Sales Audit Questions
to Ask:
The Four Witnesses

The quality of a sales audit is decided entirely by the quality of its questions. Ask soft, flattering ones — "Are we following up enough?" "Do the reps feel supported?" — and you get soft, flattering answers that confirm what you already believed and change nothing. Ask sharp ones that can only be answered with a number or an uncomfortable truth, and the audit does its job: it surfaces the leak you have been working around. This guide is the questions, but it is more than a list. For each one it tells you who to ask, why it matters, and — most importantly — how to read the answer, because in an audit the answer you get is rarely as revealing as the way it is given. A hesitation, a round number with no source, a story instead of a figure: those tells are often the real finding.

The organizing idea is that you are interrogating four different witnesses, and each knows a different part of the truth. Your data knows what actually happened. Your reps know what happens on calls that the data never captures. Your lost customers know why they really did not buy, which is almost never the reason recorded in the CRM. And you — the founder or leader — know things about your own involvement that you have never said out loud. A complete audit questions all four, and cross-examines their answers against each other, because the gaps between what the data says, what the reps say, and what the customers say are where the most important findings hide.

The Principle: Ask What Can't Be Answered With Hope

Before the questions, the rule that makes them work: a good audit question cannot be answered with hope, opinion, or effort. "Do we follow up enough?" invites a comfortable yes. "What is the median number of touches before a deal closes versus before it dies?" demands a number, and the inability to produce one is itself the finding. Every question below is built to resist a hopeful answer — to force either hard evidence or an honest admission that the evidence does not exist. When you design or borrow audit questions, apply this test: if the question can be answered with "I think we're fine," it is the wrong question. Replace it with one that requires a figure, a documented fact, or a name.

There is a reason flattering questions are the default, and naming it helps you resist it: sharp questions create discomfort, and most people — including the person running the audit — would rather avoid it. Asking a rep to walk through a botched qualification, asking a founder whether the business depends on them, asking the data a question you suspect it cannot answer: each risks an awkward moment. But the discomfort is precisely the signal that you are pointed at something real. The questions that make the room go quiet are the ones earning their place in the audit. Train yourself to move toward that quiet rather than away from it, because the comfortable question and the useful question are almost never the same question.

Questions to Ask Your Data

Your data is the witness that cannot lie, only be missing. Ask it: What is the conversion rate at each pipeline stage, and where is the steepest drop? What share of pipeline is older than 1.5 times the cycle? What is the real win rate against qualified pipeline, by source? How often do deals slip their close date? What are the top lost-deal reasons by count, and what is the no-decision-versus-competitor ratio? The tell here is not just the answers but whether you can get them at all. If a question about your own funnel cannot be answered from your data within minutes, you have found a measurement gap — and a measurement gap is a place where problems live undetected, because you cannot fix what you never see.

Questions to Ask Your Reps (and How to Read Them)

Your reps know what the data does not capture — what actually happens on calls. But their answers must be read carefully, because reps narrate in ways that protect themselves and their deals. Ask: Walk me through how you qualified this specific deal. What did the buyer say was their reason for looking? What objection do you hear most, and exactly how do you respond? Why did your last three losses actually die? The tells matter as much as the content. A rep who cannot articulate the qualification on a live deal reveals a qualification gap. A rep whose objection-handling is improvised differently every time reveals an undocumented motion. A team whose members each describe a completely different sales process reveals that there is no process at all — just a collection of individual improvisations wearing the same job title.

Questions to Ask Your Lost Customers

The most underused witness in any audit is the lost customer, because they know the one thing everyone else only guesses at: why they really did not buy. The CRM almost always records the polite, convenient reason; the truth is usually different and far more useful. If you can reach a handful of recent losses, ask: What were you actually trying to solve when you took the call? What made you hesitate? Did you choose a competitor, build it internally, or simply decide not to act — and what tipped that decision? The pattern across even five or six honest answers is often the single most valuable finding in the entire audit, because it comes from outside your own narrative. Founders are frequently stunned to learn that deals they recorded as "lost to a competitor" were actually lost to indecision they could have broken with better urgency framing.

