Let's go ahead and get started. I know this is going to be an exciting session that we've all been looking forward to, and you know, there's other reasons, but mainly because of Sandeep, and he's always a highlight of this conference for me, so Sandeep, thanks so much for being here.
Yeah, absolutely.
I'm Rishi Jaluria, I cover software here. Sandeep, as you know, CEO of Clearwater. Before jumping into the real questions, look, there's an elephant in the room around just the potential, you know, the rumors and reports we've been seeing. Would you care to comment on any of that?
Tim, I thought that would be the last question of the day after we get to the business and the GRR and all of that. But look, I think the company, and I'll give you my perspective on it. I think the company has obviously performed really well. I think it has continued to grow over a protracted period of time. But if you listen to Rishi and all the other analysts in the industry, they have us at an EBITDA of $331 million for next year. The company has about 300 million shares, give or take a little bit. It's buck 10 of EBITDA per share. We were at $16 or something, and that's 14 and a half times next year's EBITDA. It just is very attractive, I think, to lots and lots of people, which is why I'm here.
But I thought it should be attractive to all investors. I'm not quite sure why it is not. And have there been interest from private equity? I've got to tell you, I've always had, we always had strong interest from private equity pretty much throughout the eight-year history I've been the CEO of the company. So there's a lot of interest. There's interest today. And I think that's probably the best comment I can make. I do think, you know, specific names, I don't know how people come up with those names, but it doesn't take a genius to figure out who could do a deal of this size, and you sort of put the name in there to be more attractive. And yeah, that's.
Perfect. I think we got that out of the way. Now we can [cross talked] get into way more fun stuff. So I want to talk a little bit about maybe before we even get into some of the acquisitions, I want to talk about like the evolution of the business from IPO, you know, until kind of the evening of these acquisitions, and you know, you've taken this business, obviously seen accelerating growth, gone through a business model transformation, not even pricing transformation from like AUM, AUA to kind of more of a base plus pricing model, and you know, dealt with interest rates going from effectively zero to super high. Can you maybe talk us a little bit about the evolution of Clearwater from IPO until now, and then we'll obviously get into the acquisitions?
Yeah, you know, Clearwater, about five years back when it did the IPO, was very much a regional company. It was a smaller company, which was mostly U.S.-centric. And I think what we've been successful with is doing a few things. One is it has become a nicely international company. That's number one. Number two is the gross margin used to be in the high 60s and frankly is now at 78.5 l ast reported. And we also said at the end of Q3 it was 82% for the core business. So the margin profile of the business has changed quite dramatically. When we started, it was 24% EBITDA. And Q4, Q1 of this year, I think we said the standalone business was at about a 35.5% EBITDA. So again, that margin profile has moved very dramatically. And finally, about 89% in Q3 converts to cash. So it's not EBITDA.
Sometimes I don't like EBITDA. I'm just an older guy, and I like cash. And so when EBITDA converts to cash at 89%, that feels like it's a solid movement. Just in terms of the businesses, the business now sells to asset managers, it sells to asset owners, it sells to insurance companies. I'm talking about the organic business. We have really strong wins out in Europe. That's grown quite a bit for us and is frankly a big lever of growth. We also have started to grow in Asia. And this year, our largest deal by far has come out of Asia. So we feel like we are making inroads there. So the business has just continued to mature. But the architecture of the business, single instance, multi-tenant, all the clients use exactly that same platform.
All the clients, the data of all of our clients is in one database, is unique and allows us to learn quickly, allows us to improve margins because of the network effect. Frankly, generative AI, that's how we have got this dramatic expansion of gross margin because generative AI can help us after it learns from all of this data, which is in one database.
Yeah. No, I think that's great. Maybe let's talk on kind of the organic NRR. I remember at your analyst day two years ago, you gave this glide path to get to 115% NRR. I still think you should have had 115% NRR t-shirts given out of that, but great analyst day. So maybe can you walk us through number one what are kind of the building blocks to get to that and sustain it? And I know you've hit that before from higher gross retention and the likes, but on kind of a sustained standalone basis, and I know that's going to get harder. What are kind of the building blocks to get to that 115% NRR?
