Our first question comes from the line of Manav Patnaik with Barclays.

I just had a question on the DLP program that you said is under review by the GSEs. I believe one of them had already signed off. So just trying to appreciate why or what the next steps for the other one are? And then with respect to that, you talked about the signed agreements with 60% -- resellers representing 60% of the U.S. mortgage volumes. Is that -- are they also interested specifically in the performance fee model? Just looking for that incremental color.

Yes. So with respect to the DLP, not a lot of new news. We're literally waiting on certification from one of the GSEs so that we can go live. The operational stuff is all set up, ready to go. And as we've pointed out, we have agreements already covering 60% of the reseller volume. We are also very close to signing 2 additional resellers. So that's -- we're literally waiting for their okay. The -- in terms of interest from the resellers in the performance model, yes, there is significant interest in the performance model. And we and they are anxious to get it out into the marketplace. We think it will be very beneficial.

Okay. And then just on the release of the 10T data, just curious, I don't know if you guys have had any feedback worth sharing from your -- from the industry or your clients in terms of that versus the other alternatives out there.

Well, they now have an opportunity to test the 10T data as provided by the GSEs. And although we'd like to see 10T approved sooner rather than later, there is kind of a time, an elapsed time between releasing the data and doing the analysis and then approving 10T. There's a lot of interest. And as you know from my earlier remarks today, third parties like Milliman and others have done the analysis and 10T is more predictive than Vantage, pure and simple. And so there's a lot of market demand for it.

And our next question comes from the line of Jason Haas with Wells Fargo.

As you know, there's now some data out there that shows VantageScore gaining some share in the MBS market. I'm curious from the lenders that are in that pilot program, are you seeing them drop FICO pulls at any point in the mortgage process? Or are they simply pulling the same number of FICO scores and just also adding in VantageScore? And then maybe when it's time to submit that loan to the GSE, they're submitting with a VantageScore?

Jason, that's a very good question. So just a bit of -- to take a step back, you know that we're not crazy about lenders choice as a policy. We think it's bad policy because it encourages gaming. And that's the primary reason that one might consider buying a VantageScore is to try to deliver to a consumer a better rate when some of the time the VantageScore produces -- is higher than the FICO Score. So we knew that gaming was going to happen and that's what we're seeing is that it is happening. And to your specific question, are we seeing volume loss? No, we are not, which suggests that they're pulling both scores. And you have to pull both scores if you're going to try to do the gaming. You need both scores to figure out which one is going to deliver a bigger benefit to the consumer. So there's not a lot of surprise in what we're seeing. You got a 2-score system, which has its own kind of structural problems. You have the gaming that you would expect. You have the most sophisticated biggest players experimenting with it. But I guess the thing to remember is you need to pull both a FICO Score and a VantageScore if you're going to do the gaming. A VantageScore by itself doesn't really get you there. We -- and I guess just to round it out, we're not seeing volume loss. So I don't think it's instead of, the VantageScore is additive to the market, makes the market bigger.

Got it. That's very helpful. That all makes sense. Can you also just comment on the mortgage origination revenue growth decelerated on a year-over-year basis. It was also down quarter-over-quarter. I'm sure part of it was just mortgage volumes being a bit softer in 3Q. I assume that was the case. But any other comments on what's driving that? And I don't know if you could give us any sort of framework for how to think about 4Q, just to level set expectations here.

Yes, that's a great question. So if you look on a year-over-year basis, we were up, we said low single-digit volumes, which I think is pretty much in line. If you take all the bureaus together and average them, that's probably what you get or take them as a total universe. So I think what we're -- we're seeing similar volumes to what the bureaus are seeing. There definitely was a slowdown as the bureaus talked about as well. As rates tick up, the volumes do slow down. So that's what happened there on a year-over-year basis and that's also what caused the quarter-over-quarter. So again, it's really just about the slowdown in the mortgage market as rates crept up.

And our next question comes from the line of Jeff Meuler with Baird.