Questions to Ask Yourself

The hardest witness to cross-examine is the one in the mirror, and the questions you must ask yourself are the ones an internal audit reliably avoids. If you stepped out of sales for thirty days, what share of revenue survives? Are you the reason your best deals close — and if so, is that a strength you are scaling or a dependency you are hiding? Is your stated ICP the customer you actually want, or the customer you have been settling for? Are you avoiding raising prices because the data says no, or because you are afraid to test it? These questions point inward, which is exactly why they get skipped, and exactly why an outside auditor is often worth their cost: they will ask the founder-dependency and pricing-courage questions that no one inside the company will.

Cross-Examine the Witnesses Against Each Other

The deepest findings come not from any single witness but from the contradictions between them. When the data says deals die at proposal, the reps say "pricing," and the lost customers say "we never felt urgency," the truth is triangulated: a value-framing problem the reps have rationalized as price. When the founder believes the ICP is mid-market but the data shows the best customers are all small businesses, the targeting has drifted without anyone deciding it. When reps describe a crisp process the data does not reflect in consistent stage conversion, the process exists on paper but not in practice. An audit that asks all four witnesses and then lays their answers side by side will find what no single line of questioning could — because each witness is honest about a different part of the truth, and the gaps between them point straight at the problem.

The Answers That Are Red Flags

Certain answers, regardless of which witness gives them, are reliable signals that you have found a real problem — and they are worth recognizing on sight. "We don't really track that" is the most common, and it always marks a measurement gap where a problem can hide indefinitely. "It depends on the rep" usually means there is no documented process, only individual improvisation. "We close most of our good deals" is a non-answer dressed as confidence; ask for the number and watch whether one exists. "They went with a competitor" as the blanket reason for most losses is almost always hiding a no-decision problem the team finds easier not to name. And from a founder, "I just need to be in the room for the big ones" is the founder-dependency break stated as a virtue.

The pattern across these red flags is that each one converts a vague reassurance into a specific, testable claim the moment you push on it. The skill is not to accept the first answer but to ask the follow-up that demands evidence: "What's the actual number?" "Show me where that's written down." "How many of last quarter's losses were truly competitive?" The first answer tells you what the witness believes or wants you to believe; the follow-up tells you whether it is true. An audit that stops at the first answer is an interview; an audit that always asks the follow-up is a diagnosis.

When and How Often to Ask

The full four-witness interrogation is a periodic exercise, not a daily one — run it when something is clearly wrong, before a major investment like scaling headcount or raising a round, or on a roughly quarterly basis to keep the engine honest. But individual questions belong in the weekly rhythm: the data questions about slippage and aging should be asked every week in the pipeline review, the rep questions surface naturally in deal reviews, and the founder questions are worth revisiting whenever the team or deal mix changes. The lost-customer questions are the ones most worth scheduling deliberately, because they require reaching out and rarely happen on their own — yet they consistently deliver the highest-value answers, since they come from outside your own narrative.

However often you ask, the discipline is to ask the same sharp questions consistently, so the answers become comparable over time. A slippage rate or a no-decision ratio is far more informative as a trend than as a single reading — a number that is drifting the wrong way is a problem developing, even if its current value still looks acceptable. Asking the same questions on a rhythm turns the audit from a one-time snapshot into a moving picture, and the moving picture is what lets you catch a problem while it is still small enough to fix with a small change.

From Questions to a Fix

A pile of answers is not a diagnosis; the final question turns it into one. Having interrogated all four witnesses, ask: what is the single biggest leak these answers reveal, and what one change would most reduce it? Resist listing ten fixes — name the one, sequence the rest behind it, and define how you will know it worked. Then fix only that, hold everything else constant, and re-measure over a full sales cycle. The discipline that makes the questions worthwhile is the same that makes any audit worthwhile: the goal was never a thorough interrogation for its own sake, but one prioritized decision you can act on this quarter. Ask sharply, read the tells, cross-examine the witnesses, and end with a single fix.

FAQ: Sales Audit Questions to Ask