Renovation budgets are where I spend the most time before a commitment letter goes out, and the borrower's number is almost never the number I lend against.
The standard workflow is a third-party scope review, usually a licensed contractor or a construction consultant the lender controls the relationship with, not someone the borrower brought. That person walks the property before closing and prices the work line by line: demo, framing, mechanical, roof, finishes. If the borrower says $48,000 and the reviewer comes back at $71,000, the loan either reprices or dies. The gap tells you something about the borrower's competence or their honesty, and either answer matters before you fund.
What most lenders underweight is the contingency conversation. A borrower who builds zero contingency into a 1960s ranch is telling you they have never opened a wall in a 1960s ranch. Knob and tube wiring, galvanized supply lines, undersized electrical panels, and pier-and-beam foundation drift are not surprises in that vintage, they are probabilities. A realistic contingency on a full renovation in an older house runs 10 to 15 percent of hard costs at minimum, and that number belongs in the budget before the loan closes, not in a draw request six weeks in when the cash is already short.
The assumption doing the most work in a renovation loan is that the borrower can finish the project, not just start it. The budget review is one half of that question. The other half is whether the borrower has liquidity outside the loan to cover overruns without coming back to you for a modification. A borrower who is 100 percent dependent on draw proceeds to fund every dollar of construction has no cushion between a delayed inspection and a contractor who walks.
The number I always want before I commit is the borrower's out-of-pocket capacity past the loan amount. What is theirs to lose if the project runs 20 percent over budget and the draw schedule runs three weeks behind?