Yes. A follow-up on that last question. Maybe this is just asking -- rehashing old news. But the 127% growth that you had last quarter, it was always hard to get there based upon market volume and based upon a $10 versus $4.95 price. So was there anything else that was inflating that number? Or is there anything like timing -- I know it's an annual calendar year rate card but anything on when timing of the pricing actually took effect last year versus this year or anything like that?

Yes. I think there's some of the timing piece. And I think there's some of how maybe our quarter cutoff versus what the bureau quarter cutoffs are. And when you have markets that are moving very rapidly when rates are moving, you can have -- every week can have different kind of volumes. I think you might see some of that. We don't have access to what the bureaus -- the data, the underlying data that they report. So I can't tell you. All I can tell you is the scores that were pulled by us on a year-over-year basis and what we charge for them. So that's really it. I think if you -- the shorter you parse apart the segments, if you get down to week by week, it's probably even less or month by month, there's less correlation and especially when the markets are moving like they are. But over time, it all works out but there are some quarters here and there where you see anomalies like last quarter.

And our next question comes from the line of Simon Clinch with Rothschild and Redburn.

Was wondering, Will, if you could talk -- you mentioned that, of course, you're not seeing any volume loss. I'm just wondering, given you are pretty much all of the market, how easy is it for you to actually monitor any sort of evidence of volume loss in that regard?

It's not that easy. It's not -- I mean, honestly, it's not that easy. But relative to our forecast and expectations, we're not seeing it.

And I think you can see that in our numbers for the full quarter that we just did. I mean the volumes we saw are pretty much in line with what the volumes that the bureaus reported on a year-over-year basis. So we're pretty confident in that. But we triangulate on a lot of different factors. We see mortgage, public mortgage data. So we try to triangulate as much as we can. I mean I think we don't have exact numbers but we're pretty confident that what we're seeing is representative of what's happening.

Okay. Great. And just following up on the other segments, the auto, card and personal loans. Could you just give us a bit of color as to how things are progressing there in terms of your sort of pricing initiatives, testing, elasticity of the market, et cetera?

Yes. Well, it's obviously early in the year for us to be suggesting where the pricing will go for next year. But as you know, we are constantly exploring where and how to get revenue growth. And each year, we get a little bit better at getting more precise, more surgical, identifying the pockets where it creates the least amount of pain, has the least amount of market reaction. And so we're obviously working our way through those things this year as we always do. No final decisions have been made about other -- about any of the sectors, frankly, about mortgage or auto or card or personal. I mean we're -- it's early days. So yes, lots of analysis going on. Our pricing team, our strategy team are working on all these things but really nothing to share with you right now.

Our next question comes from the line of Surinder Thind with Jefferies.

Just switching gears over to platform. Can you maybe talk about just the next-generation product, the next-generation platform? And just the client conversations you're having in terms of the uptake. Obviously, it seems like there's some clients that are moving from non-platform to platform as well. And just any color on the dynamics. Is that something that we should just expect to accelerate as more features, functionality and go forward? Or how should we think about that?

Yes, that's exactly what you should expect is for it to continue to accelerate. We've got tremendous interest and tremendous uptake. And you can see in our numbers, tremendous platform growth. We do continue to release into the market new features and functionality that just increase the use cases. And with that, the utility of the platform because it's all -- as I said earlier, it's reinforcing the more use cases you have on the platform, the more value you get out of it, the lower cost it is to bring on additional functionality. And so our whole land and expand strategy is built around that. It's like get started and then work with our customers to get the full benefit out of it. We absolutely see more growth.

I guess just following on from that is, would you -- now that platform is larger than non-platform and you're also seeing some volatility from the licensing component. Would you start considering like end-of-life certain products at this point? Or how do we think about that? Understanding the strategy here. So...

