Cutting our approved lender list from 38 to nine for volume tiers, or keeping the wide shelf

Pulled our placement data for the last four quarters because a couple of our private lenders started dangling tier pricing and I wanted to know what we'd actually be giving up.

38 active approvals. 71 percent of closed volume went to six of them. Another 19 percent went to five more. That leaves 27 lender relationships that between them did about 10 percent of our volume, and every single one of those relationships still costs us something, annual renewals, portal logins nobody remembers, guideline updates my two loan officers are supposed to read and don't.

The case for cutting. Two of the six big ones will move us up a comp tier at volume, roughly 25 to 40 basis points better on our side, plus a named contact who answers instead of a shared inbox. Fewer lenders means my team actually knows the boxes cold and stops sending files that were never going to fit. Turn times improve because we're a name to them.

The case for keeping the shelf. That 10 percent tail is where the odd files went. Rural properties, a mixed-use file with a weird occupancy story, two files where the borrower's entity structure scared everyone else off. Those are the deals that make people call us instead of a portal, and half of them came in as referrals from a borrower we saved once. If I cut to nine, the eventual weird file walks and I don't get the referral either.

There's also concentration risk. Two of my top six changed their box materially in the last year with about a week of warning. If they'd both been half my volume I'd have had a bad quarter.

How are people actually running the list.