Prepared remarks
Good day and welcome to the Q3 2026 FICO Earnings Conference Call. Please be advised that today's conference is being recorded. It is now my pleasure to introduce Dave Singleton. Please go ahead.
Good afternoon and thank you for attending FICO's third quarter earnings call. I'm Dave Singleton, Vice President of Investor Relations and I'm joined today by our CEO, Will Lansing; and 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 an 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. A replay of this webcast will be available through July 29, 2027. Our quarterly investor presentation is available in 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 third quarter earnings call. With another strong quarter, we are increasing our fiscal 2026 guidance. We reported Q3 revenues of $674 million, up 26% over last year, as shown on Page 5 of our investor presentation. For the quarter, we reported $237 million in GAAP net income, up 30% and GAAP earnings of $10.45 per share, up 41% from the prior year. We reported $277 million in non-GAAP net income, up 31% and non-GAAP earnings of $12.18 per share, up 42% from the prior year. We delivered free cash flow of $370 million in our third quarter. Over the last 4 quarters, we delivered $961 million in free cash flow, an increase of 28% over the prior 4-quarter period. In Q3, we returned significant capital to shareholders through share repurchases with repurchase spending exceeding 3x the historical record quarter. Including our accelerated share repurchase plan, we bought back $1.96 billion or 1.705 million shares at an average price of $1,149 per share. At the segment level shown on Page 6, Scores segment revenues in our third quarter were $459 million, up 41% versus the prior year. While B2B Scores were the key driver of growth, we also experienced continued growth in B2C Scores. In our Software segment, we delivered $215 million in Q3 revenues, up 2% year-over-year. Results included 66% platform revenue growth and a 25% decline in non-platform revenue. Steve will provide additional revenue segment level details later. Beyond the financial results, we continue to make meaningful progress against the strategic priorities that position FICO for long-term growth. With more than 70 years of innovation, FICO has been the trusted backbone of high-stakes decision-making, turning data into intelligence and intelligence into better business outcomes. That leadership continued this quarter with the GSE release of the FICO Score 10T data sets and UltraFICO general availability. Fannie Mae and Freddie Mac recently released expanded historical level data sets for FICO Score 10T, enabling mortgage ecosystem participants to independently evaluate credit score performance using real-world GSE mortgage data. Independent analysis by Milliman, a leading global actuarial and risk management firm, confirmed these findings, concluding that FICO Score 10T outperforms Vantage 4 on all three key statistical measures of predictiveness and across every origination year studied, both individually and in aggregate. Milliman found that FICO Score 10T delivers more than a 10% predictive advantage for first-time homebuyers and more than an 8% predictive advantage over Vantage 4 for the most recent origination years, an especially important cohort because it exhibits some of the highest default rates in the data set. FICO's predictive advantage is not driven by access to different data. FICO Score 10T and Vantage 4 are built on the same underlying data sets. The comparison of the models can be found in our investor presentation on Page 10. The difference lies in FICO's decades of experience developing predictive credit risk models and how FICO transforms that data into a more accurate assessment of default risk. That predictive advantage has meaningful implications across the mortgage ecosystem. For lenders and originators, a more predictive score enables better risk assessment, more confident lending decisions, improved portfolio performance and the ability to responsibly expand access for borrowers. For investors and capital markets participants, stronger default prediction supports more accurate risk measurement, pricing confidence and ultimately a more resilient housing finance system. For consumers, it enables more precise risk-based pricing, broader access to credit and better borrowing outcomes. The FICO Score 10T Adopter Program provides lenders with historical data and makes FICO Score 10T available at no additional cost alongside classic FICO, enabling production testing and validation within existing workflows. Ecosystem participants actively opt in and FICO works directly with them to evaluate and prepare for production deployment. The program has grown to 70 lenders, spanning both conforming and nonconforming mortgage markets. The program now represents about 55% of the volume generated by the top 50 mortgage originators, $587 billion in eligible annual originations based on 2025 HMDA data and more than $1.87 trillion in eligible annual servicing. Complementing the adopter program, FICO has expanded the technology infrastructure supporting FICO Score 10T adoption. To further streamline implementation and deployment, FICO Score 10T is now integrated into Optimal Blue's market-leading mortgage platform and LoanPASS's automated product pricing and eligibility platform. These integrations enable lenders to leverage the industry's most predictive credit score throughout the