Yes, absolutely. So we've talked in the past about our end of life and our migration strategy. And at that -- in quarters past, what we've said is that we're not forcing migration. We're not cannibalizing legacy to achieve growth in the new platform. And that remains true. However, there's a tremendous benefit to FICO in simplifying our product set, our catalog. And we have some products that are old in the tooth that really should be end of life. And so we're finally getting around to doing it. We have the capacity to move our customers on those older products to newer, better products that are going to wind up, well, to the platform and to the functionality they get with the platform and they'll be able to get more functionality at a lower cost. So long answer to your question but yes, we are -- we have an active end-of-life strategy that we're working through.

Our next question comes from the line of Ashish Sabadra with RBC Capital Markets.

Just wanted to follow up on that earlier question around the VantageScore market share in conforming loans. If we look at some of the bigger players there, particularly UWM and Rocket, the market share for VantageScore seems to be closer to 20%. I understand it's both pulls initially but like are they still using both pulls when they close the loan? And also like the pace of adoption, like how does that compare to what your expectations were? And how do we -- like any thoughts around what FICO can do to regain its -- the market share?

Yes. So I think, first of all, we have a week of numbers that have a 2 handle on them, okay? So that -- so it will take some time for us to really digest and see what the steady state is. That said, we don't anticipate a lot more than the 20s or we don't anticipate more than the 20s for Vantage share because, as I said earlier, it's really tied to the gaming. And when you do the math on what percent of the time a consumer is advantaged by using a higher VantageScore than a FICO Score, that number, which in our mind, represents a theoretical maximum is in the 20s. So I think we'll just have to wait and see how things play out. I think it is a reality that Vantage is selling scores and is -- or sending them along with FICO Scores, I should say. And gaming is a reality and we're going to have gaming. I mean the rules of the road, the environment that we're operating in is a gaming environment. The FHFA has put it in place. The GSEs have accepted it. And so that's the world we have. And so I think we can -- we should all do our own math but I can tell you our math says that the number is in the 20s.

That's very helpful color. And from a pricing strategy perspective, is there things that FICO can implement to help influence that going forward? Or does that change your pricing strategy for '27 or going forward?

Well, of course, there are many, many things that FICO can do and we're not in a position to share them on this call. But we -- of course, we have a lot of work going on in the strategy for everything we do, for how we monetize our IP, for how we set our prices, for what features and functionality we put in the products. All of that is on the table. And so we have lots of degrees of freedom in how to respond.

And our next question comes from the line of Faiza Alwy with Deutsche Bank.

First, I just wanted to ask about the revenue guidance raise because it seems like it was pretty significant relative to what it should be this quarter. And so I'm wondering if there's anything out of the ordinary, maybe you're expecting some licensing revenues in the fourth quarter or maybe you had previously assumed the direct program would come through and that might cause a lag in revenue. So just a bit more color on what led to the revenue guidance raise.

Yes. I think it's more of the latter. I mean we had planned that if the DLP was to go live with the performance piece that we would push some revenues out, more revenues out. So as the year goes along, that hasn't happened yet. So there's that piece. And I think, frankly, actually, the mortgage market has been better than what we had originally guided to and our volumes are better than what we had originally guided to. We -- there were a lot of concerns about rates. The rates haven't -- they haven't gone down but they haven't gotten worse. So the mortgage markets held up fairly decently. And for a period of time, it actually popped a little bit but now it's back to kind of low single-digit growth. But that's really where it's coming from.

Okay. Understood. And then just to follow up on the Vantage pilot. I guess with 10T sort of in the picture now, how are you thinking things will evolve? Are you -- is there -- are you anticipating a separate pilot for 10T? Would lenders pull kind of all 3 scores for gaming purposes? Just curious on how you expect things to evolve from here.

Yes, that's a great point, Faiza. It really is because if we have a 3 score market, then there's an incentive to pull all 3 scores and see which one produces the best outcome for the consumer. How it will actually shake out? It's hard to say but it's hard to imagine retiring classic and not anytime soon. And so most likely, 10T will come into the mix and you'll have classic and Vantage and 10T. And you're absolutely right. I mean that's kind of a shopping -- a score shopping environment.