Yeah. So I will, why don't I use that to also contrast where Enfusion is, because Enfusion, if you notice, has an NRR 102%, 103%, something in that zip code. And sometime back, Clearwater was in that kind of zip code of 103%, 104%. So what are the elements? One is high GRR. You have to first start by, I'm not going to allow clients to leave me. That's the wrong word. I'm going to go delight clients so they don't leave me. But that feels like step number one. If your GRR is 98%, you have to grow from there. But if your GRR is 93%, then you have to get 7% growth just to stay alive and come back to 100%. So it starts with really high GRR. And Clearwater's GRR has been 98% for, I think, 25, 26 quarters, something of that dealt, right? Consistently that.
I think it was 97 Q1 . It was 99 Q3 . But you get the message. It's a really strong NRR growth. Then I think what we did is we changed our commercial model, like Rishi alluded to. What does that mean? When we were smaller, like 100 million revenue, you simply bought Clearwater. You know, if you signed up, you got the platform. Thank you. Use what you want. And that's obviously not the best commercial model. So now we have a model which says you buy the accounting and regulatory reporting. But if you want compliance, there's a separate package for that. If you want to buy risk, it sort of is broken up into modules and made specific for a given market. That allows for about 3.5%-4% of growth. Because in that also is price increase.
I think it's totally reasonable for us to say, hey, we're going to increase your price 4% year on year on account of inflation. So the commercial model should give you 3.5%-4%. That's point number one. The second thing is really building product. What else do clients need, which is very adjacent to what you got? So we started to build LPx. We built MLx. We built Prism. And frankly, we built four more. Those failed. But LPx and MLx and Prism now contribute to a real chunk of revenue growth. The third thing which we have is asset growth. So we do still price on assets. And so if you go to a large asset manager, you get one desk. You can get the next desk and the next desk and the next desk. And that is what we call asset growth, right?
So you put all those pieces together, and that's how you sort of build your NRR on a sustained basis. New products, a smarter commercial model, GRR, additional products, right? We do also have an AUM tailwind. So because our pricing is somewhat influenced by AUM, if we look at assets, they just grow 5% on a yearly basis. And you'll get 2%, 3% of growth just on revenue from the AUM tailwind. Now, that's something we don't usually factor in, but some years you will see us grow at 23%, some at 20%. That's a little bit affected by what the AUM tailwind is. And we don't control it. We don't model for it. But if interest rates go down, the price of assets go up, generally speaking. But that is the other element, which is sort of in our model for Clearwater.
Enfusion is a little bit different, but similar.
Yeah. Okay. Maybe let's jump into Enfusion. So what was the rationale behind buying Enfusion, right? You saw this asset. You said, all right, this is an asset we want to own. And how has Enfusion so far trended relative to your expectations?
Yeah, I think that, look, our TAM is not hard to figure out. You can go to NAIC. You can take down a list of every company. You can look at our take rate. You know the AUM. You can figure out our TAM here. And our TAM is about $23 billion, give or take a little bit. I don't care if it's $24 billion, $22 billion, something up there. But we are expected to be mid-900s next year. And the point of that is, at that scale, you have to start competing with and replacing anybody and everybody in the world. And some of the largest players, the largest players in Europe, for example, they sell the whole system front to back. The U.S. is different. Some people will buy a front office from someone, middle office, and back office. You can sort of buy pieces.
But in Europe and increasingly in the U.S., people want one full solution. Clearwater's ability to compete for that was not high because we never had a front office. They also want a risk system because alternative assets are growing. So if we think about what's growing, it's alternative assets. So if we don't have a risk system and you don't have a front office, then your ability to compete for those larger deals goes away. We announced we could be , I think, in, I think, second quarter of this year. And the reason we announced that was that was a full replacement with Enfusion, Clearwater, and Beacon all together replacing an end-to-end system. And that, I think, is going to be important as we sort of continue to grow. It's not a 2026 issue, but 2027, 2028.
If you can't get those very large deals across the world, I think you sort of tie your hands behind your back. Now, I also think that our contention was that Enfusion is poorly run, but they have a great platform and they have a really, really good team. And so could we do it better? And to some degree, I think we have proven that. I think when we bought Enfusion, people said the financial profile is much weaker than ours. And the question was, why is it weaker? And people said gross margin was in the high 60s. And they said, you know, if you look at EBITDA is lower, cash flow generation is lower, and revenue growth is lower. Those are the four elements.
And I think what we were able to show was that we got synergies of $20 million and the revenue was $200 million, $204 million. So right away, the 20% EBITDA became 30% EBITDA right away, literally at close. So we were able to improve the profile of that very meaningfully. I think we also said that the gross margin has improved meaningfully such that the integrated business gross margin is now 78.5%, right? Which is super impressive. Converts to EBITDA really nicely, converts to cash really well. So I think we have fixed the other three elements. And what's left is revenue, which we told people will take two years to get to. But at that point, though, it would be awesome financially, but strategically it's much more important that we can sell all pieces. But even financially, I think it's very great.