mortgage life cycle, including loan eligibility, pricing, hedging, trading and portfolio evaluation. This allows lenders to adopt FICO Score 10T using the platforms and workflows they already rely on today. Now turning to UltraFICO. We recently announced the general availability of the next-generation UltraFICO Score developed in partnership with Plaid. The new score combines the FICO Score with consumer permissioned cash flow data from Plaid's network of more than 12,000 financial institutions, giving lenders a more complete view of credit risk on the same score scale they already use today. Our initial target market for this score is subprime and near-prime consumers across card, personal loan and auto lending. Our analysis shows that 79% of nonprime applicants with a history of positive account balances saw higher scores under UltraFICO, reflecting the score's ability to recognize positive financial behavior that isn't captured by traditional credit file data for this population. We also found a 7% relative increase in approvals with no incremental risk and a 15% relative performance lift for prime applicants with limited credit histories, showing that predictiveness improves beyond what traditional credit scoring alone can capture. Since general availability began just this past May, we're still in the early days of adoption. A pipeline of lender interest exists today. As we continue to build that pipeline, we expect to onboard clients for testing. Another strategic priority is the FICO Mortgage Direct Licensing Program, which is still under review by the GSEs. This remains the key milestone for the program to go live and for lenders to realize cost savings through performance model pricing. Lender interest in the program remains strong and we continue to expand reseller participation. We signed direct license agreements with partners and resellers representing about 60% of mortgage volume and we're in active negotiations with the remaining material resellers that would bring us closer to 90% of mortgage volume once finalized. This past quarter, we hosted FICO World 2026, where customers and partners echoed a consistent theme. AI adoption is accelerating at an unprecedented pace, reshaping how businesses operate and how consumers interact with financial institutions. Three structural forces are driving this shift: the need to operationalize AI at enterprise scale, rising regulatory demands for governance and explainability and evolving customer expectations for personalized real-time decisions as AI agents emerge. For our customers, the real challenge is not investing in AI or experimenting with AI. It's turning their AI investment into business outcomes and measuring business value while keeping every decision governed, explainable and auditable. Customers are answering the challenge by integrating FICO Platform, the world's leading AI decisioning platform for the financial services industry, into their enterprise architecture and building their business solutions on FICO Platform. The FICO Platform is differentiated by a number of things. First, FICO leverages 70 years of domain expertise in financial services. Second, FICO Platform benefits from proprietary data sets such as our fraud consortium data spanning thousands of financial institutions. Third, FICO Platform clients that leverage multiple use cases benefit from a compounding feedback loop that can create a more complete picture of the customer, utilizing the always-on and always-available AI-powered customer profile engine. Fourth, our FICO Platform architecture enables responsible AI through decisions that are auditable, transparent and explainable, allowing clients to more easily adhere to governance and regulatory requirements. Fifth, FICO Platform decisioning capabilities are deeply embedded into enterprise workflows, delivering complex decisions in real time at scale in milliseconds and with a high degree of reliability. Our investments are focused on development and distribution of market-leading and differentiated intellectual property. These include the development of FICO Platform and technologies such as focused sequence models and focused language models. This requires limited CapEx as we leverage cloud providers for scalability. We continue to deliver healthy year-over-year growth in bookings, ARR, DBNRR and enterprise platform clients, demonstrating real-world value for our customers and tangible results from our investments. Our near-term focus has been on driving top line growth, while our long-term focus is on driving margin expansion. We've advanced two initiatives that will support these objectives. First, in July, we expanded our collaboration with Accenture by pairing the FICO Platform with Accenture's experience in risk, AI and industry operations. This partnership will help enterprises turn investments into real business results, faster decisions, stronger risk controls and outcomes that hold up under regulatory scrutiny. Our immediate focus is go-to-market and enablement with a phased-in geographic rollout. Second, later this calendar year, we anticipate the general availability of our next-generation FICO Platform, which includes our enterprise fraud solution. With incremental IP and expanded distribution, we anticipate greater penetration of FICO Platform within our current 500 named target accounts and an expansion of our operating market beyond those accounts. I'll now pass this to Steve to provide further financial details.
