Loading…
Loading…
Good day, and thank you for standing by. Welcome to the Second Quarter 2026 FICO Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Dave Singleton. Please go ahead, sir.
Good afternoon, and thank you for attending FICO's second quarter earnings call. I'm Dave Singleton, Vice President of Investor Relations, and I'm joined today by our CEO, Will Lansing; our CFO, Steve Weber. Today, we issued a press release that describes financial results compared to the prior year. On this call, management will also discuss results in comparison with the prior quarter to facilitate understanding of the run rate of the business.
Certain statements made in this presentation are forward-looking under the Private Securities Litigation Reform Act of 1995. Those statements involve many risks and uncertainties that could cause actual results to differ materially. Information concerning these risks and uncertainties is contained in the company's filings with the SEC, particularly in the risk factors and forward-looking statements portions of such filings. Copies are available from the SEC, from the FICO website or from our Investor Relations team. This call will also include statements regarding certain non-GAAP financial measures.
Please refer to the company's earnings release and Regulation G schedule issued today for a reconciliation of these non-GAAP financial measures to the most comparable GAAP measure. The earnings release and Regulation G schedule are available on the Investor Relations page of the company's website at fico.com or on the SEC's website at sec.gov. And a replay of this webcast will be available through April 28, 2026. We have refreshed our quarterly investor presentation with additional content, which is available on the Investor Relations section of our website. We will refer to this presentation during today's earnings announcement.
I will now turn the call over to our CEO, Will Lansing.
Thanks, Dave, and thank you, everyone, for joining us for our second quarter earnings call. We had a very strong quarter and a great start to the first half of our fiscal year. Based on our results and outlook, we are increasing our fiscal 2026 guidance. We reported Q2 revenues of $692 million, up 39% over last year, as shown on Page 5 of our investor presentation.
For the quarter, we reported $264 million in GAAP net income in the quarter, up 63% and GAAP earnings of $11.14 per share, up 69% from the prior year. We reported $297 million in non-GAAP net income, up 54% and non-GAAP earnings of $12.50 per share, up 60% from the prior year. We delivered free cash flow of $214 million in our second quarter. Over the last 4 quarters, we delivered $867 million in free cash flow, an increase of 28% over the prior fourth quarter period.
In Q2, we continued returning capital to shareholders through share repurchases, buying back $605 million or 484,000 shares at an average price of $1,251 per share. At the segment level, shown on Page 6, our second quarter score segment revenues were $475 million, up 60% versus the prior year. While B2B scores were the key driver of growth, we also experienced the sixth straight quarter of growth in B2C scores. In our Software segment, we delivered $217 million in Q2 revenues, up 7% over last year.
Results included 54% platform revenue growth and a 12% decline in non-platform revenue. Steve will provide additional revenue details later in this call. Last week, we issued a statement on our website in response to the FHFA and FHA update on credit score modernization. We applaud the FHFA and FHA initiative to get FICO Score 10T into the market in the coming months.
FICO Score 10T is the most predictive credit score for all borrowers, including first-time home borrowers. FICO Score 10T incorporates rental and utility payment history, enabling more consumers to qualify for mortgages. To support the goal of increased homeownership and bring the benefits of increased competition to the marketplace, we updated our FICO Score 10T performance model pricing in the FICO mortgage direct licensing program from $4.95 per score plus $33 funding fee to $0.99 per score plus $65 funding fee. We anticipate the release of FICO Score 10T data and the time line provided by the FHFA and GSEs.
In the last quarter, we added 11 more lenders to our FICO Score 10T early adopter program. As a reminder, through this program, FICO Score 10T is made available for free with the purchase of classic FICO. The 55 lenders in the program account for more than $495 billion in annual serviceable originations when evaluated using 2025 HMDA data and more than $1.6 trillion in eligible servicing. We're moving closer to the go-live dates of our next-generation Cash Flow UltraFICO Score with our strategic partner, Plaid and the FICO mortgage direct licensing our reseller partners.
We continue to actively work alongside participants to support testing on both initiatives. As AI adoption accelerates, we recognize the need of stakeholders to weigh the associated opportunities and risks. At FICO, we view AI as a tremendous opportunity that we've committed significant resources to for several years. In the Scores business, AI is limited by strict regulatory requirements on credit underwriting outcome explainability and model governance.
In addition, our scoring models are supported by proprietary data access, mainly with the credit bureaus and deep ecosystem integration. Across both businesses, FICO has been issued 137 AI-based patents, which include patents and blockchain technology that are helpful for traceable and explainable decision-making, the type of market-leading innovation that will be in high demand as businesses seek ways to safely deploy AI analytics in highly regulated industries. In our software business, as shown on Page 13, FICO Platform is architected from the ground up to be agentic-by-design. That foundation delivers decision grade analytics, deep domain expertise, and an enterprise platform that clients depend on for precision, consistency, explainability and trust.
