Tyler Technologies, Inc. (TYL) Earnings Call Transcript & Summary

Good afternoon, everyone. I'm Pete Heckmann. I'm one of the tech analysts at D.A. Davidson. Thanks for joining us today at the Technology and Consumer Conference. Today, we're going to be chatting a bit with Brian Miller, long-time CFO at Tyler Technologies. Brian, thanks for participating again.

Yes, you bet. Thanks for having us.

Yes. Well, what I wanted to start on is get back to a little bit of the basics. As most of us know, Tyler has a strong brand and market position, providing automation software to the public sector, primarily municipalities. But can you talk to us a little bit just how the business breaks down between municipal customers, state-level customers, customers of the federal government?

Yes, sure. Tyler's historic presence has really primarily been at the local government level. So cities, counties, school districts, local agencies make up today about 75% of our business -- 70%, 75%, and that's kind of where our roots were and where we grew for the first couple of decades. More recently, we've expanded our presence at the state level through -- primarily through the acquisition of NIC back in 2021, which gave us some strong state relationships that we're now leveraging to sell more software into state governments. Most of our business at the state level is currently under a transaction or a self-funded model. So at the state level, we primarily have -- provide state portals and have the interfaces for citizens or businesses to conduct business with state governments, renewing your license plates or getting a fishing license, and that's all funded through convenience fees as opposed to being a budget item. And then the federal level is less than 5% of our revenues. And that came to us a few years ago through an acquisition of a company called MicroPact. Primarily what we provide at the federal level is a -- through an application platform, a low-code platform that's used to manage a variety of business processes for federal agencies, things like background checks and security clearances. Our platform is used by a lot of agencies, including the Department of State to manage those, EOC claims, veterans benefits at the state level, but still a pretty small and targeted presence at the federal level.

And then from an international perspective, I know you have some customers in Canada, maybe a customer or 2 in Australia, but by and large, primarily a domestic business.

Yes, about 98% domestic. As you said, most of that 2% is in Canada. We have some property tax solutions in provinces in Canada, school bus transportation solutions, court system in Australia. But currently and in the near future, certainly expect to continue to be mostly domestically focused. There's a lot of runway ahead of us in a very, very fragmented market in the U.S.

Sure. So let's speak to that. I mean, in the past, you've talked about in terms of sizing the TAM, thinking about like the number of municipalities and the number of systems that they would need. Can you talk a little bit about that and your penetration within that?

Yes. So even though we have, by far, the biggest customer base of anyone serving the public sector and by far, the broadest product offering of anyone serving the public sector, we still have a relatively small market share depending on how you define it in individual products, it varies. But still pretty small penetration. The government market, especially the local government market has historically been very fragmented for a long time before Tyler was really served by a lot of niche companies who are -- and still, they have a wide swath of market share, companies that are focused on a narrow product focus and often narrowly geographically focused. So someone that does court systems in California or tax systems in New York and New Jersey. So stepping back a bit, there are about 88,000 local government entities in the U.S., 37,000 cities and towns, 3,200 counties, about 15,000 school districts and then 30,000 agencies, airport authorities, water districts, fire districts, those sorts of things. They each have multiple software systems. Obviously, a city would have a -- everybody has an ERP system of some sort. A city would have a public safety system, a court system, a licensing and permitting system. Schools would have ERP, bus transportation, student information system. So the average jurisdiction could probably -- if you think about big suites of products, 8 to 10. So if you take the 88,000 local government entities times, if you just take 5 major systems, there's 450,000 solutions. We have about 44,000 installed solutions. So you could kind of think of it as maybe a 10% market share. Now to be fair, there's a wide -- a fairly sizable slice of that, that's really small places. So the 37,000 cities and towns, the ones that are like under 1,000 people probably are -- wouldn't be a meaningful market. But still, it's a very fragmented market with us being the leader, but still a small market share. Now that's kind of the existing market. Our win rates are fairly high. Typically, with our flagship products, we win more than 50% of the new business. So as these old systems turn over, we're winning a larger and larger share of those. And so our market share continues to grow with a really long runway ahead of us.