Thanks, Will and good afternoon, everyone. As Will mentioned, our Scores segment revenues for the quarter were $459 million, up 41% from the prior year. As shown on Page 17 of our presentation, B2B revenues were up 49%, primarily attributable to a higher mortgage origination score unit price. In the prior year quarter, FICO recognized approximately $16 million on a multiyear U.S. license renewal on our insurance score product. Normalized for that, Scores revenues were up 49% and B2B revenues grew 59%, respectively. Our B2C revenues were up 5% versus the prior year. In our mortgage origination scores business, third quarter volumes grew low single digit versus the prior year. Our mortgage originations revenues were up 97% from the prior year. Mortgage origination revenues accounted for 71% of B2B revenues and 62% of total Scores revenues. Auto originations revenues were up 15%, while credit card, personal loan and other originations revenues were up 9% from the prior year. For your reference, Page 18 of our presentation provides quarterly trending for Scores segment metrics. Turning to our Software segment. Our software ACV bookings for the quarter were $29 million, as shown on Page 19 of the presentation. On a trailing 12-month basis, ACV bookings reached $128 million this quarter, an increase of 39% from the same period last year. We continue to see strong growth in our sales pipeline. Our total software ARR, as shown on Page 20, was $816 million, a 10% increase over the prior year. Platform ARR grew 62% versus the prior year to $413 million and represented 51% of our total Q3 '26 ARR, while non-platform declined 17% to $403 million for the quarter. For the first time, platform ARR dollars exceed non-platform ARR dollars, marking an important milestone that reflects the successful execution of our long-term strategy and the returns we are realizing from our strategic investments. Excluding migrations, platform ARR growth was in the mid-30% range, reflecting strong execution in new customer wins as well as expanded use cases and volumes from existing customers. In our non-platform business, ARR declined year-over-year, driven mostly by migrations and to a lesser extent, end-of-life products. Our dollar-based net retention rate in the quarter was 109%. Platform NRR was 148%, while our non-platform NRR was 82%. Platform NRR was driven by a combination of new use cases, increased usage of existing use cases and migrations. Third quarter Software segment revenues detailed on Page 21 were $215 million, up 2% versus the prior year. Within the segment, our SaaS revenues grew 21%, driven by continued strength in FICO Platform. Our on-premises revenues declined 16%, driven by lower point-in-time revenue as we had fewer non-platform license renewal opportunities compared to the prior year quarter. Our professional services revenues declined 24% as the prior year quarter includes revenue from the completion of a large deal milestone. Normalizing for point-in-time revenue and professional services revenue, the Software segment revenues grew 10% versus the prior year. Platform revenues exceeded non-platform revenues for the first time in FICO history. Year-over-year platform revenues grew 66%, driven by success in our land and expand strategy. Excluding migrations, platform revenues grew in the high 30% range. Non-platform revenues declined 25%, driven by migrations and lower point-in-time revenue. As a reminder, platform and non-platform revenues exclude professional services revenues. From a regional perspective, 91% of total company revenues this quarter were derived from our Americas region, which is the combination of our North America and Latin America regions. Our EMEA region generated 6% of revenues and the Asia Pacific region delivered 3%. Operating expenses for the quarter, as shown on Page 22, were $312 million this quarter compared to $289 million in the prior year, an increase of 8% quarter-over-quarter, driven by marketing for FICO World and some personnel expenses. Our updated guidance includes fourth quarter operating expenses that are modestly higher than in our third quarter due to incremental front-end loaded marketing expenses to support the launch of our new partnership with Accenture, as well as some anticipated one-time restructuring charges. Our non-GAAP operating margin, as shown on Page 23, was 62% for the quarter compared with 57% in the same quarter last year. We delivered year-over-year non-GAAP operating margin expansion of 479 basis points. The effective tax rate for the quarter was 24.6%. 