I know many of your shareholders, you know, we diluted 15%. Revenue grew 77%. EBITDA grew 84%. Some of it is organic. But what's wrong with the story? We continue to generate really solid cash. We are paying back the debt. If we diluted 15% for a 77% growth, let's assume 20% of that was organic. 57% growth comes from this inorganic vehicle. I think it's a really good deal for shareholders. So I feel strategically it was super important, but also financially from an investor point of view, this should be a really, really good story. But obviously, the current stock price doesn't show it. So I'm just annoyed. Okay. Sorry. I didn't mean to say that. You know, I've said that in private settings. Like, how are you doing? I'm like, I'm just annoyed. I'm like, why are you annoyed?
I started going to my lecture about 331 million, 300 shares. What the hell? People just give me something about generative AI or something.
Definitely want to get into AI in a minute because I know that's been a huge passion of yours since day one, and there's a lot of opportunity here. But I want to stick with Enfusion, kind of the potential to accelerate the growth, and I remember at the analyst day, you kind of laid out the building blocks of how do we get from around 12% Enfusion growth today back to that 20% watermark, and understanding that takes place not immediately. That takes a little bit of time. Can you walk us through the building blocks to get there and your confidence in those building blocks?
Yeah. Look, I think the first thing you've got to do is you've got to improve gross revenue retention. And that means you have to lower churn. There's no complicated science. You have to do that. The benefit of that is every percentage point of less churn you have in that business, that adds to the growth by 1%. It's like quite simple. The other massive benefit is if you can delight clients, there'll be lower churn and your sales should go up because they'll refer you to others. So our priority one was, which we announced as soon as we closed, was we are merging the operations team to try and improve churn and improve customer delight. And it is delight. It's not customer satisfaction. If you're satisfied with the restaurant, you don't sort of refer it. If you're delighted with the restaurant, yep, then you refer it.
So getting customers to be delighted, I think, is sort of strategy number one. Really original idea here. But yeah, we feel that is strategy number one. Number two is commercial model. I think the commercial model is like Clearwater's was about three years back. And so we have that playbook. We're implementing it right now. We expect to launch it in the February timeframe. It's launched already. We expect to be able to roll it out in February. We'll start with the new clients and then go to all the other clients. And so over 2026, you'll see it roll out, have some impact. In 2027, you should see it in full glory. So I think there's 3.5%-4% points of improvement there.
I think we've offered, because many hedge funds want it, to do the data ingestion like we do for Clearwater and do the aggregation and do all of that work for them, and we now sell that as an add-on. It's doing really well this year, and we expect that booking to double next year, and that could add 2%-3% of growth sort of on an ancillary basis. What technology would you use? Helios. We already built it for ourselves and Clearwater. We're just going to use the same technology for data ingestion and for aggregation, then what we do really well at Clearwater is client reporting. It doesn't take a genius to figure out. Most hedge funds don't want to do reporting. They'd rather have a digital portal, which clients can make their own, and so we feel that can add a point or two of growth.
And then finally, one of the biggest elements where they spend money is risk. And so being able to take our risk assets and launch risk for hedge funds is not trivial, but it's not a difficult exercise because the model's all there for municipal bonds, for equities, for CLOs and derivatives and bank loans and private credit and private debt. All of that exists. So if we can take that and also sell to clients, so we feel like there's 12% improved churn, that'll get you something. Then you go do the commercial model fix, that'll get you something. Sell risk, that'll get you something. Manage services. And then finish it off with going out and doing client reporting. So again, our strategy is many irons in the fire.
It's not always precisely the way I said it, but if you did all of it, you would get to that number and better, and if you get three or four of them to click, that'll be great, and so we think there is a systematic way to get there. We feel like we did it. Clearwater did it to NRR from the low 100s to 115, 116 last year in, so we think there's a path, and so we feel really confident about our ability to deliver that.
Yeah. Yeah. Absolutely. No, that makes a lot of sense, and maybe I want to ask about integration timeline. So at the conference in Boise, which by the way, everyone should be going to that. Not enough of us on this side of the equation go there. It's a great conference. You did lay out a timeline, right, in terms of both interoperability and then full integration. Can you walk us through what that timeline looks like? Because I think a lot of people may have missed that. And what gives you confidence that you can achieve that?