These principles are nonnegotiable for our primary target market, the highly regulated financial services industry. FICO Platform is the world's leading AI decisioning platform for financial services recognized as such as a leader by Gartner, Forrester and IDC. Its agentic architecture power is a real-time, always-on customer profile engine that delivers hyperpersonalized consumer experiences where every interaction can inform and improve the next. There are over 150 clients globally using the FICO Platform across multiple connected use cases to power their customer experience, business critical operations, risk management and fraud monitoring and prevention.
FICO Platform brings together multiple functions within an enterprise in a common operating environment and enables them to operationalize AI at scale to drive real business outcomes. Financially, a substantial majority of our nearly $315 million platform segment annual recurring revenue is driven from FICO Platform. Financially, a substantial majority of our Platform segment annual recurring revenue, approaching $350 million and growing rapidly is driven by the FICO Platform, reflecting years of proven commercialization. FICO transformed 70 years of proven deep domain knowledge into validated expandable AI that powers the most consequential business decisions with that expertise embedded directly into the agents, models and guardrails that operate on the platform.
Fico Platform accelerates client innovation by providing clients with the ability to build, test, optimize and monitor decisioning across the enterprise. With FICO AI-guided operations, clients create a self-reinforcing cycle of value generation, reinvesting outcomes back into the platform by enabling additional use cases, driving further value for their businesses. FICO Platform's marketplace and FICO Assistant unlock broader capabilities that compound with scale. Every new model, agent and integration from the ecosystem strengthens the customer profile engine and accelerate consumption of proprietary capabilities across the platform.
At FICO, AI is already driving meaningful results today while creating significant opportunities that we are well positioned to capture. I'll now pass it back to Steve to provide further financial details.
Thanks, and good afternoon, everyone. As Will mentioned, our Scores segment revenues for the quarter were $475 million, up 60% from the prior year. As shown on Page 16 of our presentation, B2B revenues were up 72%, primarily attributable to higher mortgage origination scores unit price and an increase in volume of mortgage origination. Our B2C revenues were up 5% versus the prior year, driven mainly by our indirect channel partners.
Second quarter mortgage originations revenues were up 127% versus the prior year. Mortgage originations revenues accounted for 72% of B2B revenue and 63% of total Scores revenue. Auto originations revenues were up 13%, while credit card, personal loan and other originations revenues were up 6% versus the prior year. For your reference, Page 17 of our presentation provides 5-quarter trending of our scores metrics.
As in the past, our updated guidance assumes conservative score volumes. And to reiterate, we do not anticipate share loss competition in any vertical. Turning to our software segment. Our software ACV bookings for the quarter were $28 million, as shown on Page 18 of our presentation.
On a trailing 12-month basis, ACV bookings reached $126 million this quarter, an increase of 36% from the same period last year. With our strong pipeline, we expect bookings in the second half of the year to exceed the first half of the year. Our total software ARR, as shown on Page 19, was $789 million, a 10% increase over the prior year. Platform ARR was $349 million, representing 44% of our total Q2 '26 ARR.
Platform ARR grew 49% versus the prior year, while nonplatform declined 8% to $440 million this quarter. Platform ARR growth was driven by both new customer wins as well as expanded use cases and volumes from existing customers. Platform ARR growth includes the onetime Q1 liquid credit solution migration and Q2 CCS migrations from non-platform to the platform. Excluding those migrations, our platform ARR growth was in the mid-30% range.
The non-platform year-over-year ARR decline was driven by migrations, end-of-life products and some usage declines. In our CCS business, which contains both platform and non-platform, ARR growth was relatively flat. Our dollar-based net retention rate in the quarter was 109%. Platform NRR was 136%, while our non-platform NRR was 90%.
Platform NRR was driven by a combination of new use cases and increased usage of existing use cases. Second quarter software segment revenues detailed on Page 20 were $217 million, up 7% from the prior year. Within this segment, our SaaS revenues grew by 19%, driven by FICO Platform. Our on-premises revenue declined 4%.
Year-over-year, our platform revenues grew 54%, driven mainly by the success of our land and expand strategy. Non-platform revenues declined 12%, driven mainly by migrations. As a reminder, our FY '26 revenue guidance reflects an expectation of lower point-in-time revenue throughout FY '26 due to fewer non-platform license renewal opportunities compared to the prior year. From a regional point of view, 90% of total company revenues this quarter were derived from our Americas region, which is a combination of both our North America and Latin American region.
Our EMEA region generated 7% of revenues and the Asia Pacific region delivered 3%. Operating expenses for the quarter, as shown on Page 21, were $289 million this quarter versus $278 million in the prior quarter, an increase of 4% quarter-over-quarter, driven by personnel expenses. We expect operating expense dollars to trend modestly upward from the Q2 run rate into the back half of the fiscal year, driven mainly by personnel expenses and marketing for both FICO World and our Scores business. Our non-GAAP operating margin, as shown on Page 22, was 65% for the quarter compared with 58% in the same quarter last year.
