition.

On ACV side, I mean, I think historically, we've talked about potentially getting up into the 25% plus range. I think you just guided to 21%, low 20s. I guess what would it take to kind of get back to that range that we [ can potentially ] see an acceleration there?

So we -- I think the 2 biggest headwinds we've had is both of which probably we could have managed better than we did but also were a little bit out of our control. One is the impact of COVID, and the hiring profile that we did for salespeople in COVID is not the hiring profile that we have now that we're exiting COVID. We need -- we're not -- selling is only partially face-to-face. There's a lot more digital engagement that needs to happen. And the team mix needs to be different post-COVID. So I think us working through that with the right sales profile probably -- we probably turned over and had to rethink about our enablement and the sales team through the process. And the second one, like I said, we could have done that better. But I don't think when -- back in June or July of 2020, I didn't predict that we were going to be talking about COVID in the beginning of 2022. So that wasn't something that -- and I jokingly say I've like mispredicted the end of COVID like 5 times, so I'm not going to try to do it again. But I do think that that's one. And then the second one is turnover in tech has just been higher, right? So like we typically have 15% turnover in our sales team, somewhere about that. And our turnover has been at least 10 percentage points higher than that in the last 12 to 18 months. So every time someone turns over, you've got to reramp them again, right? And we're an enterprise seller. So think people don't ramp in 1 or 3 months. It typically takes a year or 2 to really get things anchored. So those 2 things are the biggest headwinds for us that have slowed down us accelerating our growth with the increased investment. And when I guided for 2022, I was naturally impacted by the fact that Hayden gave us notice he was leaving like 2 days before we released earnings. So I kind of did tweak our guidance a little bit to be -- to take into account that we would have -- there has to be some impact from a go-to-market leader leaving. Although we haven't had any turnover since he's announced, and I think we'll be pretty good on the turnover front, it's still a disruption. So we had to take that into account.

Okay. That's good to hear. I think we'll leave it there. We're running up against time. Ken, I want to thank you so much for being here today, and I want to thank everybody in the room for being here as well. So thanks again.

Awesome. Thanks, everyone.