Really long runway, yes. And so in terms of -- one of the things that you've talked about in the past is that there's a very high number of municipalities that may be running on older in-house systems that are harder and harder to maintain as well as software from vendors that no longer support that software. And so talk -- without an incumbent provider to compete against, talk about that opportunity? And what are the catalysts to get them off those systems?

Yes, that's really kind of the key, and that's one of the ways that the public sector market really differs from the private sector and how they use software and how they think about replacing it. We think if you look back at all those hundreds of thousands of systems that are being used and catalog where they came from, you'd find that -- probably more than 60%, we believe, of those systems are kind of fall in the legacy category, either they were homegrown systems, some of these written in the '70s that were written in COBOL in languages where there are no more programmers around. And some of those in very, very large places or systems from a vendor who's no longer competitive, someone who was competitive 10 or 15 or 20 years ago. So a lot of systems. Those are still being used. They may still be in business collecting maintenance and supporting those systems. But at some point, they didn't invest in the next generation of technology, and that may have been a couple of generations ago. And so they don't have a product anyone would buy today. So as those systems finally get to end of life, and that's the big difference between public sector and private sector, governments don't have competition. They don't really think about these traditionally on an ROI basis, and they don't like change. So they tend to use these systems until they kind of literally are dying and they are forced then to replace them. It becomes a nondiscretionary decision. So it creates a very steady kind of flow of business. But when it's replaced, then those systems won't be replaced by an upgrade with the existing vendor, they become a new opportunity for somebody like Tyler, and that's where we win a high percentage of those. Historically, it's kind of hard for us to accelerate that process or to create demand, but it creates a very stable kind of demand. We might talk about this a little bit more, but in this environment of focus on government efficiency, that's a good thing for us. In terms of our longer-term view, we believe that increasingly, governments will -- are starting to change kind of the way they think about it and starting to kind of have an ROI lens and say, okay, if I replace this, I don't have to do it for 10 years, but if I do it today, I can see these benefits of providing better service, doing what I have to do with fewer employees and creating efficiencies that are starting to potentially accelerate the replacement of some of those systems.

And certainly, we've heard a lot about cybersecurity and so the kind of the dual pressures of needing to have more secure, more modern systems, but then also automating more. The federal government did have a stimulus program, ARPA that I think created -- I think it catalyzed some spending. And can you talk about that? I think that those funds had to be earmarked by the end of 2024 and have to be spent by 2026. But certainly, it seems like when you start to read about some of the concerns about hacking at the local and state government level, it suggests that there may be other -- whether it's either stimulus or other things that could potentially maybe not accelerate, but certainly encourage upgrades?

Yes. I think there is -- just broadly, there is a sort of a growing theme of digital modernization. And even if you read the executive order that created DOGE, it's got a whole paragraph in there that says, we need better technology. We need to upgrade software and we need better connectivity between agencies, and that's at the federal level, but that certainly applies at the state and local level as well. So there's sort of a recognition that the way they get more efficient is through technology. And so there does seem to be a growing focus on that. Yes, the ARPA funds provided money for -- initially intended to offset impacts of COVID. As it turns out, governments really didn't suffer a lot of impacts. There weren't massive shrinkage of revenues. They didn't have massive costs at the local level around public health. And so it kind of became a whole lot of free money. It was about $350 billion that was given to state and local governments. Everybody got something. I'd say a lot of it was spent on onetime things. So a lot of infrastructure kinds of things, but a wide range of things, things [indiscernible]. Some cities gave every teacher a onetime bonus with their ARPA funds. It's hard to quantify exactly what the impact on us was. It clearly was a positive. We think it helped create what we've said over the last couple of years has been a pretty active market, but continues to be active at about the same level in terms of new RFPs and just general demand environment. So it kind of created a confidence around their budgets when they might have otherwise had more concerns around it. We think there were some incremental sales, but not a massive amount of incremental sales. And because we are pretty much an all recurring revenue model now that somebody using ARPA funds to enter into an agreement with us, maybe pay for the services around it are signing up for a long-term recurring revenue stream. So it's not like with us that we're one time or one-time matter. We did see a little bit of growth in hardware last year. We saw some hardware around some of our school bus systems. And we've called the relatively low-margin revenue, we've called out that, that's expected to decline this year because there was some onetime pushes. We also saw a little bit of a pull forward of demand around that deadline for committing the funds in December that there were some deals that would have otherwise occurred probably in Q1 or Q2 that were pulled forward, not huge, $10 million in kind of that range. But -- so we don't see a big dropoff post ARPA, but it did create a bit of a tailwind, which is that money gets spent over the next couple of years will continue to play out.

