Issue No. 17 | July 28, 2026

Intelligence for F&I agents, administrators, and dealership finance professionals.

THE VEHICLE MARKET NUMBER: 25%

Per Dark Sky Data’s cohort-based retention modeling, at an 8% annual cancellation rate, a portfolio of contracts turns over roughly a quarter of itself every three years just to hold flat, before any new volume is added. Applied to VSC and GAP business, that same compounding dynamic provides important context when finance companies evaluate reserve rates and renewal terms on CLIP and participation programs.

A single blended cancel rate across your whole book can look stable while hiding meaningfully different behavior by vehicle type, term, and vintage, exactly the kind of composition that can become important during a finance company’s renewal review.

THE F&I ANGLE

A blended cancel rate tells a finance company what happened to your book. It does not tell them why.

Two dealer groups can show the same blended cancellation rate and mean very different things for reserve risk. One has consistently durable contract performance across its existing vintages. The other has a wave of new volume masking rising cancellations in older vintages. Segmenting cancel and loss data by vehicle type, term, and origination vintage—the same cohort logic private equity uses to separate real customer persistence from masked churn—shows a finance company which story is true for your program.

As finance companies apply more scrutiny to program reserves across the credit cycle, dealer groups bringing that cohort-level detail into renewal discussions are negotiating from a stronger position than those relying on a single blended number for the whole book. That detail also travels well beyond a single renewal conversation, since the same cuts support pricing decisions on new product structures and F&I menu changes.

FROM THE BLOG

Customer Retention Metrics: Beyond the Headline Numbers

The post is written for private equity diligence, but the underlying cohort framework applies directly to the way finance companies evaluate a program’s cancellation performance. A blended number hides whether the book is genuinely holding or being propped up by new volume. Segmenting cancellation and loss data by cohort answers that question the same way account-level classification does in the post.

THE DESK STAT

At 8% annual cancellation, a quarter of a book turns over every three years by default, before any growth target. Do you know your program’s cancel rate by vehicle type and vintage, or only for the book as a whole?

Until next Tuesday,

Reply here and tell me whether your last program renewal used cohort-level cancel data or a single blended number. I read every response.