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 $305 million in cash and marketable investments. Our total debt at quarter end was $5.58 billion with a weighted average interest rate of 5.64%. This includes the June issuance of a $1.5 billion term loan to fund the accelerated share repurchase. As a result, we expect fourth quarter interest expense to be higher than in the third quarter. As of June 30, 2026, 60% of our debt was held in senior notes, while 40% of our debt was held in term loans or a balance on our revolving line of credit, both of which are repayable at any time. As Will highlighted, we had a record quarter for returning capital to our shareholders through buybacks. As shown on Page 25, in Q3, we repurchased 1.705 million shares for a total cost of $1.96 billion. In the near term, we will be using cash to pay down debt. Beyond that, we continue to view share repurchases as an attractive use of cash. And with that, I'll turn it back to Will for his closing comments.
Thanks, Steve. Elevated interest rates and ongoing affordability challenges continue to weigh on the mortgage market, keeping loan originations below historical norms. While the macro environment remains fluid, our strategy, innovation and execution remain disciplined and consistent. I'm quite pleased to report that today, we're raising our full year guidance as we enter the fourth quarter. As shown on Page 26 of our presentation, revenue guidance is now $2.53 billion, an increase of 20% versus prior year. GAAP net income guidance is now $850 million, with GAAP earnings per share of $36.86, an increase of 30% and 39%, respectively. Non-GAAP net income guidance is now $979 million, with non-GAAP earnings per share of $42.43. Those are increases of 33% and 42%, respectively. With that, I'll turn the call back to Dave and we'll open up for Q&A.
Thanks, Will. This concludes our prepared remarks and we're now ready to take questions. Operator, please open the lines.
Questions and answers
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 resellers representing 60% of the U.S. mortgage volumes. Are they also interested specifically in the performance fee model? Just looking for that incremental color.
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. As we've pointed out, we have agreements already covering 60% of the reseller volume. We are also very close to signing two additional resellers. So we're literally waiting for their okay. In terms of interest from the resellers in the performance model, yes, there is significant interest in the performance model. 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 the industry or your clients in terms of that versus the other alternatives out there.
They now have an opportunity to test the 10T data as provided by the GSEs. Although we'd like to see 10T approved sooner rather than later, there is a time lapse between releasing the data and doing the analysis and then approving 10T. There's a lot of interest. As I mentioned earlier, third parties like Milliman and others have done the analysis and 10T is more predictive than Vantage. There is 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?
That's a very good question. We're not supportive of lender choice as a policy because it encourages gaming. The primary reason one might consider buying a VantageScore is to try to deliver a consumer a better rate when Vantage is higher than FICO. We expected gaming and that's what we're seeing. To your specific question, are we seeing volume loss? No, we are not, which suggests that lenders are pulling both scores. You typically need both scores if you're going to try to game the system. A VantageScore by itself doesn't really get you there. We're seeing a two-score environment with the expected structural problems and experimentation by the most sophisticated players. But we are not seeing volume loss; VantageScore appears additive, not replacing FICO.
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 assume part of it was just mortgage volumes being a bit softer in 3Q. Any other comments on what's driving that? And could you give a framework for how to think about 4Q?
If you look on a year-over-year basis, volumes were up low single digits, which aligns with what the bureaus report overall. We saw a slowdown as the bureaus described; as rates tick up, volumes slow. That explains both the year-over-year and quarter-over-quarter movements. It's primarily the slowdown in the mortgage market as rates crept up.