Right. One of the things that surprised me a little bit is that because you do have 7 areas or so of competency and most municipalities need each of those. It does -- because of the way the purchasing is siloed, there are definitely some municipalities where you have 1, 2, 3 major products, but there's a lot where you just have one. So talk about some of the barriers to kind of selling an integrated solution.

Yes. We've talked about that as one of the primary growth pillars as we look out over the next 5 years and beyond is cross-selling and leveraging our installed customer base. So as I said, we've got the broadest product offering, and we've got the broadest customer base, but our average customer has 2 or 3 products from us and could have 8 or 10. We're talking about major product suites. There's also an opportunity within a suite of products to sell more things. You're correct that they are typically sort of bought in silos by an agency had a department, had a police chief for the public safety system, a tax assessor for the tax system, a courts administrator for the court system. But there is a value to having multiple products from Tyler and an added value that we think gives us a competitive edge. Our products have common foundational elements that work together more effectively than if you were putting together systems or had multiple systems from different vendors, things like dashboards and security and sign-on and workflow engines. We have common technologies like a data and analytics platform that sit across all of our products. In some cases, products are very closely integrated like courts and justice and public safety. And so there's very significant benefits if you're one of the 55% of the country that has our court system and you're looking for a public safety system that out-of-the-box integration creates significant additional value. Also, references and reputation are a really big part of the buying process because governments don't have competition, they talk to each other a lot, and they take a lot of comfort in buying the same system that 5 towns around them have successfully implemented and use. And they talk to each other a lot. So if you're successful and you have a good reference -- good reputation and strong references, that's a big plus. If you're a newer entrant, your references are across the country or you have bad references, they all know that as well. So we do very well in that category. And so that creates an additional competitive advantage as we continue to try to cross-sell. We've done a lot of structural things in the last couple of years to remove barriers or create better incentives internally to encourage cross-sell, things like aligning sales commission structures to provide the right incentives to multiple sales reps across different products. So we think we've put into place a better framework to accelerate cross-sells. We also -- as we make acquisitions or as we build things, some of the things like we're doing around AI right now gives us more things to cross-sell, more products to put in the same sales reps bags, so we can do that really efficiently with relatively little incremental selling costs. So that's all a part of the model as well. And then things like the payments platform that we got with the NIC acquisition as we are accelerating the process of pushing that payments platform into our software customer base.

Well certainly, that was one of the synergies from the NIC deal that we thought was pretty exciting. And I would say you've outperformed my expectations there. But I think hundreds of payments deals each year for the last 3 years. And so that's becoming -- while each deal might be relatively small, that's attaching payments has become a pretty powerful part of the growth algorithm.