We delivered year-over-year non-GAAP operating margin expansion of 712 basis points. The effective tax rate for the quarter was 25.7%, and we expect a full year operating tax rate of 25% to 26% and an effective tax rate of around 24%. At the end of the quarter, we had $272 million in cash and marketable investments. Our total debt at quarter end was $3.64 billion with a weighted average interest rate of 5.5%.
This includes the March issuance of $1 billion in senior notes due 2034, which used some proceeds to fund the redemption of $400 million in senior notes that were due in May. As of March 31, 2026, 93% of our debt was held in senior notes. We had $265 million balance on our revolving line of credit, which is repayable at any time. We anticipate interest rate expense dollars to trend modestly upward from the Q2 run rate into the back half of the fiscal year.
As Will highlighted, we continue to return capital to our shareholders through buybacks, as shown on Page 23. In Q2, we repurchased 484,000 shares for a total cost of $605 million, representing the single largest quarterly repurchase in dollars in FICO history. We continue to view share repurchases as an attractive use of cash. With our recent $1.5 billion Board authorization, strong free cash flow and unutilized revolver, since April 1, we have bought an additional $170 million or 164,000 shares at an average price of $1,040 per share. And with that, I'll turn it back to Will for closing comments.
Thanks, Steve. As we approach the start of FICO World 2026, which is going to happen on May 19 through the 22nd in Orlando, we look forward to showcasing our continued innovations. The event brings together customers and partners from around the world to explore how real-time scalable decision-making is transforming consumer engagement. We remain focused on enabling deeper customer relationships through always-on personalization that drives strong business outcomes.
The conference also provides a forum to connect with industry experts, share best practices and advance initiatives that drive financial inclusion. We had a great first half of our fiscal year, and I'm pleased to report that today, we are raising our full year guidance as we enter the third quarter. As shown on Page 24 of our presentation, revenue guidance is now $2.45 billion, an increase of 23% versus prior year. GAAP net income guidance is now $825 million with GAAP earnings per share of $35.60, an increase of 27% and 34%, respectively.
Non-GAAP net income guidance is now $946 million with non-GAAP earnings per share of $40.45, an increase of 29% and 35%, respectively. With that, I'm going to turn it back to Dave, and we'll open up for Q&A.
Thanks, Will. This concludes our prepared remarks. We're now ready to take questions. Operator, please open the lines.
[Operator Instructions] Our first question is going to come from the line of Jason Haas with Wells Fargo.
I'm curious to start, Will, if you could talk about the philosophy behind adjusting your pricing model going to the $0.99 upfront. I appreciate some commentary.
Yes, absolutely. So that's a step in the direction we've been talking about now for several years. I mean we have historically charged upfront for score. That's the historical way we have always charged for our IP.
But what that does is it doesn't spread the cost across the rest of the value chain. And so a lot of the beneficiaries of the IP are not really paying for it. And so we have that cost concentrated upfront. The whole idea behind moving to the performance model was to give us more flexibility so that we could distribute the value -- the monetization of that IP over more players across the chain.
And so that's really what we've done. In this most recent move to $0.99 plus a $65 funding fee, the idea was to encourage adoption of FICO 10T because we think that the most powerful thing that we can do is really get FICO 10T established. And obviously, it's already established in the non-performing market, but we'd really like to encourage wide use of 10T. And so this kind of pricing is designed to encourage that.
Great. That certainly makes sense. And then now that VantageScore is available to be used on the conforming mortgage market, do you expect -- what percentage of lenders do you think would shift fully away from FICO to just using VantageScore? Or do you see most lenders, if they are going to use VantageScore, do you see them also pulling FICO during the mortgage process and then submitting the score ultimately that's most favorable to them to the GSEs.
I suppose we'll see how it turns out. But if you think about the decision process for those who purchase scores, if they're after the most predictive score, 10T is the answer to that. If they're after price, then I think we have parity, 10T at $0.99 is at parity with Vantage at $0.99. And so on both predictability and price, we think we're highly competitive and frankly, don't see good reasons to switch. Now depending on how the FHFA decides to handle the gaming problem, there may be opportunities for Vantage based on the gaming.
And so we'll just have to see how that unfolds. Although our analysis suggests that in a gaming scenario, if there's true consumer shopping for the best rate and the system is going to be gamed in that way, that originators and lenders would wind up pulling both scores.
Our next question will come from the line of Manav Patni with Barclays.
Will, for the 10T adoption, obviously, that $0.99 is only available through the direct loan model that you have, DLP model. Can you give us an update on when that's going live, what the feedback right now is with lenders and kind of adoption that you expect there?
Yes, absolutely. So there's a few pieces to getting the direct license program live, and they're mostly in place. We're working on the last kind of final details now. So we have 3 of the top 5 major resellers signed up.
We are in deep discussion with the other 2 and fully anticipate that all 5 of the big resellers will be able to provide the direct license program. We also see a great deal of interest from the lender community for this performance-based pricing model. So there's a pent-up demand, and we anticipate quite a lot of usage of this model once we get direct up and running. We do still need FHFA final sign-off on having the resellers calculate the score.
