Deal Review Questions Every Sales Manager Should Ask (and How AI Is Changing Them)

The best deal reviews run on six to ten fixed questions, agreed with the team in advance, asked on every deal, every time, covering qualification, stakeholders, competition, and forecast confidence. This article gives you the strongest deal review questions in each category, annotated with what a good answer sounds like, what a bad answer reveals, and the follow-up that does the real work. It also covers something most question lists ignore: AI now answers the factual half of these questions before the meeting starts, which changes what the human half of the session is for. If you’re a sales manager choosing your team’s question set, or a rep who wants to know what’s coming, this is the question bank, with the judgment layer attached.

The two questions I never skip, and the reason this whole page exists: “Who has veto power that we haven’t met?” and “Did we hear that from them, or about them?” Everything below is, in some sense, an elaboration of those two. (For the process these questions plug into, start with the complete guide to running deal reviews.)

How Should You Structure Deal Review Questions?

Not as a forty-question interrogation. The structure that works: pick six to ten with your team, mapped to how your deals actually die — if you lose to unmapped stakeholders, weight the stakeholder set; if you lose competitively, weight that one. Publish the questions in advance and never rotate them without advanced notice. When the team knows exactly what’s coming, they hunt the answers all week, and the questions do their work before anyone joins the meeting. Ad hoc questions produce ad hoc preparation.

Then honor the don’t-dwell rule: when a rep doesn’t have an answer, don’t excavate, flag it, assign it as a task, move on. “I don’t know” must always cost less than a confident bluff, or your sales organization will be systematically bluffed. The gap is the finding, and it becomes the action item for the next session.

One framing rule above all: questions test the deal, not the seller. The moment questions become gotchas, people learn to sanitize, and management hears what it wants to hear — which is worse than hearing nothing. A deal review is a coaching conversation with evidence, not a deposition.

Which Two Questions Belong in Every Deal Review?

• “Who has veto power that we haven’t met?” Approvers and veto holders are different lists. The people who own the purchase can say yes; the people around them, a CIO with a platform-standardization agenda, a CFO with a spend freeze, a procurement lead with a preferred-vendor policy, can say no from outside the room. I learned this on a seven-figure deal we lost at final-two, when a CIO nobody had mapped surfaced an objection that had existed, unasked, for six months (the full story opens the main guide).
• “Did we hear that from them, or about them?” Every reassuring claim in a deal record — “they’re comfortable with it,” “budget’s approved,” “the economic buyer is on board” — is either firsthand or secondhand, and the difference is the difference between evidence and hope. Secondhand comfort is how that CIO stayed invisible: our champions told us everyone was fine with our architecture. Nobody had heard it from the man who wasn’t.

If your team adopts nothing else from this page, adopt these two. They take thirty seconds in every deal review, and in fifteen years of B2B sales I’ve never run a session where they didn’t earn it.

What Qualification Questions Should You Ask?

Qualification questions establish whether there’s a real sale underneath the activity — a prospect with a problem, a motivation to act, and a reason to act now.

• “What problem is the prospect solving — in their words, not ours?” Good: the rep quotes the customer’s language from a specific call. Red flag: the answer arrives in your product’s messaging — projection, not validation. Follow-up: “When did they last articulate the problem themselves, unprompted?”
• “What happens if they do nothing?” The status quo is every seller’s biggest competitor. Good: a concrete cost — lost revenue, compliance exposure, a team drowning in workarounds. Red flag: “they know they need to solve this.” Deals without a do-nothing cost lose to no-decision more often than to any rival. Follow-up: “Has anyone on their side said that out loud, or is it our math?”
• “Why now — what triggered this evaluation?” Good: a nameable event — a leadership change, a contract expiry, an incident, a strategic initiative. Red flag: no trigger. Evaluations without triggers drift, because nothing forces a decision by any date.
• “What evidence supports budget — and who releases it?” Good: the buyer has named the budget line, its owner, and the approval path. Red flag: “they wouldn’t be talking to us otherwise.” They would, and they do. The follow-up connects to the stakeholder set: has the economic buyer confirmed intent to spend, or only intent to evaluate?
• “What’s their decision criteria — and who wrote them?” Good: the rep knows each criterion and its origin. Red flag: criteria that read like a competitor’s feature list, a strong signal someone else is shaping the evaluation. Follow-up: “Which criterion actually decides it?”

Which Questions Reveal Who Really Decides the Sale?

Stakeholder questions exist because B2B buying committees routinely run six to ten people, and the sale is decided by the map, not the org chart.