Yes, it really has. They were kind of -- besides NIC just being a good business on its own that we were able to acquire for a reasonable valuation, and you followed NIC -- you were one of the -- a very small handful of analysts that followed NIC. So you have a good perspective on that. They -- we saw 2 major synergies there. One, the opportunity to leverage their deep state relationships in the 28 states where they have enterprise state agreements and sell more Tyler software into state governments where we historically didn't have a big presence, didn't have deep sales organizations, but we had products that would apply at the state level. And so we've had success with that. Still, I'd say, in the relatively early days of that, and we talked about in the fourth quarter that we just are in the process of building a new state-focused enterprise sales team that's going to sort of bridge those relationships with all of the Tyler products we have to kind of kickstart that. But then taking the NIC expertise and the technology they had in the payment space and leveraging that and what you were referring to because Tyler has a lot of software products that present bills, process transactions. So things like utility billing systems, we have thousands of those, municipal court systems, managing traffic tickets and fines and fees, licensing and permitting systems, anything your local government licenses you for, property tax systems, parks and recreation, all kinds of systems that are presenting bills, creating transactions. But historically, we didn't have that technology or expertise around payment processing. And so we really, for several years, kind of addressed that through being a reseller and bringing in third-party payment processors like Chase or Elavon or OpenEdge and getting a revenue share from them. But then NIC brought us this deep experience and robust payment platform that's managing tens of billion -- or processing tens of billions of dollars of payments each year at the state level. And so what we've done is integrated that technology with our software solution so that we have a fully integrated software and payment solution that provides more value than sort of a generic, sort of horizontal payment processor. The biggest value being the automation of reconciliations from the transactions and the payment as well as providing better reporting, more tailored reporting, better analytics around their payments, in some cases, better security features. So providing a payment solution that governments value more highly that we can sell with a new software solution or back into our installed base of customers who have those systems from us and get premium pricing, better margins, better pricing than historically in some of the more, I guess, commoditized payment arrangements. As you said, we're still -- we've had our go-to-market down for a few quarters, but still building a lot of momentum there. But we've done 1,500 payments deals in the last 7 quarters with Tyler software customers that have added about $50 million of ARR, and we're adding a couple of hundred every quarter and doing some enterprise-wide payment deals as well. So still in the early days of it, but definitely encouraged by it's playing out the way we had envisioned when we made the acquisition.

Definitely. It's funny, over the years, companies have given guidance 1 year out, and we had felt like it was a best practice for companies to give 3-year aspirational guidance, but a year 2023 Investor Day, you gave 7-year goals, right? And a big part of that included the discussion about moving from a traditionally on-prem software model into the cloud. Talk about where you are in that transition? And then why did you decide to go 7 years out? What gave you the confidence to be able to go that far out?

Yes, that was a new one for us as well. And 2023 was sort of an inflection point for us in the cloud transition. It had been a couple of years since we had the last Analyst Day. And in between there, we have done -- we've been through COVID. We've done the NIC acquisition, which was by far the largest acquisition in the company's history. We were -- we had gone from being sort of a cloud agnostic. We'll sell software in the cloud or on-prem, didn't really care to being cloud first. That kind of took place in 2019. So there were a lot of initiatives underway that we thought it was important to make sure investors were aligned with how we were thinking about, how we saw those playing out. And so yes, we gave sort of 2-year targets for 2025 and 2030 targets as well. It kind of aligned with internal processes. We have a Tyler 2030 vision that we had been an internal process. So we actually were doing some early planning towards that and setting targets for us internally. So we were comfortable sort of aligning all those. All those targets we talked about were organic. We do M&A. That's the biggest use of our free cash flow. But we don't have quotas or we're disciplined and opportunistic about M&A, so we didn't layer those in. So there will be M&A, but those were organic targets. Around the cloud transition, yes, 2019 really was a year where we were -- half of our new sales were cloud and half were on-prem. Having sort of proclaimed ourselves as cloud first at that point and really starting to only sell new customers, for the most part, cloud software and launching a process to accelerate the migration of our on-prem customers to the cloud. 2023 was really the year where we hit the trough in margins from the pressure, from the licenses that big upfront money going away and us building up a recurring revenue stream that is from the average customer about 2x what the recurring revenue stream would have been on-prem. So we crossed that point. Also sort of hit the trough in terms of revenue growth with the pressure from licenses going away and now having that recurring stream build up. So we set out targets for both revenue growth and talking about sort of low double-digit growth in recurring revenues, 10% to 12% growth there, SaaS revenues growing high teens through 2030, free cash and then margin expansion, roughly an average of 100 basis points a year, although not -- would not be linear, but going from a 23% operating margin in 2023 to a 30% plus margin in 2030 and similar kind of free cash flow margins. So a significant increase in free cash flow with a target of $1 billion of free cash flow by 2030. At our investor -- sort of session at our user conference last month, we kind of updated everyone on where we are with respect to those '25 targets since we've now got guidance out there for '25 and how we're looking towards the 2030 targets. I guess the headline is, in all respects, we're either on track or ahead of track to achieve those. And so for '25, certainly, especially on the free cash flow, we're well above it. We talked about a free cash flow margin of 17% to 19%. Our guidance is for 24% to 26%. So -- but on the revenue growth, the cloud transition, some of the aspects about that, exiting our data centers, we're pretty much right on track with all that. So we have a high degree of confidence about being able to achieve or exceed those 2030 targets. And we'll expect to have another kind of full Investor Day next year where to the extent we revise any of those, that would probably be the point.