Sorry, just to follow up on that. Would you charge extra -- would 10T be an additional price? Or would it -- would you offer it for free alongside FICO classic?

As you know, today, we have a tremendous amount of volume in our 10T pilot program, right, 55%, I said earlier. So 10T is increasingly being used. 10T is the most predictive score. If you care about credit risk, 10T is the answer. If you care about gaming, 10T might be the answer some of the time, classic might be the answer some of the time and Vantage might be the answer some of the time. And that's a gaming world. In terms of how we price it and how we bundle it, today, as you know, we bundle 10T with classic. And if you pay for classic, you get 10T free. And I would imagine we would continue to do something like that because we want to encourage adoption. That said, we have not made final pricing decisions. Everything is on the table. One could easily imagine a revenue jump at FICO if we were to charge separately for the 2. But that's not the current plan. The current plan is more of a continuing with the bundle as we have it today.

And our next question comes from the line of Alexander Hess with JPMorgan.

I want to maybe ask about -- we'll start with the 10T question, of course. There have been a bunch of analyses out there -- I'm sorry, of classic FICO versus Vantage and whether or not there is a meaningful coupon differential at -- obviously, when you look at one VantageScore at a certain given number, say, at an 800, of course, there's a differential versus a FICO at 800 for a mortgage loan, given the presumably shadow LLPA grid that exists for them. But then when you correct for those sorts of factors, it seems that a lot of the spread goes away, at least by some of our team's analysis. How do you think about the mortgage level and then the pool level spreads for firms that are using the competitor's score?

Yes, that's a great question. And frankly, I'd refer you back to your own MBS traders to really get the insights there. But as a matter of theory, there's a lot less history with Vantage. Vantage has never been through a down cycle. And so there's more uncertainty around it. And so investors, if they're rational, should analyze paper that's not as well understood. So that's the theory behind the 30 basis points estimate that we've kind of talked about in the past. How it will really shake out? It's hard to say. I mean it really kind of depends on the rationality of the market. Now today, Vantage is such a small part of the market, it's not clear that any of it is really providing real signal. As you know, what has been securitized with Vantage is largely mixed into much bigger FICO pools. So I would just encourage you to talk to your traders and see what they say. But they are rational. And so in the -- as they dig into this, I think we ought to see the differential appear.

And then maybe thinking about the monetization across the Scores business holistically, Will, obviously, there's a price point for the FICO and mortgage applications. But there are entire pools of the market where I think you guys would say you're pretty undermonetized and even underpenetrated. Rental comes to my mind but maybe something else comes to your mind, overseas perhaps. How are you thinking about maybe -- at some point, the mortgage market does reach an equilibrium on FICO pricing. I don't know when it is. But how are you thinking about the next leg or future legs of monetization for Scores holistically and what might that look like?

Yes. I very much appreciate that question because too often, people think about our Scores IP as being single threaded through mortgage. And yes, obviously, we have a lot of mortgage concentration but we do have a lot of opportunity in other verticals and we have a lot of opportunity with new scores that rely on different data sets that can score new populations and provide new avenues for us to make money. We're working on all those things. I mean, I mentioned UltraFICO and our partnership with Plaid. UltraFICO is a next-generation score. It's a consumer permissioned score that captures everything that you get today in a FICO Score, the credit file data and everything, all the caloric value that you get out of the credit file but it augments it with cash flow data. And when you do that, you get much better insights. And so that's very much a next-generation score. It's not widely adopted. We're just getting going now and you know that the adoption curve for new scores, it can take 4 years for a new score to really get established. And so we are very much driving ahead with UltraFICO 2. There's an UltraFICO 3 that's being readied that has some additional functionality. And so that's one example. But they're -- we're -- there's also FICO 11 in the lab. So we're constantly innovating, constantly trying to figure out how are we going to get a little more signal. There really are limits to how much more signal you can derive from the credit file. I mean that's -- it's been pretty well picked over. And so we tend to focus on other data sets to get more predictive power.

