What a small team can actually absorb.
Next Monday, every new appraisal report on a loan sold to Fannie Mae or Freddie Mac has to use UAD 3.6. Eight weeks later, FHA is expected to start taking two new credit score models. Somewhere in between, most of us are closing out the year.
Each of these is manageable on its own. All of them land on the same few people.
The near-term risk on the first one is practical: appraisals ordered now that won't reach the Uniform Collateral Data Portal before the 2nd. After that date, a 2.6 submission comes back rejected.
Large lenders have a project office for this. A mandate arrives, someone gets assigned, a vendor gets a ticket, and production barely notices. On a team of four, or a credit union mortgage department of ten, the people learning the new appraisal format are the same people working tomorrow's closings. There's no bench to pull from.
I've come to think about change capacity the way I think about capital. Each quarter there's a fixed amount of it. Every new system, rule, or workflow draws it down. When you overdraw it, the cost shows up somewhere you didn't plan for, like a missed condition or a close that slips a week.
Most organizations budget dollars carefully and budget change hardly at all. Leadership approves the new LOS module, the policy rewrite, and the regulatory deadline in separate meetings, and nobody adds them up. The team does that math, usually in November.
Adding them up is a leadership job. The people closest to the work will absorb whatever you hand them, because that's what good teams do, and they rarely say it's too much until something breaks. A leader who wants to run the function, and not just get through the quarter, has to make the trade-off visible before the team makes it quietly.
When we moved off spreadsheets at Cook Inlet, the sequencing mattered more than the software. Staff-facing changes went first, so the team learned the tools before a borrower ever saw them. Borrower-facing changes came second. Tuning came last, after we'd lived with the system for a while. That order is the Build phase of the framework I built for how mission-driven institutions modernize without losing what makes them matter, and it's a big reason the rollout stuck.
So here's the question I'd put on the next leadership agenda. What are we asking this team to absorb between now and March, written as one list? If the list is longer than the team can carry, something has to move. I'd rather choose what waits than find out in a file review.
WHAT I'M WATCHING
FHA's credit score change is next in line.
For case files dated on or after January 1, 2027, FHA is expected to accept VantageScore 4.0 and FICO 10T alongside Classic FICO, according to HousingWire. HUD has described it as an end-of-year launch, so watch for the final guidance.
Take: For credit union and community bank lenders, start this one now. It touches your LOS and your credit vendor at the same time, and it arrives right behind UAD 3.6.
The CDFI Fund's FY27 number.
The House Appropriations Committee's FY27 bill set the Fund at $276.6 million, $47.4 million below FY26, per OFN. The Senate has historically recommended more.
Take: Any 2027 plan that leans on federal dollars deserves two versions, one at this year's level and one at the House number. Layered state and FHLB capital is what keeps the second version from turning into a smaller program.
FROM THE FIELD
What I learned running a rollout with a four-person team is that the first phase should be boring. Nothing a borrower could see went live until the team had used the new tools on real files and told us what was broken. Some fixes were configuration. Others were decisions to keep a step manual on purpose, because automating it would have cost more training time than it saved. Testing on real files also meant the team watched the system answer problems they had named themselves, which did more for adoption than any training session.
The harder lesson was saying no to good ideas mid-rollout. A new report or a nice-to-have integration is reasonable on its own. Stack enough of them and a phase you planned in weeks stretches into months.
I'm curious how other teams handle this. Do you keep one list of everything your lending team is absorbing, and who decides what waits? If you've found a way to make that list visible to a CEO or board, I especially want to hear it. Hit reply. I read every one.
|
If this was useful, forward it to someone building in this space.