And our next question comes from the line of Jeff Meuler with Baird.
A follow-up on that last question. The 127% growth that you had last quarter was hard to get there based on market volume and based upon a $10 versus $4.95 price. Was there anything else that was inflating that number? Or is there any timing on when pricing actually took effect last year versus this year?
There is some timing and quarter cut-off effect. When markets move rapidly and rates shift, week-to-week and month-to-month volumes can vary. We don't have access to the underlying bureau data that others report, so there may be differences in cutoff dates and reporting windows. Over time it all works out, but when markets move quickly you can see anomalies like last quarter.
And our next question comes from the line of Simon Clinch with Rothschild and Redburn.
Will, you mentioned that you're not seeing any volume loss. Given your market position, how easy is it for you to monitor any evidence of volume loss?
It's not entirely straightforward, but relative to our forecasts and expectations, we're not seeing volume loss. We triangulate from different sources, including public mortgage data and internal pulls, and the full quarter numbers are consistent with bureau-reported volumes on a year-over-year basis. We are pretty confident our observations are representative.
Okay, great. Following up on other segments—auto, card and personal loans—could you give color on how things are progressing in terms of pricing initiatives, testing, and elasticity?
It's early to predict next year's pricing outcomes. We constantly explore where and how to get revenue growth, becoming more precise and surgical each year to identify pockets that create the least market reaction. We're analyzing mortgage, auto, card, and personal segments, but no final decisions or announcements yet. Our pricing and strategy teams are working on these issues, and we have nothing specific to share today.
Our next question comes from the line of Surinder Thind with Jefferies.
Switching gears to platform. Can you talk about the next-generation product and client conversations on uptake? Are migrations from non-platform to platform expected to accelerate as more features and functionality are released?
Yes, that's exactly what you should expect: continued acceleration. We have tremendous interest and uptake, as you can see in our numbers. New features and functionality expand use cases and platform utility; the more use cases on the platform, the more value customers derive and the lower the marginal cost of adding functionality. Our land-and-expand strategy is working and we see continued growth.
Following on that, now that platform is larger than non-platform and licensing shows volatility, will you consider end-of-life for certain products? How should we think about that?
Yes. We've discussed end-of-life and migration strategy in past quarters. We are not forcing migrations to achieve growth, but simplifying the product catalog benefits FICO. Some products are outdated and should be wound down. We have capacity to move customers to newer products on the platform that provide more functionality at lower cost. We have an active end-of-life strategy and are working through it.
Our next question comes from the line of Ashish Sabadra with RBC Capital Markets.
A follow-up on VantageScore market share in conforming loans. With some larger players showing Vantage around 20% share, are they still using both pulls when they close the loan? How does the pace of adoption compare to your expectations? Any thoughts on what FICO can do to regain market share?
We have early data indicating Vantage may land in the 20s as a theoretical maximum based on the percent of consumers advantaged by Vantage over FICO. We have numbers in the 20s and expect gaming behavior; I don't anticipate Vantage exceeding the 20s in steady state. Vantage is being sold and used alongside FICO, and gaming is a reality in the environment created by lender choice. In terms of actions, many things are on the table—pricing strategies, product features and monetization approaches—but we are not sharing specific tactics on this call.
Our next question comes from the line of Faiza Alwy with Deutsche Bank.
First, I wanted to ask about the revenue guidance raise because it seems significant relative to what you saw this quarter. Is there anything out of the ordinary—are you expecting licensing revenues in the fourth quarter or was there previously assumed DLP activity that caused a lag? More color on what led to the revenue guidance raise would be helpful.
It's more that the mortgage market has been better than we originally anticipated and volumes have held up versus our prior outlook. We also planned that if the DLP went live with the performance piece earlier in the year we might push some revenues, and that hasn't occurred yet. Rates haven't worsened materially and mortgage markets have held up, returning to low single-digit growth, which contributed to the raise.