And our next question comes from the line of Kyle Peterson with Needham.

I wanted to start off on the DLP program. I know we've -- gaming has come up several times on this call already. But I guess just how have those conversations with potential lenders gone? I guess, do you think there might be any slowdown in adoption from that by people that are gaming, given at least right now, if they buy from the bureaus, they can get the Vantage for free and save the buck. So I guess just any thought on if gaming would impact the adoption curve on DLP would be good.

Yes. Good question. We don't think so. We think that the benefits of the performance model with DLP are pretty significant from a cost standpoint. What it really does is encourages more customer acquisition, more speculative effort to identify potential borrowers. It broadens the market, more access. So it does all those kinds of things, which are very desirable to the big lenders. So we see continued interest in it. And I don't think the gaming is going to put any kind of a damper in it.

Okay. That's helpful. And then I guess just a quick follow-up on capital return. Obviously, good to see the ASR big signal. And I think you guys have said a couple of times now that kind of prioritizing leverage reduction in the near term. I guess, should we interpret that as that you'll likely be more or less totally out of the market for the next couple of quarters? Or -- any color there on how much you guys want to prioritize and for how long you guys will be prioritizing debt versus incremental buybacks?

Yes. I mean we drive -- we have a lot of free cash flow. So we delever pretty quickly. So we'll see. We'll update you next quarter on what it looks like then and we'll determine where we're at. I mean it's probably not likely we're going to buy any shares -- additional shares beyond what's already in the ASR this quarter. But then when we give our next quarter results, we'll talk about where we are there because we do delever pretty quickly. So this is an acceleration certainly of our buyback but you've also seen our EBITDA and our cash generation grow pretty dramatically. So we're delevering even quicker than we normally did.

Our next question comes from the line of George Tong with Goldman Sachs.

You're on the cusp of having your 10T score go live pending DLP approval. Can you share feedback you've received from lenders on 10T's pricing, including the funding fee component?

Well, we have explored the performance model and the funding fee component with lenders. And as you can imagine, there's -- some like it and some don't like it so much. And so we know that there's appetite for it. I think that is one of the driving forces behind DLP adoption, will be the opportunity to get the funding fee model and performance model. So -- but is it for everyone? No, it's not -- I mean not everyone will love it, particularly if you hardly pull -- if you don't pull many scores per closed loan, it's less attractive. I mean that kind of goes without saying.

And our next question comes from the line of Ryan Griffin with BMO Capital Markets.

Just on the software business, I was wondering if you could talk about the retention rate dynamics. Is that the best way to think about that just land and expand? And then on the ACV bookings, are you still expecting those to accelerate in 2 half versus first half?

So in reverse order, bookings, yes, we see continued acceleration. But the first part of your question is, yes, land and expand is very much our strategy. You see it in the DBNRR. We are doing a little more migration now than we were, say, 3 or 4 quarters ago. Some of our CCS business is migrating. So there's a bit more migration and that's why you saw the legacy retention rate go down. But on balance, we're pretty happy with the way it's all playing out. I mean, 146% (sic) [ 148% ] DBNRR on the platform business and that being now the bigger half of our business, that's pretty good. We have very, very low churn.

And for the follow-up, just on the recent trigger loan legislation, we heard that shook things up between the pre-qual market and the hard inquiry market. I was just wondering to what extent that impacted the mortgage volume side of your business?

We haven't seen a lot.

There's a little.

I mean, we noticed a bit but I wouldn't say anything dramatic.

And our next question comes from the line of Owen Lau with Clear Street.

Going back to software, you mentioned booking was strong, platform ARR growth accelerated but non-platform was weak. Should we expect these kind of like divergence continue? I mean, because the delta was pretty high. I'm just thinking about how to think about this going forward.

