Maybe it’s an inconsistency in the management presentation. A founder who hedges on a question no one else noticed. A number that doesn’t reconcile with the narrative. A data room that’s missing documents that should be obvious to include.
Too junior, and raising it risks being seen as obstructionist. Too senior, and stopping the deal requires fighting a committee that’s already committed to the thesis. Either way, the deal has momentum. The room is excited. You have a feeling. Not a proof.
This is the scenario where most bad deals survive. Not because the evidence wasn’t there. Because the person who sensed it didn’t have a vehicle to surface it without making it personal, and didn’t want to be the one who killed the deal on a hunch.
The worst deals are not the ones that fail in diligence. They are the ones that pass diligence and fail in year two. By the time the market understands what happened, momentum had already overridden the judgment of the one person who saw it coming. The IC heard the thesis. Capital was committed. Teams were mobilized. At that point, a finding is no longer an opportunity to reprice. It is an obstacle to manage.
The architecture of that problem has a solution, and it is not being louder in the next committee meeting.
You authorize a 72-hour diagnostic. Not from the diligence budget. Out of discretionary. No committee approval, no procurement chain, no questions asked about what you’re looking for. The protocol runs quietly. The evidence window closes. You get a memo.
The doubter on the deal team has one job: not to be right in retrospect. Anyone can be right in retrospect. The question is whether you have a mechanism to surface what you know before the price is set and the capital is committed.
No career risk. No confrontation. No committee approval needed. A 72-hour window while the leverage still exists to act on what it finds.
The only losing move is saying nothing and being right.