But we don't anticipate any issues there because the math is identical and the score we've tested and the score calculated by the resellers is the same score as that calculated by the bureau. It's on the same data. It's the same methodology. So although I can't give you a date, I can tell you that we're closing in on it.
Okay. And then just in terms of the historical 10T data coming out sometime in the summer, maybe just some help on how that process works? Like will there be another pilot like they're doing now with VantageScore once 10T is out and we're only looking for something realistically in 2027 for both to be ready to go fully live, I guess?
Well, the FICO 10T data, as you know, is with the FHFA and the GSEs, and it's up to them to decide when to release it. There's certainly a lot of market sentiment for being able to evaluate 10T and Vantage at the same time. And certainly, by the time the GSEs accept -- truly accept Vantage, I think the market would like 10T to be available as well. So there's some market pressure to get this done, but I don't have the time line.
Our next question will come from the line of Simon Clinch with Rothschild & Co Redburn.
Well, I was wondering if you could just cycle back to the question. I think it was Jason asked about the pricing of 10T. And your comments that it's at parity VantageScore. I was wondering if you could talk about the philosophy or like how you think lenders will treat the success fee in that kind of situation and how we should think about that dynamic in that sort of comparison?
Well, I think the beauty of the way we've structured this is that mortgage originators and lenders have a choice. They can continue to buy the score the way they always have on a per square basis or if they prefer, they can move to the $0.99 plus funding fee. And the idea there is that it encourages very widespread use of the score in the prospecting phase, in the customer acquisition phase and figuring out who's qualified for a mortgage. And frankly, with the goal of trying to encourage more housing and more mortgages, making the upfront score cost very low is likely to support that.
And so it really is up to the lenders, which model they prefer, and we leave it to them. We are -- I've said before, we're largely indifferent as between the 2 models because it's about revenue neutral for us either way. But I think that each model meets the needs of different customers for the score in different ways.
Understood. And just as a follow-up to the reseller readiness right now. I mean, I understand we're getting close to go live to come into place. A bit -- I would love to get a bit more color on is just, I guess, sort of what has -- relative to initial sort of expectations, it feels like it's taking longer than expected.
And I was wondering if you could talk about sort of what has been behind some of the prolonged process here.
I think that some of the expectations were a little on the optimistic side. We certainly didn't think it was going to happen in a couple of months. We thought that it would take a while to put this together. It's a pretty complicated program, not a complicated program, but it's -- there's enough moving parts that require validation and testing that we knew it was going to take some time.
This much time, I would say, we actually believe that it would be up and running by now. I would say that we're close and as I said earlier, it's really up to the FHFA to sign off on the calculation of scores by the resellers and then we're pretty much there.
Our next question comes from the line of Surinder Thind with Jefferies.
Well, just following up on the timing of 10T. Just to understand, is there a sequence of dependencies before the FHFA kind of makes it available in the sense of like releasing the historical data. Obviously, you got to have the systems and everything ready. But are there other things that we should be aware of?
Or is it just kind of once the systems are ready, they can release it, whether or not the historical data is available?
No. I would say that there are not a bunch of additional things that no one knows about. I think we have to get the 10T data out so that people can test it and then the GSEs have to accept 10T, and that's it. That's all that's required.
Got it. And then in terms of just switching away. Can you maybe talk a little bit about the outlook for expenses here. I noticed you talked a little bit about incremental scores, marketing expense what should we expect there? And then other than kind of the step-up that's related to the annual FICO World Conference.
Yes. I mean it's not all that material. I mean there'll be some expense. I mean it's not I think you can kind of back into it when you look at our guidance numbers, but it's not all immaterial.
But we've got some -- there's some additional personnel expense. We got expenses around FICO World. There's some other types of marketing we're doing. When you see more growth on the software side, that comes at a -- that's not 100% margin either, right?
There's cost of goods sold. So you're going to see some expenses there. But none of it's all immaterial.
Our next question comes from the line of Faiza Awa with Deutsche Bank.
So first, I wanted to ask about the very strong growth that you saw in mortgage revenue this quarter, up 127%. I think we know about your pricing but it implies pretty strong volume growth. So I'm just curious if you can talk a little bit more about some of the factors there.
Yes. I mean we had decent volume growth. I think it was a pretty good quarter. There was a period of time that where interest rates dropped a little bit.
We saw a little bit of an uptick here and I think it's consistent with what you hear from the bureaus as well. So it was a decent volume quarter, probably better than we expected when we gave our guidance. But again, we guide very conservatively because it's really difficult to know what those numbers might be.
Okay. Understood. And then just on the software side of the business, again, pretty strong bookings, really strong ARR growth on the platform side. So again, give us some context in terms of what you're seeing there?
Are you seeing higher [indiscernible] and I have noticed that you alluded to growth or maybe focusing outside of financial services. And I'm curious if you're sort of thinking your approach there at all?