Yes. I mean the short answer is yes. I mean we had held off on migrating legacy to platform for many, many quarters, partly because we didn't have the capacity to handle it on the new platform. We are so busy bringing on new customers. But again, kind of going back to one of the earlier questions about end of life, there really are some legacy offerings that ought to be wound down over the next several years and we're actively doing it. And we're going to give our customers a better alternative on the platform. So yes, I would say you will continue to see the divergence that you're seeing right now. It's -- I mean we don't think it's a bad thing. As long as on balance, we're going up, I think we're pretty happy. We will have continued migrations. We have a plan for migrations. We have a team that makes sure that they go smoothly. But all that said, I don't want you to get the impression that the growth in platform is being driven by us cannibalizing the legacy because the truth is, our pipeline is growing. It's expanding. It's growing. We -- the land and expand strategy works. We have a lot of net -- we have a lot of new. But unlike a year ago, 2 years ago, some amount of the platform growth is coming from migration.

Got it. That's helpful. And then quickly on the VantageScore pilot program. I'm wondering if 10T is actually dependent on the VantageScore pilot or there's a path to run both in parallel? I mean if 10T is further delayed, how would that impact the implementation of other FICO Score you just mentioned like UltraFICO 2 or FICO 11?

Well, so obviously, the FHFA and the GSEs are going to have to decide when they approve 10T. I mean it's approved but when they accept it, right? That's on them and that's their schedule. That said, we know that 10T is absolutely the best score in the world for measuring credit default risk. And you're seeing it in the nonconforming market. You're seeing tremendous adoption of 10T in the nonconforming market. And so to the extent people care about credit default, 10T is the answer. When the GSEs and the agency side decide that they want to mix that in is up to them. I think that they are actively working on it. I think they want it out there but it's going to take a certain amount of time for them to get comfortable. They just released the data, so it's going to take them a little while.

And our next question comes from the line of Scott Wurtzel with Wolfe Research.

Just one for me on the direct license program and in terms of the remaining resellers that are still out there to be signed up. Just wondering if you can give us kind of an update on where the process stands with those 2 and maybe what's unique about them that's maybe taking a little bit longer.

We have 2 big ones that are signed. We have 2 big ones that are almost signed, very, very close and then we have the tail. So that's where we stand. We're pretty close.

And our next question comes from the line of Sean Kennedy with Mizuho.

On software, I was wondering how impactful partnerships like Accenture are for platform growth. Does it significantly help FICO Platform's customer reach? And is there any particular type of customer that you're targeting in terms of geography or size?

Sean, thank you for that question. We have talked for many years about FICO's challenge in distribution. We're IP-rich and distribution poor and we've always had so much more IP than we can sell with our direct sales force. And that's as true today as -- it's not quite as true today as it was in years past but it's still true. And so for several years now, we've been very focused on how do we partner with SIs who can take our IP to market with us, for us, different approaches. And we are super pleased to now be in this significant strategic partnership with one of the top SIs in the world, where they're going to be going to market with us, with our IP and their capabilities. They have relationships that we don't have. We have relationships and can send work their way. So it's really truly a strategic partnership for both of us. We're super excited. I think it's the beginning of seeing the indirect side of our business grow. I think you're going to see increasingly, we'll wind up getting -- monetizing our IP through partners and not just through our direct sales force.

Got it. That's great to hear. And then the platform retention rate has really inflected positively this year. Is it partially due to AI and FICO's capabilities there? Or are there certain platform products that are growing significantly faster than other ones?

Just repeat the first sentence you said, it just got muffled on our side.

Oh, sorry, I was saying is with the net retention rate of platform like in this trajectory, is it partially due to AI and FICO's capabilities there? Are there -- or are there certain platform products that are growing significantly faster than others?

So I would say yes and no. So it's not yet because of AI, although we have lots of AI coming in the platform. I'd say that the platform growth we have right now reflects the current state, the current state of the platform and the functionality and capabilities that we can bring to our customers with what we have today. I mean, they get immediate payback. They're up and running very fast and they get immediate payback from it. The AI enhancements to the platform, if you want to call them that, are coming. They're close. And will that result in an uptick in platform growth? I don't know. It's certainly going to continue the growth. There's a lot of appetite for it. We've got all kinds of great AI-driven capabilities for our customers who are on the platform.

