IC questions cluster into a small set of recurring categories, with customer durability raised more often than anything else. Deal teams routinely prepare detailed market sizing that is rarely queried.
What does an IC actually ask?
Fewer things than a memo contains, and more consistently than deal teams expect. Committees develop habits, and those habits cluster around the assumptions that have hurt them before.
Customer durability sits at the top of that list in most houses, because a revenue base that leaves after close is the failure mode that is hardest to recover from.
The habits are house-specific and knowable. Anyone who has taken three deals through the same committee can list the questions that will come, and that list is worth more to a deal team than a generic diligence checklist. Where a team is presenting for the first time, asking the partner sponsoring the deal what the last three memos got asked is a five-minute conversation that reshapes a fortnight of work.
Which questions recur across deals?
Customer durability, the repeatability of the growth rate, management capability, and what happens if the base case does not arrive. Sector-specific questions sit underneath those and vary, which is what the diligence scope should follow.
Knowing the cluster in advance is worth more than preparing comprehensively, because it directs effort at the questions that will actually be asked.
Management capability is the question with the least reliable preparation. It is usually addressed with a reference call arranged by the seller and a page describing the team's tenure, neither of which tells a committee whether these people can run the plan being underwritten. Where the plan depends on a capability the incumbent team has not demonstrated, that is a question worth answering before the meeting rather than during it.
What prepared material never gets used?
Detailed market sizing, most often. It is the section deal teams invest most in and the one committees query least, partly because everyone assumes it is directionally right and partly because it rarely changes a decision.
That is not an argument for skipping it. It is an argument for not spending the marginal hour there.
The pattern holds because sizing is a question the committee has already delegated. They assume the number is directionally right, they assume the deal team has not misled them, and they know the decision does not turn on whether the market is eleven billion or nine. What it turns on is whether this company holds its share of it, which is a different workstream entirely.
What kind of evidence lands?
Specific and attributable: what a named category of person said, with a sample and a date. Aggregated market data rarely shifts a view because the committee assumes it is broadly correct anyway.
A single quoted customer with the context of who they are and when they said it does more work than a page of survey output.
Attribution is what makes a finding usable under challenge. One that can be traced to a category of person, a sample and a date can be defended when someone asks how confident the team is. An unattributed finding cannot, and a committee's response to an unattributed claim is not to reject it but to discount it — which is worse, because nobody says so out loud.
How should you present a gap?
Named, with what it would take to close it — usually a stated number of further conversations — and what the team believes in the meantime. Committees accept a known gap far more readily than they accept discovering one during questions.
Presenting a gap also protects the rest of the memo, because it signals that the absence of a caveat elsewhere is meaningful.
The corollary is that a memo should distinguish between what was tested and what was assumed. Most read as though every claim carries the same weight, which forces the committee to guess. Marking the two or three claims tested against primary evidence, and saying plainly which ones were not, gives the discussion somewhere to go other than a general debate about confidence.