I would not say that moving to other verticals is driving the growth. It's really primarily in financial services. And it's across a wide range of use cases. And we continue to have success.
And the model that we've been experiencing just continues to be strong, which is a financial institution will adopt the platform and make it the kind of the heart and soul of the way they interact with their consumer customers and then discover just how powerful it is and then get more utility out of it, the more use cases they put on it. And so it's the land and expand strategy, which we have for that business is working really nicely. And the customers have tremendous satisfaction, and that's driving the growth.
Our next question will come from the line of Jeff Meuler with Baird.
From an earlier question, it sounds like the answer may be TBD depending upon what FHFA decides to do. And I don't know, do we have to wait for the selling guidelines. But the question is, what's your understanding? Because I think the language is the enterprises cannot accept scores from multiple models.
But have they said anything about if an underwriter can pull scores from multiple models earlier in the process? Or is that waiting for the selling guidelines to know the answer?
I think that's waiting on the selling guidelines. I mean I can't speak for the GSEs on that.
Okay. And then do you have any sense of what went into the approval process of the 21 initially approved lenders for Vantage 4.0. Were they asked to apply by FHFA? Is there any sort of like commitment, how intensive of a process it is? Just trying to figure out if that's a meaningful signal or not.
We don't really have a lot of detail around that program. Obviously, we weren't invited to be part of it. And so we just don't have the details. It remains to be seen what happens there. Our understanding is a fairly manual process.
Our next question comes from the line of Ashish Sabadra with RBC Capital Markets.
I know you just announced the FICO 10T pricing, but I just wanted to understand what's your pricing strategy over the midterm? Is there still a gap between price and value and as you think about it, how do you think about closing that gap? Would you also consider alternative pricing algorithms, including a percentage of the loan amount for the success fees. So any color there?
As you know, we've talked about a lot of different approaches to pricing for our IP. And those are under constant evaluation and study. And the balancing act is, we don't want to shock the market. We don't want to make precipitous changes.
In fact, we don't love change. We -- the market works really well the way it is today, and so we don't like change. That said, there is a case to be made for low pricing upfront. There's a case to be made for shifting around the monetization of the IP across more than just the first purchaser.
And so we're always evaluating those kinds of things. Our philosophy has not changed. What you see is the first couple of steps in the direction of what we've been talking about for several years now.
That's very helpful color. And then maybe just on the VantageScore LNPA grids, FHFA mentioned that they are taking into account proper credit risk accounting in order to make sure -- and that's why those matrices are different compared to FICO. I was wondering as based on your experience, what are the key credit risk that they would consider when they are designing these matrices? And why should FICO or FICO 10T get a preference?
Well, so again, I can't really speak for the way the GSEs are thinking about it. But what we believe is that in these LLPA grids, if you're going to account for risk, there's going to be price differential. There's going to be gaming that goes on. What kind of risks might be accounted for?
I don't know how they account for them exactly, but certainly, you could have very different credit default risk for Vantage versus FICO. You could have very different prepayment risk for Vantage versus FICO. As you know, Vantage only goes -- the Vantage data only goes back to 2013. It's never been tested through a full cycle. And so there's a lack of understanding, not for want of trying, but there's just -- the data is not there to understand how Vantage will operate through a full cycle.
And so I'm not really sure -- what does that mean? It means that downstream, investors are going to demand some kind of a premium for the lack of understanding around the prepayment risk and default risk. How that gets translated into the LLPA grid, the G fees, hard to say. And then because the pricing will be different for FICO and Vantage, and we guess that sometimes Vantage will have better pricing for consumer and sometimes FICO will have better pricing for consumer.
It's going to create some real headaches for the GSEs. So we'll see. We'll just have to see how they solve that problem.
Next question will come from the line of George Tong with Goldman Sachs.
With the direct licensing program, it sounds like you're awaiting FHFA approval. Are there other implementation hurdles they have to overcome among the top 3 resellers that have signed up so far? And can you talk about why the remaining 2 out of the top 5 are taking a bit longer to sign up?
I would say that there are not other factors, nothing meaningful. So we're really just waiting on approval from the PSCs and from the FHFA. And then in terms of the 2 that haven't signed, I can't get into the details, but we're very close.
Okay. Got it. And then with respect to your outlook, can you elaborate on what assumptions are baked into your full year guide with respect to VantageScore adoption, the timing of the direct licensing model going live and performance fee adoption?
Yes. We anticipate no loss of volume to Vantage in this fiscal year. That's in our -- that's assumed in our guide. We are -- as I said earlier, we're in roughly the same place financially, whether they go with the per score model or the performance model.
So it's revenue neutral. There's a little bit of a timing difference because with the performance model, the funding fee would trail the initial fees. So I mean there's some minor differences, but I would say on balance, it's pretty close to a wash between the two. So it doesn't really matter when the adoption occurs.
I suppose you could argue that if the adoption of the direct license program is delayed, that's beneficial to FICO in the very short term from a timing standpoint, but we don't think about it that way.