And our next question comes from the line of Curtis Nagle with Bank of America.

Terrific. Will, maybe just a question for you. Just following up on all the commentary in terms of gaming -- potential gaming in the system. I guess anything you're seeing anecdotal, whether it be discounts placed on Vantage securitized loans, maybe that's a little hard to see because of commingling. But just anything else that you think is suggesting that's occurring in the market?

I'm not sure I know how to interpret that question. We know that there's gaming. We expected gaming. We're seeing gaming. We think there's a limit to how big gaming can get. And so I don't know what -- the anecdotal is what you see with Rocket, UWM. I mean that would be your anecdotal evidence of how much is happening and who's doing it. But it's -- I mean, it's -- it is a reality that there will be gaming. I mean the structure that's been put in place invites lenders and originators to score shop and so they will.

And our next question comes from the line of Craig Huber with Huber Research Partners.

First I wanted to ask, on the performance model, given that we're almost into August here, just talk a little bit further about just the usage of it out there, the feedback that you're getting. Where are we at on that right now, please?

Yes. So just to be clear, the performance model is to be distributed through the direct license program with the resellers. That program is not yet live. And we thought it would be live months ago but it's not yet live because it's waiting on the certification from one of the GSEs. So I can't speak to usage of the performance model because it's not available yet. It's theoretical until it gets certified. That said, are we hours away, days away, weeks away? I mean this isn't that hard a thing to do. The market wants the model. It gets us -- we're happy because it gets a lower price point out into the market, makes us more competitive. The lenders who want it really want it. The resellers want it. So I think we're in a waiting game here to get certified.

And there's really no commentary you can give like on the nonconforming part of the market for the usage of it? Is that all tied to the conforming piece, so there's not really much uptake there either, this is also a waiting pattern?

We have not offered it. Just to be clear, we have not offered it there. It's offered in -- it is to be offered through the direct license program and that is not live yet.

Okay. And my other question on the software side. I just wanted to understand this a little bit better. Obviously, your software revenue in aggregate was up 2%. Your on-premise and SaaS software up, call it, 5% year-over-year. Your costs look like we're up about 12% year-over-year, similar cost growth to the March quarter. Just what's going on there with the cost growth significantly outpacing the revenue growth?

Well, so there's a couple of pieces there. So on the revenue side, we have a lot less point-in-time revenue, significantly less point-in-time revenue. So that's essentially revenue that -- it's lumpiness. There's still a little bit of that lumpiness in the model. We have less of it than we had in the past and we'll have less going forward. So there's that piece. And then this quarter, we had a pretty significant uptick from FICO World, which is a bigger event than what we had last year. And then we had some other kind of AWS that has increased too as the SaaS piece has gone up. And we've done some investing on that side. So I mean, there's -- we've done some investing and you're going to see the margin growth probably next year, you'll start to see some growth off of that. But there is some lumpiness on the software point in time side that will give you some lumpiness in the margin.

And our next question comes from the line of Rayna Kumar with Oppenheimer.

So even with the DLP, the credit bureaus are likely to remain large customers of FICO. So I'm just wondering if you can comment on how your relationship with them has evolved over this process and where you stand now?

That's a great question. We get along great with the bureaus. They are our partners. We get a lot of revenue from them. They are our channel partners for our Scores IP. We sell into lots of other verticals with them and it's been a strong, healthy relationship. And then particularly with Experian, we have a big consumer business together. So I would say healthy and strong relationship. But at the same time, we're now competing in mortgage scores and that's not a secret. We're obviously doing it. They've been pushing Vantage for 20 years and now they're finally getting a little bit of traction in mortgage because of the Lender Choice program. And so is that a thing that stands between us and being the best of friends? Yes. But I would say the relationships are strong, healthy and we will compete in this space.

Thank you. Ladies and gentlemen, thank you for participating. This does conclude today's program and you may now disconnect.