Yeah. I think that they sort of laid out a really simple three-phase plan, and why was that? I said, look, phase one, can everybody just go back to doing what the hell they were doing before these acquisitions? There's nothing wrong with the Clearwater story. Grow it 20%, improve gross margin 50 basis points, improve EBITDA 200 basis points. What the hell's wrong with that? Let's just go back to doing that, and Enfusion, can you just go back to doing what you were doing, but add these five elements, which should help you sell more to current clients? How hard a thing is that? So let's go back and try and do that. So initially, the idea is phase one, just go back and do what you were doing and sort of get growth in that way. All these platforms at this point talk to each other.
And so we made the investments in R&D so that each one of these platforms talk to each other. What you'll see in phase two, which is really, I expect, and we have set Q1, Q2 of next year, is actually natively integrating these pieces, which means that if you have a position data in Clearwater, you can use that, suck it into Beacon and do the risk analysis and push it back for reporting purposes. Same thing with risk. So really, that is sort of phase two where it's natively integrated. Phase three, which is supposed to be launched in 2027 Q1, is having a single security master for the entire investment management layer. So when you think about this right now, front office systems have their own description about a security. Middle office has their own description about a security.
And the back office and the risk have their own description about a security. That same security. And that's why there's so much money spent on Wall Street on reconciliation. Because front office systems don't talk to middle office systems, don't talk to back office systems. What we are building is a single security master borrowing from Enfusion, Clearwater, and Beacon, pulling it together and having a singular security master, a singular data ingestion system where all of those systems have to stop doing it. The benefit of that is really, really massive. Why is that? Forget the efficiency and the cost. Obviously, that becomes massively better. But from your client perspective, right now, Clearwater waits till the end of day and then sucks in all the trades and does the accounting, then sucks in all the trade and does the risk reporting.
But if all of it was a single security master, the moment you do a trade, it'll immediately show up in your middle office. There's no translation needed. It'll immediately go to your risk systems. And if you're trading derivatives and you're building a portfolio, it could give you a portfolio refresh of your entire system on the fly. It'll completely change people's ability to do transactions in terms of efficiency. Think about how many times you see a report which you don't agree with because you think somebody else in the company is very wrong. It'll just go away because it's a single security master. It'll be a single data ingestion system. So I feel like it is transformative. And I've got to say, Rishi, that's why we were building it.
We sort of sit back now and say, the biggest value of that will be all the data is in one database. All the operations we are talking about doing, all the calculations we do, is in one database. Can we use generative AI to massively learn from all this data being in one spot, and to give you a sense for it, every calculation we have done for the last 15 years is in that database on one platform. Every data reconciliation we have done, all of our 200 people who do it every day, all of their work is on that one platform, and so the agent's ability to learn is massively higher, which is why I'm a little bit embarrassed with our gross margin in Q3. I think we had laid out in August, September, pardon me, that our long-term margin expectation is 82%.
The core business delivered 82% in Q3. I mean, the power of agents to learn and bring it to fruition is extraordinary, I think. We have it because our agents can learn from all of that data as in one spot, I think.
Yeah. Absolutely. Maybe let's jump into AI. Look, you're probably one of the few companies, at least in my coverage universe, that has actually quantified the impact of AI on gross margins in a positive way. Can you walk us through how you've been able to get gross margin leverage so far from AI? And as we look out over the next couple of years, right, what do you think you might be able to get even more incremental leverage, even as you have to bear certain costs of inferencing on the product side? We'll get on the product side in a second.
Yeah. Yeah. Look, I think I don't know how much software you all, obviously, you all cover software. But in our industry, which is investment accounting, analytics, middle office, risk, and all that, it is still the vast majority of is single user, single system. What that means is if you buy an accounting system in Germany from any of the large providers, you get that installation. What does that mean? I'm the only one who's using that instance of the software. Rishi buys the same software. He builds his own system. Everybody has their own platform, even though it's on the cloud, right? Single user, single. The provider here in this case doesn't have access to my database or my portfolio or to Rishi's portfolio or to Camille's portfolio or anyone's portfolio. So how does it learn? I don't know.
In Clearwater's case, all the information about everything we do is in a single portfolio, is in a single database. We are the ones who are doing it. Clients are not touching the software or the data or the accounting till it gets to the reporting layer. So our ability to learn, I think, is fully transformative. How does it work in real life? So let's say I'm a reconciler for Clearwater. What happens is data comes in from 4,000 sources every day, and there's something wrong. Something doesn't match. 93% of the time, the platform itself will fix that. But sometimes it doesn't work, and people have to go at it. And we have 250 some number of people who look at this every day and fix it. So like what?