Yes. And we do have some lag built into the guidance based on the assumption that performance model will go live, and we'll have some revenue that's pushed from late this fiscal year and early next fiscal year because again, that the [indiscernible] timing cost described...
Next question is going to come from the line of Alexander Hess with JPMorgan.
Could you start with the 127% year-on-year growth in mortgage? I understand that your rack rate is widely known. Layer on top of that volume assumption is still a bit below. So maybe were there any prior year pricing adjustments have feathered into the present fiscal year.
Just anything that might have given that an extra booster is this sort of the rate you guys think you can continue at these volume levels?
Yes. I mean not really. I mean there might be some difference in the unit cost. I mean there's some without getting to a lot of detail that some people are on a little bit lighter rate last year, and we're up to the full wrap rate this quarter. But it's primarily just the new rate and then the additional volumes we saw.
Got it. And then maybe shifting to usage of the FICO score overall. I know there were some remarks about stepping up expenses for the Scores business, introducing the new version of UltraFICO. If you could just talk about your investments in innovation in the Scores business and how that sort of benefits the franchise you guys have there? That would be super helpful.
In the scheme of things, the investments and incremental expense is not large, okay? I mean, just to be really clear. That said, we are constantly investing in innovation, developing new scores. UltraFICO is -- although we've talked about it for several years, it is very much on our minds, and we have a plan, which we're going to talk about at FICO World next month.
But I can't go into the details now, but UltraFICO is likely to be a pretty significant factor in the Scores business in the future.
Next question is going to come from the line of Kyle Peterson with Needham.
I wanted to just start off on software. The platform growth remains really impressive. Bookings are really good. I know the non-platform was kind of ran off maybe a little faster than we expected in the second quarter in a row.
But I guess should we expect this trend to continue where the platform growth is accelerating, the non-platform is running off? Or do you think it will kind of return to flattish non-platform and historical platform growth? Just I guess, the moving pieces there would be helpful.
It's a good question, Kyle. And we've talked about this in the past. There's -- we have the platform growth, which comes from selling the platform often to customers -- generally to customers we already have, but not necessarily for the same things that they've been doing with us on the legacy side. And so there's new growth in platform, which look like new deals with customers that we know and occasionally with customers that we've never met before.
And then there's migration from our legacy applications to platform. And I would tell you there that we are not forcing that migration. We're not even really encouraging that migration because we have our hands full with the growth in the new platform. And so we really leave it to the customer.
It's the customer's choice. If the customer comes to us and wants to renew for 3 more years, a legacy application that is working extremely well for them, we are all for it. And it's a highly profitable business for us, and it's good. If they're ready to make the move, we're happy to help them make the move.
And so we work on that, too. I think there is a balance there. I think there at some level, there's a bit of migration that happens from the legacy business to the platform business. And so that would explain higher growth on the one side means a little bit lower growth -- a loss of business on the legacy side.
But I wouldn't say it's a huge factor. I just think that the two are kind of in balance at this level now. We're not pushing it with our thumb on the scale one way or the other. That may change in the future.
But for now, we're very happy with the growth on the platform side.
Got it. That's helpful. And then as a follow-up, I wanted to switch over to auto origination Scores revenue. I guess, it did decelerate a little bit this quarter.
Obviously, I think the comps are getting tougher, but I want to see at least directionally, if you guys could give a little bit more color on what drove the year-on-year detail between tougher comps, pricing changes in calendar year '26 or any changes in origination volumes or trends that you guys are seeing?
It's really the tough comps. The volumes are not growing as rapidly as they were. The pricing is relatively consistent. The '26 price increase is consistent with '25. I think what you see is that the comps are difficult, and there's probably a little bit of mix shift there in terms of the pricing tiers that some of the lower unit cost pricing tiers have gained the volume from those that are higher unit costs.
So there's some of that happening in the auto industry in general.
Our next question comes from the line of Craig Huber with Huber Research Partners.
We've talked about this in the past, but can you just update us on your understanding, what's the data show you in terms of what the market share out there is for VantageScore in credit cards, autos, personal loans, and also nonconforming mortgage loans. What's their market share right now and we'll go from there.
I guess it all depends on how you measure it because if you ask them, they would tell you they have significant market share and all those things. Near as we can tell, nobody is paying for VantageScores and the bureau send along the VantageScore for free when someone buys a FICO score. So when you see the big VantageScore volumes that Vantage talks about, you should know that they're largely unpaid for. So are they -- is anyone using them?
I don't know. Is anyone paying for them? Our sense is not much and so it's pretty hard to triangulate on what their market share is. I mean I think it's trivial is what I would say.
And I think you see that in our numbers, right? I mean if there were -- we were losing market share, you'd see it in our numbers, and you don't see any that. We have to report our results or audited. They don't have that same obligation, so there's a lot of scrutiny on what we produce, and we back it up with actual numbers that are verified.
So just to be clear, if you had a ballpark, you think it might be 5%, 10% market share? Maybe it sounds like not even that...