And remind me for the 2030 goal in terms of having customers into the cloud, were we talking about 80% of ARR by 2030?

Well, what we said is -- and today, we're about -- we are 85% recurring revenues. But in terms of the on-prem versus the cloud, we said at the starting point being 2023, we expected that of those customers that were still on-prem, that 80% to 85% of them would convert to the cloud by 2030. And we're probably, at this point, maybe about 25% of those have moved. So we're kind of on track. We said it would be kind of follow a bell-shaped curve trajectory and that the peak would likely be in '27 and '28. So both the number of clients slipping from on-prem to the cloud and the average size of those clients is continuing to grow through that '27, '28 time frame.

Right. And on the balance sheet side of it, you did incur some debt with the NIC deal, focus the cash flow on rapid deleveraging, converted back to net cash, what, 2, 3 quarters ago. So how do you think about capital allocation from here? You've done a few little acquisitions. And certainly, over the years, it's been clear that acquisitions are -- you're very competent at sourcing, negotiating and integrating acquisitions. So do you see more acquisitions in the future? And if so, kind of how would you characterize them in terms of size? And maybe if possible, what niches are you looking at?

Sure. Yes, our balance sheet is in great shape. We've always taken a pretty conservative approach to our balance sheet. Given the high level of recurring revenues in a very stable market does make it easier to take that approach. We've typically had little or no debt other than around a couple of times when we've had larger acquisitions and have been able to delever pretty quickly. I think at the time we did the NIC acquisition, which was a $2.3 billion purchase price, we had a fair amount of cash on the balance sheet. We did a convert, which will mature next March, which is our only debt today of $600 million. And we did some term debt and have the flexibility to prepay that and pay that off well ahead of schedule early in '24. So that's given us now the really strong balance sheet with 0 net leverage, net cash. And a lot of flexibility around financing future needs. The biggest use of our free cash flow and what would potentially drive additional financing needs is M&A. We've done about -- I've been at the company for 27 years. We've done about 60 acquisitions over that timeframe, most of which would fall in the smaller kind of tuck-in category and a few that were larger. I'd expect that going forward, most of -- at least in the next 2 to 3 years, most of the acquisitions would still kind of fall in that tuck-in category, although that can range in size from a deal that's a few million dollars to something that could be tens or a few hundred million even. I think the more than $1 billion kind of acquisitions, we're probably still a little bit further off just from what we sort of see on the horizon today and from how we've been focused from a management attention perspective on a lot of these big initiatives around the cloud and around payments. As we get a little bit further down the road, I think the possibility of those larger acquisitions kind of increases, but maintain a very consistent discipline around valuation, which has been difficult at times with -- especially with some of the PE activity around strategic fit and around cultural fit is a big part of it as well. I think there are lots of good targets out there. We've tended to be really successful, especially in acquisitions where we identify a gap in our product portfolio, maybe something adjacent to one of our core product areas. Sometimes the technology. We've done some acquisitions that have brought in some AI technology. Often, it's a partner, somebody we've already worked with in an adjacent space, a company we know already and that knows Tyler. So we've been really successful in those kinds of acquisitions. I think that's kind of where our focus stays. We don't have any big gaping holes today. But there are a lot of peripheral things that are applications that can add value to our existing product suites. Generally, these are things that we can leverage our existing sales force. So the same sales force can now have more products to sell, so we don't have to hire a lot of sales reps to sell more and things that we can leverage our customer base. So now we can sell them back into that gives us more cross-sell opportunities. And so we're looking for businesses that we can grow -- even if they're relatively small today, they can grow substantially faster than Tyler's core growth rate. And we think there's a good pipeline of those. And we have a reputation as a good acquirer and a good manager of those businesses, and that track record is really strong. So I'm really talking about specific areas that we're looking for shopping for something, but we've always got a pretty good white space list. I would say that with our increasing focus on state-level business through the NIC, although we already have a lot of Tyler products we can sell to state governments. Other applications that are more state-focused could be of interest to us. We acquired, for example, a company in the outdoor recreation space, so software to manage campground reservations and hunting and fishing and all those sorts of things after the NIC acquisition, and that turned into what was -- is now the biggest contract Tyler has ever done with the California State Parks organization for a couple of hundred million dollars. So again, a really small company. We bought a product from, and we're able to leverage into some big recurring revenue numbers. So focus is really consistent with what we've done over a long period of time, just the average size, I think, continues to grow to really be impactful for us.

