The F&I sales training industry has been selling the same playbook since about 2004. The core idea is a script with objection handling, mapped against a standardized menu of products, coached toward a PVR number.
It made sense in 2004. In 2026 it is actively teaching F&I directors the wrong priorities.
What changed
Three things, in this order.
Customer information density. A customer walking into F&I in 2004 was mostly reliant on the F&I director to explain what they were buying. A customer walking in today has already Googled the product, priced alternatives online, read Reddit threads, and formed an opinion. The "teach them, then close them" script assumes asymmetric information that does not exist anymore.
Lender posture. The profit levers in 2004 were GAP, service contracts, and a fat reserve on the finance rate. The fat reserve is mostly gone. Most manufacturer captives cap dealer reserve at 1–2 points. You cannot build a shop around reserve anymore. You have to build it around products that actually match the customer.
Compliance posture. The FTC Safeguards Rule and state-level predatory-lending statutes have tightened what you can say and how you can say it. A script written in 2004 is going to walk a junior F&I director into language their state AG does not like.
What replacements look like
The new F&I playbook is customer-matched, not product-menu-matched.
For a 72-month deal with a 580 FICO buyer in a 2020 used SUV, the product set that fits is narrower than the full menu. GAP is almost always right. A tire-and-wheel is usually right in the Midwest. A service contract has to be priced against the vehicle's actual reliability data, not the generic menu price.
For a 36-month new-car lease with a 720 buyer, GAP is already baked in. A tire-and-wheel is a harder sell. A pre-paid maintenance is a narrower upsell, and the right one to lead with.
A script cannot do this. A script is one script. The matching has to be per-deal.
Where the AI actually helps
This is the work a model is good at. You give it the deal — vehicle, FICO band, term, borrower profile — and ask for the three products most likely to close clean, in order, with the customer-matched reasoning attached.
Not the script. The reasoning. The F&I director still delivers it. But now they are delivering a reasoning that fits this customer rather than a script that fits "customers in general."
The difference shows up in cancellation rates. Products sold with customer-matched reasoning cancel at a fraction of the rate of script-sold products. That is the PVR metric F&I training should be optimizing and generally is not.
The number
In the operator deployment, we watched F&I product cancellation rates drop roughly 35% after we swapped the standardized menu for customer-matched recommendations. The top-line PVR number did not move as much as an F&I trainer would have predicted. The net PVR after cancellations moved considerably.
That is the number your DMS does not report and your F&I trainer does not care about. It is also the number an acquirer's finance team cares about, because it is the one that shows up on the P&L three months after the sale.
The F&I vendors who are still selling scripts are going to find that out the hard way when the first big dealer group runs the cancellation math. Probably in 2026.