Ballpark, I would call it 2%.
Okay. So then on the nonconforming part of mortgages, you're saying probably the same thing, right, roughly that...
No. On the nonconforming part of mortgages, they don't particularly any care at all. Just to be really clear, in the nonconforming market, the lenders use FICO Classic and they use FICO 10T, and they don't use Vantage.
So what's -- all the worry out there about AI, put that aside for a second, all the worry out there that VantageScore is going to take significant share just because of the changes from the government standpoint. The rest of the market here is -- you guys have been -- VantageScores have been going up against FICO for 20 years, right, since 2006. You're telling me it's roughly 2% market share, give or take.
We don't know. No one know.
What's going to change, though, but what's going to change here on the conforming mortgage side of things here that they're going to get significant market share. I mean that's the theory out there for a lot of people. What's the case there that you can possibly see?
Look, I am not going to make the case for how Vantage takes market share because I think we're competitive on price. We are far more competitive on predictiveness. We have a better score than Vantage. There's not a good reason for them to take any share at all.
Okay. Let me just -- my final question and is why did you lower the upfront fee down to $0.99 from $5 then?
Two reasons. One is to be competitive with Vantage and to have a low entry point and encourage widespread use of the score and second, to encourage adoption of FICO 10T. A pretty classic approach to launching a new product is to price it so that people use it.
But again, you're not worried at all that Vantage is going to take any meaningful share from you on the conforming mortgage side, right? That's what you're saying?
That is correct.
Our next question comes from the line of Ryan Griffin with BMO Capital Markets.
I'm just wondering if you have any feedback to share from the securitization market in terms of reference in light of...
Everyone has done their own market checks, and we have to. And I would say that the securitization market is not ready to accept Vantage. It's -- there's some hurdles to be overcome. And so we'll see how that all unfolds.
I don't have a lot of insight there. I mean the market is still all FICO. I think something like 20 mortgages have been securitized with VantageScore paper and -- which is obviously less than 1%, less than 0.1% of the most recent securitization. So it's not real yet.
We'll have to see how the market reacts.
And I know we're getting some data released over the summer from the GSEs. I was wondering what you're expecting that relate the tail and how you think it might validate the predictiveness in FICO?
Well, I think that I can't give you a date for when the FHFA will release the FICO 10T data to the marketplace. But we're certainly not standing in the way. We provide the data and we're ready to go. In terms of validating the predictiveness, we have white papers posted on our website that actually analyze FICO 10T versus Vantage and provide insights on credit default risk and prepayment risk and the differences.
We qualify 5% more borrowers. I mean there's a lot to see there. That's already been done. But then if you don't believe FICO because it's self-serving, I'd encourage you to look to third-party analysis as they come out because I'm sure they will.
And you're going to see a lot of analytic work around this topic in the coming weeks and months.
Our next question comes from the line of Owen Lau with Clear Street.
So the AI disruption narrative hasn't gone away. Could you please talk about why it's very hard for whatever Vantage or a third-party AI platform to come in and create a more predictive credit score, which will be adopted by lenders and consumers if they can offer a lower price?
Okay. So there are two different things there. One is AI versus the current credit scoring system. And the second is within that, more predictive.
So first, I would say, with respect to AI displacing the FICO score, we have a really well-defined body of law and the fair lending laws, which are designed to protect consumers to ensure that there's not discrimination, ensure that consumers are treated fairly. And that requires compliance with all kinds of things that our scores take into account. I mean just one small example would be red lining, which is not allowed in the United States. Is it a predictive factor?
Yes, it's a predictive factor, but it's not allowed. And so you can't use red lining as a factor in a credit score. Well, AI doesn't -- AI would find 100 other ways to get to the same result. And so the regulators are not going to be comfortable with AI making underwriting decisions when they're not explainable when it's a black box, when they can't demonstrate that discrimination is not occurring.
So that's kind of the core problem with using AI and underwriting. I mean AI is great in a lot of things. But using it in underwriting, the biggest play is that it's going to get around the rules and regulations of the fair lending laws. Now you're probably aware that FICO scores carry with them 32 reason codes.
So when a consumer turn down for credit, they get a letter and/or the line is not increased on a request or whatever, they get a letter, and the letter says, here's why. And that reaches into the FICO score and the reason codes and those reason codes are shared with the consumer. And so there's a level of comfort with the regulators and with the consumer that they understand what's going on. I would also point out that the experiment with AI and some of the black box underwriting that was undertaken several years ago by Upstart ended with the CFPB shutting it down.
So I think there's some real challenges, not that it will be this way forever. And we are prepared for the day when AI is appropriate in underwriting. We have patents in the area of explainability and ethical AI. And so I think we're in an advantaged position, but I would not hold my breath.
I think that's going to take a long time. And then on predictiveness of the score, I would tell you that our latest and greatest score is more predictive than Vantage. And frankly, more predictive than any other score out there. The only asterisk I would put on that is there are lenders who build proprietary scores on top of FICO, and they leverage their first-party data.