Right. And with your high win rate, I mean, win rate can kind of vary based on the functional area. But with high win rates, does it make sense to do like a consolidating acquisition to give customers a migration path? Or is it that you kind of feel like they're coming anyway?

It depends. On one hand, we haven't done many consolidation type acquisitions, especially because over the years, we've really focused on integrating the products we have to work more tightly together and creating products that look and feel the same. So adding -- so we have industry-leading solutions in public safety and ERP and property tax. So adding another solution there complicates things. It brings you customers in that space, but either you're going to have to merge into one solution or you have to maintain multiple solutions and integrate 2 systems into your others now. But the other side of that is it brings us customers that now create more cross-sell opportunities that we can sell more things to. So it could be that more consolidation acquisitions. I don't see a flood of them, but I think there are some areas where that could be an area that we jump into a little bit. There are a lot of these, like I said, these niche players that are kind of legacy products that you would likely not continue to build those products going forward, but have a migration path into your existing products. And the cloud does make that easier as we get down to having one version of each product in the cloud, it should simplify things for us.

Definitely. And over the last 5, 6, 7 years, certainly, you've been winning larger contracts, California Parks, North Carolina, you did major implementations in like Cook County, Harris County. Talk about those bigger contracts and kind of the relative expectations in terms of implementation. But then also, even though you're winning these bigger contracts and bringing them on, like the customer concentration is still very low, right? I mean your largest customers.

Our biggest customer probably wouldn't have annual revenues, even like a Cook County might be our largest, but they have less than $10 million a year probably of annual recurring revenues. At the time we sell them a product, there may be a higher around services in the old days, licenses could be a multimillion dollar license upfront. But today, yes, we don't have any customer concentration to speak of. We've got 14,000 customers, different cities, counties, school districts. In some area -- yes, the sweet spot is really midsized from kind of lower mid to upper mid, and that's where most governments are. There's only -- I can't remember the number, but it's surprisingly small cities that are even more than 250,000. It's only like 35 cities in the country that are that size. So most of the cities are right there in the middle part of the market. And that -- to the extent that we compete with the big horizontal players like Oracle and SAP in the ERP space, they don't scale down very well. So that's really where we work well. Now some of our products, though, like Courts & Justice, we are by far the leading provider of court case management systems, about 55% market share, and we have 8 of the 10 largest counties in the country. We have L.A. County, Cook County, the 2 biggest, Miami, Atlanta, Dallas, places like that. Property tax systems, we're a very clear leader there, especially at the high end of the market. So we have New York City's property tax system, which is the biggest taxing jurisdiction in the country. Cook County's property tax system as well. And public safety, we've had a lot of success in moving upmarket there since we acquired that business. Last year, we won 4 or 5 state police agencies, which are all Tier 1 kinds of clients. So we've got a good mix. The big deals are lumpier. They don't -- especially from a bookings perspective, they don't come along every -- there's not several of them every quarter. And the timing on those can be really long sales cycles and a little bit difficult to predict. Last year, we had several really big deals. So it was a really strong bookings year. This year, it doesn't look like as many of the big deals. It doesn't mean the market has fundamentally changed. It's just kind of the timing of when they come out. Our biggest deal last year was a court system for the state of Kentucky, which I think is our 17th state court system. But it's the first one that's really been in the market in like 3 or 4 years. And so that was like a $29 million deal over several years, 6 years, I think. They actually used ARPA funds to prepay 6 years worth of SaaS fees and paid it all upfront. So we are -- we do have a strong presence there, but still the mid-market is really where most of the deals are.