• “Who signs the agreement — and when did we last speak with them directly?” Good: a name and a recent date. Red flag: a name and “our champion handles that relationship.” Follow-up: “What would it take to find time to meet them before the next stage — can our champion introduce us, or does the introduction need to come exec-to-exec?”
• “Whose job gets harder if we win?” The question almost nobody asks. Every purchase creates a loser somewhere — the owner of the incumbent software, the team whose workflow changes, the person selling the alternative internally. Good: a name and an engagement plan. Red flag: “nobody, everyone wants this.”
• “Who on the committee is unengaged — and what’s the plan for each?” Good: a stakeholder map with a not-engaged column and actions against it (the five-column map is in the prep guide). Red flag: engagement defined as “was on a call once.” Communication with one enthusiastic contact is not committee coverage.
• “What has our champion actually done to prove they’re a champion?” Good: evidence of spent capital — meetings brokered, our case argued in rooms we’re not in. Red flag: champion defined as “the person who likes us.” A contact who takes your calls is a friend; a champion takes risks. Follow-up: “If our champion left tomorrow, does this deal survive?” The pause is usually the answer.
• “How does this committee approve a purchase — consensus, sponsor-driven, or procurement-led?” Good: the buyer walked us through the mechanics, including legal and security gates and their typical timelines. Red flag: assuming the org chart is the decision process. They rarely match.

What Competitive Questions Help You Win the Deal?

Every deal with a named competitor should get its own session — that’s the pillar’s non-negotiable trigger — and these questions structure that segment.

• “Who else is in this deal — and how do we know?” Good: evidence, not vibes. Red flag: “no one,” without a probe — buyers rarely volunteer the competition. Follow-up: “Anything in the prospect’s language or requirements that sounds like someone else’s messaging? Objections arriving strangely well-formed?”
• “Why is the buyer considering them?” The highest-leverage competitive question, because the answer reveals how the customer has framed their problem — a buyer evaluating a point solution is solving a different problem than one evaluating a platform, and if that framing shifts mid-deal, the deal you thought you were winning stops existing.
• “What’s our differentiator for this buyer specifically — validated with the buyer?” Good: one axis, confirmed. I once won a deal in Japan against point solutions that looked stronger on paper because we found the one criterion that decided it — global reach for a customer expanding across Asia — and made the whole strategy about it. Selling competitively isn’t being better everywhere; it’s finding the axis that decides this deal.
• “What will this competitor most likely do next — and what’s our counter?” Good: a stage-specific prediction from the battlecard: their late-stage discount play, their proof-of-concept move, their exec fly-in. Red flag: no prediction — you’re playing them reactively. This is where a current battlecard converts the conversation from memory into accumulated team knowledge; it’s the loop we built Playwise HQ around.
• “What did the last deal against this competitor teach us — and are we using it?” Red flag: silence — which means everyone fights this competitor from zero, and your wins are heroics rather than a system.

What Forecasting Questions Keep the Pipeline Honest?

Forecasting questions separate evidence from momentum — they’re how a deal review protects the whole pipeline’s credibility, one deal at a time.

• “What’s the path to close, step by step — and does the buyer agree with it?” Good: a mutual action plan the customer has edited or confirmed, with dates and an implementation timeline on the far side. Red flag: a plan that exists only in our CRM. A close plan the buyer hasn’t seen is a hope with milestones.
• “Is the close date the buyer’s date or our date?” Good: the date anchors to a buyer-side event — a contract expiry, a fiscal deadline, a launch. Red flag: it anchors to our quarter-end. Follow-up: “What did they say about timing, word for word?”
• “What evidence puts this deal in its forecast category?” Good: each load-bearing claim tagged verified or assumed (the seven-factor scorecard is in the main guide) — that’s how you assess the probability of a deal closing accurately. Red flag: “it’s been commit for two months.” That’s a description, not evidence.
• “What’s most likely to cause a slip — and what are we doing about it this week?” Good: a named risk with a live action. Red flag: “nothing, it’s on track.” Deals genuinely on track can still name their most likely slip cause; deals that can’t haven’t been examined. Add the gate check: “Has procurement, security, or legal been scoped — or are we treating them as a formality?” That’s where close dates go to die.

Which Questions Uncover Hidden Deal Risk?

Beyond the standing set, three inversions surface what direct questions miss:

• “If we had to disqualify this deal today, what would the reason be?” Good: a specific, known soft spot. Red flag: “we wouldn’t.” Every deal has a disqualification case; not knowing it means not knowing the deal.
• “If this deal slips, what will we say the reason was?” The pre-mortem in one line. The honest answer usually names the real risk faster than any direct question — and it’s remarkable how often the room already knows, and how rarely anyone had asked.
• “What are we assuming that we’ve never actually checked?” The question that would have saved my logistics deal. Ask it last, leave a silence, and let it be uncomfortable. The first answer is usually deflection; the second answer is usually the finding — no deal review should end in a surprise that this question could have surfaced.

And one listening question, triggered by the buyer’s language rather than a schedule. When a prospect says “we can accommodate that,” “that’s workable,” “shouldn’t be a problem,” ask: “What would their preferred version look like — and what happens down the line if preferred doesn’t happen?” Accommodate is the vocabulary of doable-but-not-preferred. The deal that taught me: a customer told us for six months they were “comfortable” with our cloud architecture, and at final-two their CIO explained what comfortable had actually meant — technically workable today, a permanent exception to their operating model tomorrow. If the buyer doesn’t volunteer what preferred looks like, the deal isn’t as safe as the notes suggest.

How Is AI Changing Deal Review Questions?

