Scores revenue is up 41%. The company plans to remove 15% of positions

FICO, the Fair Isaac Corporation behind the widely used credit score, posted $458.9 million in Scores revenue for its fiscal third quarter ended June 30, 2026, an increase of 41% year over year. Company-wide revenue reached $674.2 million, up 26%.

An October 6 Form 8-K then disclosed a plan to eliminate approximately 15% of positions across the company. The plan is prospective, not evidence that precisely 15% of Scores personnel had been dismissed. A strong quarter and a broad restructuring can coexist without implying the same metric or employee group.

Meanwhile the US Federal Housing Finance Agency widened lenders' ability to use the competing VantageScore 4.0 model for eligible mortgages. The question is not simply why a growing company would cut roles, but what happens to a historically powerful model when customers gain a viable approved alternative.

A 97% revenue increase with low-single-digit volume growth

FICO's Q3 announcement reported 41% Scores growth and 49% business-to-business scoring revenue growth. The company attributed much of the B2B increase to higher unit pricing for mortgage-origination scores.

On the earnings call management supplied the sharper contrast: mortgage-origination score volumes rose in the low single digits year over year, while revenue from that same activity grew 97%. Those figures refer to a mortgage-origination subset, not the entirety of the Scores segment.

Growth led largely by higher revenue per transaction is economically different from growth driven by an explosion in transactions. Both lift reported revenue, but the sustainability of pricing power may face a new test if buyers can compare and choose competing models. The figures alone cannot predict FICO's future revenues.

The regulatory change is lender choice—not the abolition of FICO

On September 9, 2026, FHFA extended VantageScore 4.0 availability to all lenders approved to originate eligible loans for Fannie Mae and Freddie Mac, without requiring separate prior written approval. For qualifying deliveries, lenders can use Classic FICO or VantageScore 4.0.

On September 30 the two Enterprises adjusted relevant upfront loan-level fees to align treatment of loans scored with either model. That did not set identical FICO and VantageScore license or per-score prices; it aligned a different component of the mortgage transaction's pricing.

Classic FICO remains permitted and the change does not eliminate its entire business. It does shift a part of the market from reliance on one broadly required model toward the ability to compare two approved options. That changes what the incumbent may have to prove in price and performance.

Rocket Mortgage made model choice a real procurement decision

Rocket Mortgage announced on September 28 that it intended to make VantageScore 4.0 its preferred model for eligible direct-to-consumer loans during the fourth quarter of 2026. It described about four months of comparative testing and 1.4 million credit reports obtained using both models during the year.

According to Rocket's own analysis, the alternative helped some applicants meet criteria and reduced costs for some customers. The average $1,600 at-closing savings applied only to borrowers who saved with VantageScore, not to every customer or an independently established market-wide average.

The lender also specified that investment-property, second-home, home-equity, FHA, jumbo and other ineligible loans would continue using FICO, while its broker business would offer both models. This is evidence of an announced buyer decision in an eligible segment—not a claim that American mortgage lending has abandoned FICO.

FICO was already redesigning distribution before the new FHFA policy

On October 1, 2025, FICO announced the Mortgage Direct License Program to let authorized tri-merge resellers calculate and distribute FICO scores directly, rather than depending exclusively on the nationwide credit bureaus. Existing bureau distribution remained an option.

The company offered a $10-per-score flat model and a performance-linked alternative costing $4.95 per score plus $33 per borrower per score when a FICO-scored loan is funded. The performance model is not automatically the cheaper option for every lender: funding outcomes and score volumes matter to total cost.

This exposes two distinct procurement questions. Which credit model should a lender buy? And if it chooses FICO, through which channel and pricing contract should it purchase scores? FICO launched the program before the September 2026 rule change, so presenting it as a reaction to that later policy would reverse the chronology.

A second transformation is unfolding inside FICO's software portfolio

FICO's Software segment revenue rose only 2% to $215.3 million in fiscal Q3, but total software annual recurring revenue reached $816 million, up 10%. Beneath that aggregate, the product mix changed sharply.

Platform ARR increased 62% to $413 million. Non-platform ARR declined 17% to $403 million, putting the platform ahead for the first time. Management said the reduction in older products was driven mainly by customers migrating to FICO Platform and, to a lesser extent, end-of-life offerings.

That should not be mistaken automatically for external customer losses. It is a product migration initiated within the company, distinct from the regulatory opening of mortgage credit scoring to a competing supplier. Both involve customers changing what they use, but the reasons and economics differ.

What must survive the elimination of 15% of positions?

FICO's October 6 SEC filing says management adopted the plan on October 1 to reduce organizational layers, simplify its operating structure, optimize processes and tools, and integrate AI-driven product development. Notices began in the week of October 5, with substantial completion expected by fiscal Q3 2027 and approximately $27 million in pretax charges anticipated for fiscal Q4 2026.

The filing does not establish that VantageScore caused a particular number of cuts, or that the reshaping has already improved execution. Position count and product-development capability are not interchangeable measures. Determining whether critical judgment survives requires a different test.

In a multi-model mortgage market, relevant capabilities may include credit-risk performance validation across borrower groups, lender-system integration and audit trails, pricing and distribution economics, model governance, and migration of legacy software customers. These are Banseog's capability categories, not announced FICO vacancies or a forecast that specific teams will grow.

BANSEOG VIEW | The market standard must become the customer's choice

FICO generated strong scoring revenue when mortgage-related revenue expanded much faster than the underlying volume, reflecting higher unit prices. Since then, policy has made alternative model selection more available and a major lender has announced a preference shift. Whether the incumbent retains its historic pricing power remains to be observed.

At the same time, FICO has been changing its distribution structure since 2025, moving software ARR toward its new platform and planning company-wide position reductions. Their dates and stated drivers are different; they should not be collapsed into a single narrative of AI replacing staff or VantageScore directly causing layoffs.

The better corporate questions are: how much growth came from usage rather than price, what measurable advantage keeps customers choosing the model once alternatives are allowed, and which teams own the evidence needed to prove that advantage after restructuring? Market position and voluntary customer preference are not the same form of competitive strength.

BANSEOG VIEW

Banseog View — Pricing power, customer choice and evidence-owning capabilities

Mortgage-origination Scores revenue +97% despite low-single-digit volume growth is not the same population as overall Scores +41%; the former signals a meaningful price effect.

FHFA broadened VantageScore eligibility but did not ban Classic FICO; Rocket's preference was a future eligible-loan rollout announcement.

Approximately 15% refers to positions across FICO, not necessarily Scores employees. The SEC filing does not quantify a causal link between the layoffs and VantageScore; monitor model evaluation, integration, pricing and governance capabilities separately.

SOURCES

Primary sources and references

As of October 13, 2026. The fiscal Q3 period ended June 30, before FHFA's September lender-choice expansion. Scores +41%, B2B +49%, and mortgage-origination revenue +97%/low-single-digit usage refer to different scopes. The 15% position plan is across FICO, not a verified Scores division headcount reduction, and is not shown in the SEC filing to be directly caused by competition. September 30 upfront fee alignment is not a move to equal credit-score royalty pricing. Rocket's planned Q4 preference applies only to eligible loan products, and its $1,600 saving is a mean among beneficiaries. FICO's direct license program launched in October 2025, before the 2026 FHFA change. Non-platform ARR declines partly reflect internal migrations. Workforce and capability mapping is Banseog's inference rather than disclosed recruiting plans.