And so they have incremental data and they get incremental signal out of that. And so there are some proprietary scores that are really excellent that are most typically developed on top of FICO.
Got it. And then maybe quickly on LLPA, have you heard of any of these 21 lenders received the updated LLPA grid from FHFA for the pilot? And do you have any expectation that when the new grid will be made public?
No idea. I have heard nothing, I encourage you guys to keep asking the questions, what's going on there? I think it's a manual process.
Our next question comes from the line of Scott Wurtzel with Wolfe Research.
Just on the guidance, I understand you're still being -- it seems like being conservative on your assumptions regarding volume. Just wondering if there had been any sort of change to your volume assumptions after the last quarter at all?
Not really. I mean, again, we tend to be pretty conservative because, obviously, there's a lot happening in the world. And if we get that number wrong, it's difficult to make that up someplace else, but not really. I mean I think we had a better second quarter volume-wise than we had anticipated when we gave guidance.
But we don't necessarily think that's going to continue. So we tend to take the same conservative approach for the rest of the year.
Got it. And then just on the buyback, I mean, the number, $600 million in the quarter was great to see along with the incremental buyback this quarter. Just wondering, I mean, how aggressive do you think or would you guys be with the stock at these current levels and given the capacity that you have?
What I can say is what we've said in the past, we're always interested in share repurchase, and we're in the market kind of all the time. And we tend not to be market timers, although we have leaned in much more heavily on an opportunistic basis. I would certainly consider our stock at these levels to be an opportunistic time.
Our next question comes from the line of Kevin McVeigh with UBS.
I wonder if you had any thoughts on, given the current shifts in the regulatory environment, do you feel like that's pretty much contained at this point? Or is there anything else you're kind of focused on as we think about whether it's FHFA or other parts that you kind of continue to manage through from a regulatory perspective?
The mortgage market is a $13 trillion market, and everyone takes it pretty seriously, and no one wants to do things that are reckless there. And so everything that happens in that market, you see coming a mile away. And I think that's kind of where we are. I think we know everything there is to know about the way this is unfolding for now.
And so no, I don't really see being blindsided by regulatory or other kinds of things in the market. I think we understand how the market is evolving. We understand what the choices are for evaluating credit in the modern market. Will things change if the GSEs get out?
I mean anybody's guess when and if that happens, and will things change? We actually don't think they'll change that much. And we think that in a world where the GSEs are private or if they were to lose the guarantee, the emphasis on credit default risk would go up, the interest in credit default risk goes up, and that's advantaged FICO because we have the best score for evaluating that. But again, these are more theoretical and down the road kinds of things.
I don't think there's any surprises ahead.
Our next question comes from the line of Curtis Nagle with Bank of America.
Most of my questions have been taken, but just maybe, Will, I guess, any stats or detail you could provide in terms of the uptake of 10T within the nonconforming market at this point for mortgages?
Yes. I don't have an updated number for you, but it's -- we have underwritten trillion.
Yes. Most of them are running in parallel with Classics because they want to be able to use the latest score and so they run in parallel with each other.
Got it. I just -- any running...
I think the number is $1.2 trillion. The latest number.
And our last question is going to come from the line of Sean Kennedy with Mizuho.
So with VantageScore, I was wondering if you could discuss a bit more about potential adverse selection, how lenders could pull both scores in the beginning of a process that could pick one or the other for the remaining initial results and the implications there for the mortgage market.
Yes, it's a good question. I think -- and of course, we don't know how this is going to unfold. I mean it is strong. It's really in the interest of the GSEs and the FHFA to prevent gaming to not have a gaming situation.
That said, in a 2-score system, it's almost inevitable. It's kind of structural that one score or the other is going to be more beneficial to the consumer at all times. And so in a world where the systems are in place to use both scores and barring other unforeseen things, there will be some people who pull both scores. And so it may unfold that way.
I think to the extent that, that happens, that's not -- I mean, it is technically share loss for FICO, but it's not volume loss. What you're really doing is expanding the market by the second pull. And so it's conceivable that Vantage could get some share that way if they don't solve the gaming problem. But I don't -- again, I don't see volume loss for FICO.
Great. And then I was also wondering just with the auto and card key loan growth, if you saw any volume weakness later in the quarter [Technical Difficulty] on the macro. And if you were seeing any consumer weakness there?
Yes. Auto tends to be pretty stable unless there's like a really disruption in the economy. A lot of the volume on the card side is really the banks that are marketing. And if they want to market more, they'll find consumers that will take it off.
So -- and that can vary quarter-to-quarter. But so far, we haven't really seen any significant weakness on the volumes. They've actually been pretty good. There's been a little bit of a falloff in the subprime, but it's been picked up throughout the rest of the prime, super prime.
So we haven't really seen any...
This does conclude today's question-and-answer session. Ladies and gentlemen, this also does conclude today's conference call. Thank you for participating, and you may now disconnect. Everyone, have a great day.