Understood. And then on the Vantage pilot and 10T, would lenders pull all three scores for gaming purposes if 10T comes into the mix? How do you expect things to evolve?
That's a great point. If a three-score market emerges—classic FICO, Vantage and 10T—there would be an incentive to pull all three and shop scores to find the most advantageous outcome for the consumer. It's hard to imagine retiring classic FICO anytime soon, so most likely 10T will be added to the mix and we'll have score shopping behavior.
Would 10T be charged separately or offered free alongside classic FICO?
Today we bundle 10T with classic FICO so paying for classic includes 10T at no additional cost, and we expect to continue that to encourage adoption. We have not made final pricing decisions and everything is on the table. One could imagine increased revenue if we charged separately for both, but the current plan is to continue bundling to drive adoption.
Our next question comes from the line of Alexander Hess with JPMorgan.
On 10T, there have been analyses showing classic FICO versus Vantage coupon differentials and that much of the spread dissipates after corrections. How do you think about mortgage-level and pool-level spreads for firms using the competitor's score?
There is less history with Vantage and it has not been through a down cycle, so there's more uncertainty. Investors, if rational, should analyze paper that's less well understood. That rationale underpins the 30 basis point estimate we've discussed previously. Today Vantage is a small part of the market and much of the Vantage-securitized paper is commingled into larger FICO pools. Traders will dig in and the differential should appear as they assess risk and pricing.
Thinking about monetization across the Scores business, beyond mortgage, where do you see future legs of monetization? Rental markets, overseas or other opportunities come to mind. How are you thinking about the next legs of monetization holistically?
Too often people view Scores as single-threaded through mortgage. We have opportunities in other verticals and with new scores relying on different data sets to score new populations. UltraFICO, which augments credit file data with cash flow data, is an example and targets subprime and near-prime segments. Adoption curves for new scores can take years. We also have FICO 11 in development and we constantly innovate to derive additional signal from other data sets beyond the credit file.
Our next question comes from the line of Kyle Peterson with Needham.
On DLP and potential gaming, how have conversations with potential lenders gone? Could gaming slow adoption given that if lenders buy from the bureaus they can get Vantage free and save a dollar?
We don't think gaming will slow DLP adoption. The performance model delivers significant cost benefits and encourages customer acquisition and speculative outreach to identify potential borrowers, broadening access. These are desirable to large lenders, and interest in DLP remains strong.
On capital return, given the recent large ASR and the priority to reduce leverage in the near term, should we interpret that you'll be out of the buyback market for the next couple of quarters? Any color on how long you'll prioritize debt reduction over buybacks would be helpful.
We generate substantial free cash flow and delever quickly. It's unlikely we'll buy additional shares beyond what's already in the ASR this quarter. We'll update you next quarter on where we stand. The ASR accelerates buybacks but our EBITDA and cash generation have grown, enabling quicker deleveraging.
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?
We've explored the performance model and funding fee component with lenders; some like it and some do not. There is appetite for it and it will be a driving force behind DLP adoption for those who value cost savings via the funding fee model. It's not universally attractive, particularly for lenders who pull few scores per closed loan.
Our next question comes from the line of Ryan Griffin with BMO Capital Markets.
On the software business, can you talk about retention rate dynamics and whether land-and-expand is the right way to think about it? Also, on ACV bookings, are you expecting acceleration in H2 versus H1?
Bookings are accelerating and we expect continued acceleration in the second half. Land-and-expand is our strategy, and it shows up in DBNRR. We're doing more migrations now compared to several quarters ago, including some CCS migrations, which affects legacy retention rates, but platform NRR is strong. On balance, we're pleased with these dynamics: 148% DBNRR on platform and low churn indicate solid retention and expansion.
As a follow-up, recent trigger loan legislation impacted pre-qual and hard inquiry markets. To what extent did that impact mortgage volume on your side?
We haven't seen a material impact; there's been a little, but nothing dramatic.