If somebody says a municipal bond paid 72 cents and another source says it paid 73 cents, there's no way to reconcile this. Computers can't do it. Humans have to do it, right? And what the system does today is that I used to have to fix it. Let's assume I took 45 minutes to do it. Now what happens is the generative AI tool says, here is what I think the fix is. And it's based on 10 years. Every time there was a difference between these two sources, this is what the likely reason is. I have two choices. I'll either say, yep, that makes sense, accept. Or I would say, I don't like it. I'm going to do my own math. And you do your own math for 45 minutes, the same 45 minutes, and then you have a choice.
You either accept it or you say, no, no, no, you were wrong. This is the right way to do it. Either way, the machine learned. But if I accept the answer of the agent two times, I'm 20% more productive. If I accept it three times, I'm 30% more productive. Each one I accept, I'm more and more efficient. And that's how we're getting gross margin improvement because we're not hiring more people because people are just becoming more and more productive as they learn, right? And I think that's where you will see I was wrong about 82% long term because I thought we'd take longer to learn. And these platforms are really powerful because they have all of this data in one spot and they can learn. And human beings aren't that good, by the way.
Half the time, human beings, sorry, not selling, that's not the right term, but you hire new people, they don't learn enough of how to price their derivatives. They'll go out and do it. And half the time, they make mistakes because they're unhappy about something. Yeah, machines learn. They unidimensionally, they learn. So I feel like using generative AI for that is really helpful. One other thing on the client side, clients come in, and every time you see a report, I'm sure you say, oh, that doesn't look right. What do you do? You go ask somebody. And in this case, you would send it to Clearwater. Right now, what happens is the agent provides an answer. And I would tell you that 80% of the time you look at an answer and you can say, I get it. Oh, that's how it came. Fine.
You can accept it. But if you don't accept it, you can send it to Clearwater just like you did before. But the point is the number of times you accept it will continue to rise. And every time you accept it, my gross margin grows up just a little bit. And so the point here is you can do it on both sides. And over time, what I'm hoping you'll do is, you know what? Clearwater platform, every time you don't understand something, they answer it instantaneously because they have this thing. So hopefully, more and more people start to buy it because of that reason. But that hasn't happened yet. But can we, in the meantime, show much higher gross margin and a constantly improving gross margin? Absolutely. And last point on investment, there is no investment.
If you invest $20 million and you get a 300 basis point improvement in gross margin in that year, the return on that is more than the $20 million, and the return perseveres every year after that, so I don't, when we sit on a plan, we don't think of it like that. We said, what are we going to produce?
Yeah. Yeah. No, that makes a lot of sense. Very thorough. I want to ask now about international, right? So you had a big beachhead deal in Europe, I think, two years ago. And that's allowed you to be really successful in Europe. You called out a big APAC deal that was really recent. Maybe can you walk us through kind of the growth vectors you have outside of the U.S. and how we should be thinking about the kind of growth patterns that are going forward?
Yeah, I think that what we sell in the U.S., just think, we already have U.S., at least we have U.S. GAAP. Nobody exactly follows it because everyone, no, I shouldn't say that. People follow the U.S. GAAP, but they have their own nuance to each one of these items. Everybody has their own rules about how they manage things, how they account for it, right? You go to Europe, the problem is a lot worse. They got French GAAP and German GAAP, Italian GAAP. Every GAAP is different. And the point is our product therefore is much more relevant, not much more. It's more relevant there because to get a comprehensive view, you would have to take five different systems and merge it together. With Clearwater, single instance multi-tenancy, you can get all of it. When you get to Asia, the reverse is true.
There's nothing similar about a Japanese GAAP and an Indian GAAP. And yet, as an investor investing in both of these countries, how do you get a comprehensive view to understand risk and what you should trade and what the current value is? Super hard, so I think our platforms are more relevant and more powerful when you look at Europe and you look at Asia. And so look, I feel really strongly that the platform is very comprehensive, but it'll never be 100% only because new asset classes are developed every day. New regulations come up every day. New compliance standards come up every day. So it's constantly learning. And to that degree, it is our beachhead is that it's our moat.
Yeah. Yeah. Absolutely. Looks like we're at a time. So I think that's a great place to jump off. Sandeep, thanks so much as always for being here. Thank you, everyone, for being here.
Thank you.