DealWise AI
Use case

The cheapest mistake is the one you find yourself

These errors are cheap to fix while the model is still yours and expensive once somebody else has found them — because what they cost is not the error, it is the credibility of everything beside it.

The checks

None of these look like mistakes once the model is built

That is what makes them worth an automated pass. They do not announce themselves — they quietly move a number you are about to defend in a room where you cannot recheck it.

  • Rent that will not reconcileIn-place rent disagreeing with the suite's own stepped schedule means one of the two was edited and the other was not.
  • Names and dates that cannot be rightTwo suites sharing a name is how square footage gets attributed to the wrong one. An expiry before a commencement is arithmetic that was never possible.
  • Costs on the wrong side of NOIThe expensive category, because NOI times a cap rate is the price and the error propagates straight into it.

One category costs real money and the rest cost embarrassment

A duplicated suite name is awkward. A date that cannot be right is awkward. Both get corrected in a follow-up email and life continues.

The NOI line is different, because NOI multiplied by a cap rate is the valuation. At a 7% cap, every $10,000 sitting on the wrong side of it moves the supportable value by roughly $143,000. A roof replacement booked as an operating expense quietly argues the building is worth less than it is.

The reverse is worse and nobody checks it. Recurring maintenance pushed below the line inflates NOI, which inflates the value — and errors that flatter you are the ones that survive review, because nobody goes looking for good news that is wrong.

Where it starts

Classification happens when the statement is read

Most of these findings are decided long before the audit runs. If the expense lines were classified correctly on the way in, there is very little for the audit to catch.

  • Capital items separated on arrivalPulled out when the operating statement is read, and placed below the NOI line rather than wherever they sat on the seller's version.
  • Every line carries its sourceSo a finding points at the document it disagrees with, rather than just flagging a cell and leaving you to work out why.

Questions people ask

What errors recur most in underwriting models?

Suites named after the property rather than the unit, in-place rent that does not reconcile to the suite's own stepped schedule, dates that cannot be right, expenses on the wrong side of the NOI line, and a missing replacement reserve. Most findings are one of those five.

Why does the NOI line matter more than the others?

Because NOI multiplied by a cap rate is the price. A capital item booked as an operating expense understates NOI and therefore the value it supports — at a 7% cap, every $10,000 on the wrong side moves supportable value by roughly $143,000. The other errors cost you an awkward email; this one costs you the valuation.

Is this checking the seller's numbers?

No. This checks the model you built — the one about to go to a lender or a partner. The stated-versus-derived cap rate check is the one that examines what the seller claimed, and they are deliberately separate.

Does it stop me continuing if it finds something?

No. It reports and you decide. Plenty of findings are deliberate — a model legitimately carrying no reserve because the buyer handles it separately, for instance — and a tool that blocks on a judgement call is one people learn to click past.

When should I run it?

Before the package goes out. The cost of one of these errors is not the error itself, it is what finding it does to everything you presented alongside it. A lender who catches a reconciliation error in your rent roll starts checking the rest of it.

Run it before the package goes out.

Build the deal from your documents, then check your own model against the mistakes that recur. It reports what it found and points at the source it disagrees with. Free plan, no credit card.

Find Underwriting Errors Before Your Lender | DealWise AI