Yes. On the court side, so I think -- remind me, but like within California, I think you had something like 28 of the 33 county or largest county...

It's a big percentage.

Yes. But I'm sure it varies, but like some deals are at the state level, and they would apply to all counties and then some are on a county by county.

Yes. So on the court space, the courts are primarily run at the county level across the country, criminal courts, civil courts. There are local courts, like municipal courts, which is kind of separate. That's traffic tickets and miss the meaners primarily. But criminal, civil, probate, family courts, those are primarily run at the county level, most places. But 40 of the states have statewide court systems. So the state has bought or built a system that all of the counties use. That doesn't mean they're all integrated. They may have 40 different instances of the same software. The other 10 states are county by county, and those tend to be the big states. So Texas, California, Florida, Georgia, Illinois, Ohio are all county by county. That actually sort of creates a bigger opportunity because the sum of all the counties is more than they just bought a state system. We have a really big presence in a lot of those Florida, almost all the major counties use our system, Texas, something like 80% of the population is in a jurisdiction that uses a Tyler system. And California, really strong presence there as well. California was interesting because they -- probably a decade ago, they started a process to go the other way to build a statewide court system that all of the counties would use. And they hired Deloitte to do that, spent a lot of money, tens, maybe a couple of hundred million dollars on a project that was supposed to be much, much smaller and never got there. And so after several years and a lot of money, the state abandoned the product or the project and told all the counties, sorry, you're on your own again. And there have been a lot of in action of systems that would have been replaced, but they were all waiting for the system that was going to come from the state. And so there was sort of a flood of business when that project failed, and we won almost all of it. So Los Angeles, San Diego, Alameda County, we won like 25 California counties in a very short period of time. And so it just goes to show there's been a lot of failures of custom build systems in recent years. And so the market seems to have figured out the good off-the-shelf software from a vendor like Tyler, it's a much safer option.

Yes. Did anyone in the audience have just a quick question that they wanted. I just have one more.

Did you say earlier that cloud customer add revenue twice what would have been in on-prem?

Yes. So the question is about the difference in revenues -- recurring revenues from a cloud customer versus an on-prem customer. So yes, as our existing on-prem customers move from on-prem to the cloud, we average about a 1.7 to 1.8x uplift. Now part of that, that's not all net revenue to us because we then, in turn, pay hosting costs to AWS for part of that. So -- and the client wouldn't view that as their cost going up by that level because they would have people that they don't need anymore or jobs they don't have to fill around systems administrators and applications. They'd have hardware they don't have to pay for, data center costs, security costs. But in a new customer, it's about 2x what they would have paid for maintenance for an annual basis. Now we don't get the license upfront. So say we would have gotten $1 million license and then $200,000 a year in maintenance. Instead, we get $400,000 a year for SaaS. So it takes about 4 years to recover the license. But then forever, we get 2x.

With the long runway that we've talked about in municipal, I guess we talked earlier some federal contracts, a little bit of international revenue. But with the long runway in the U.S., there's really no need for you to do those -- force yourself to do those deals...

No. We've done opportunistic like a court system in Australia. They have a similar justice system. They speak English. It was not a big development effort. It was a logical sale. And we saw the potential that at some point to potentially sell more systems there. But -- and tax systems in Canada. They have a very similar property tax system as us. We are doing a system in Montreal right now. So it's the first one we've had to do in a foreign language. So it's dual language, but same technology. But yes, in terms of just like sending sales reps all over the world to try to sell products that's not on our horizon.

Yes. Okay. Well, thank you, Brian. I appreciate your time today. Thanks, everyone, for participating, and we look forward to additional developments.

You bet. Thank you.