Our next question comes from the line of Owen Lau with Clear Street.
Going back to software, platform ARR growth accelerated while non-platform was weak. Should we expect this divergence to continue given the magnitude of the delta?
Yes, expect continued divergence. We've held off on migrations for many quarters because of platform capacity constraints while onboarding new customers. Now we have capacity and are migrating certain legacy offerings that should be wound down. Migration will continue, but the platform growth is not solely cannibalization; we have strong pipeline and new customer wins driving platform adoption. We have a migration plan and team to ensure smooth transitions.
On VantageScore pilot and 10T, is 10T dependent on the Vantage pilot or is there a path to run both in parallel? If 10T is delayed, how would that impact implementation of other FICO Scores like UltraFICO 2 or FICO 11?
The FHFA and GSEs will decide timing for 10T acceptance. 10T is the best score for measuring default risk and has seen adoption in nonconforming markets. When agencies decide to mix it in is up to them. Their recent data release means they are actively working on it, but it takes time for them to become comfortable. UltraFICO and other score innovations are on separate paths and not contingent on 10T's regulatory timing.
Our next question comes from the line of Scott Wurtzel with Wolfe Research.
On the direct license program, can you update where the remaining resellers stand and what's unique about them that's taking a little longer?
We have two large resellers that are signed, two more that are very close to signing, and then a tail of smaller resellers. So we're pretty close to finalizing the remaining agreements.
Our next question comes from the line of Sean Kennedy with Mizuho.
How impactful are partnerships like Accenture for platform growth? Do they significantly help FICO Platform's customer reach and is there a target customer type by geography or size?
Partnerships are essential to address our historical distribution gap. We're excited about our strategic partnership with Accenture, a top global system integrator, which will take our IP to market with their capabilities and relationships. This partnership should expand indirect distribution, open new customer relationships, and help monetize our IP beyond direct sales. It's the beginning of scaled indirect distribution growth.
Regarding net retention inflecting positively this year, is that partially due to AI capabilities or are certain platform products growing faster?
The current platform growth reflects the platform's existing state and the immediate payback customers get today. AI enhancements are coming to the platform and will continue to support growth, but today's retention uplift is primarily driven by platform functionality and land-and-expand success. There's strong appetite for forthcoming AI-driven capabilities.
Our next question comes from the line of Curtis Nagle with Bank of America.
Following up on gaming commentary, are you seeing any anecdotal evidence, such as discounts placed on Vantage securitized loans or other signs suggesting that gaming is occurring in the market?
We know gaming is occurring; we expected it and are seeing it. Examples include behaviors at Rocket and UWM, which are anecdotal evidence of gaming. The environment created by lender choice invites score shopping and gaming, and that's the reality we're operating in.
Our next question comes from the line of Craig Huber with Huber Research Partners.
On the performance model, given we're into August, can you talk about its usage and feedback? Where are we at on adoption?
To be clear, the performance model is distributed through the Direct License Program, which is not yet live because it's awaiting certification from one of the GSEs. I can't speak to real usage because it's not available yet. The market wants it and it's attractive to lenders and resellers; we're waiting on certification to go live.
Is there any commentary you can give on nonconforming market usage of the performance model, or is that also waiting on the conforming certification?
We have not offered that model broadly in the nonconforming market via DLP. The performance model offering is tied to the Direct License Program, which is not yet live.
Our next question comes from the line of Rayna Kumar with Oppenheimer.
Even with the DLP, the credit bureaus are likely to remain large customers of FICO. How has your relationship with them evolved over this process and where do you stand now?
Our relationships with the bureaus are strong and healthy. They are partners and large customers in many areas, including our consumer business with Experian. At the same time, we are now competing in mortgage scores. They've been promoting Vantage for years and now have some traction in mortgage because of lender choice. So while there's competition in scores, the broader relationships remain solid.
Thank you. Ladies and gentlemen, thank you for participating. This does conclude today's program and you may now disconnect.