Draw decisioning should be a verdict, not a checklist

Draw review at most private lenders ends in an inbox. Someone looks at the upload, makes a judgment, and the wire goes out. The decision is not written down. The evidence is not packaged. The reasoning lives in a Slack thread or a phone call.

That works until it does not. The first time a loan goes into workout, the first time an investor asks for the audit file, the first time a draw admin leaves the company, the gap shows up immediately.

The case for a verdict

Every draw should produce a verdict in one of three states: pass, review, or fail. The verdict is computed from the verification signals (photo verification, budget reconciliation, attestation, prior-draw consistency) and the lender's own thresholds. The verdict is what flows downstream — to the LOS, to the wire instruction, to the audit packet.

What goes into the verdict

• Photo verification result for every uploaded image
• Budget line-item reconciliation against the original scope
• Borrower attestation, captured and stored at submission
• Comparison against every prior draw on the loan
• Override events, if any, with the reviewer's name and reason

Why review is the most important verdict

Pass and fail are easy to design for. Review is where the lender's expertise actually lives — it routes the draw to a human, attaches the specific signals that triggered the route, and asks for a decision. The override gets logged. That log is the audit trail.

What a credit committee will defend

A draw that funded because the system passed it, with the evidence attached, is defensible. A draw that failed and was overridden, with the reviewer and the reason on record, is defensible. A draw that funded because somebody glanced at it on a Tuesday is not — and the workout team will find out at the worst possible moment.

“The point of draw decisioning is not to eliminate human judgment. It is to make sure the human judgment is written down.”