AI hasn’t replaced deal review questions — it has split them in two. There’s a factual layer (what happened, who’s engaged, what was said) that AI now answers before the meeting, and a judgment layer (what does it mean, what do we do) that stays human. Sales managers who understand the split run shorter, sharper sessions; those who don’t spend meeting time asking questions AI answered yesterday.

Questions to ask your AI before the session, not your rep during it: What changed on this deal in the last two weeks? What concerns did the buyer raise on calls that aren’t reflected in the CRM? How does engagement look across the committee — who’s gone quiet, whose email response times are stretching? What’s the volume and recency of activity relative to deals that actually closed? AI deal summaries reconstruct the story from transcripts and email threads in real time, and AI-based scoring can predict the probability of a deal closing from signals no human tracks manually. Let AI do that archaeology; it’s better at it than anyone in the room.

What AI hands the human questions: with the factual layer pre-answered, the session’s questions move up a level. AI can flag that the economic buyer hasn’t been on a call in five weeks; only a human can ask why, and decide whether the fix is an exec introduction or a champion play. AI can detect that a competitor’s name appeared on a transcript; the “why is the buyer considering them?” conversation is judgment. The guidance I give teams adopting AI tools: every question AI can answer gets asked before the meeting; every question that requires deciding something gets asked in it.

What AI cannot ask. The CIO objection that killed my seven-figure deal would never have appeared in any AI summary, because it never appeared in a transcript — nobody had asked the question, so there was nothing for AI to find. AI answers what was said. The best deal review questions — the veto question, the from-them-or-about-them test, the never-checked assumption — surface what was never said. That’s the layer no software reaches, and it’s precisely where a sales manager earns their seat in the room.

What Questions Should Sales Managers Ask Themselves?

The sets above test the deal. These five test the review — the manager’s own weekly self-check:

1. Did I read the transcripts (or run the AI summary) before the meeting, or am I about to spend it on archaeology?
2. Did I choose the one load-bearing assumption to pressure-test, or am I planning to wander?
3. Am I testing the deal or testing the rep — and would the rep agree with my answer?
4. Did “I don’t know” cost anyone anything today?
5. Did what this deal taught us — about a competitor, an objection, a stakeholder pattern — get captured anywhere the next seller will find it?

If the answer to the last one is consistently no, your sessions are working single deals and building nothing. That’s the difference between a review habit and a review system — and honestly, between a manager who inspects and a manager who coaches. (How to facilitate the session itself, script by script, is here.)

What Do Red-Flag Answers Sound Like in the Meeting?

A Starter Set of Questions to Ask (Copy This)

If your team is choosing its first fixed set today, start with these eight and adapt:

1. What’s changed since last session?
2. What problem is the prospect solving, in their words — and what happens if they do nothing?
3. Who signs, and when did we last speak with them directly?
4. Who has veto power that we haven’t met?
5. Who else is in this deal, why is the buyer considering them, and what’s our counter?
6. What’s the path to close, step by step — and has the buyer agreed to it?
7. What’s most likely to cause a slip, and what are we doing about it this week?
8. What are we assuming that we’ve never actually checked?

Key Things to Remember

• Six to ten fixed questions, chosen with the team, published in advance, never rotated — consistency is what makes reps hunt answers before the meeting.
• Questions test the deal, not the rep. “I don’t know” must cost less than a bluff, or you’ll be bluffed.
• The two never-skip questions: who has veto power we haven’t met, and did we hear it from them or about them.
• Approvers and veto holders are different lists; the sale is decided by the map, not the org chart.
• The status quo is the biggest competitor — every deal needs a do-nothing cost in the customer’s own words.
• Competitive questions decide deals on one validated axis, not a feature war — and what each deal teaches must be captured for the next one.
• The close date is only real if it’s the buyer’s date, and the close plan only real if the buyer has agreed to it.
• Ask AI the factual questions before the meeting; save the judgment questions for the humans in it. AI answers what was said — the best questions surface what was never said.
• End every session with the never-checked-assumption question, and let the silence sit.

Frequently Asked Questions

• What questions should you ask during a B2B deal review? Six to ten fixed questions covering qualification (the problem, the trigger, the budget), stakeholders (who signs, who vetoes, who’s unengaged), competition (who’s in, why, and the counter), and forecast confidence (the path to close and the evidence behind the category) — asked on every deal, every time.
• What are the best questions every sales manager should ask? The two never-skip questions — “Who has veto power that we haven’t met?” and “Did we hear that from them, or about them?” — plus the closing inversion, “What are we assuming that we’ve never actually checked?”
• What review questions uncover deal risk? The inversions: “If we had to disqualify this today, why?”, “If this slips, what will we say the reason was?”, and the soft-yes probe for any “we can accommodate that” language in the notes.
• What questions identify buying committee gaps? “Who has veto power that we haven’t met?”, “Whose job gets harder if we win?”, “Who is unengaged and what’s the plan for each?”, and “If our champion left tomorrow, does this deal survive?”
• How does AI change deal review questions? AI answers the factual layer before the meeting — what changed, who’s engaged, what the buyer said that never made the CRM — so the session’s human questions move up to judgment: why, what it means, and what to do next. AI can’t ask about what was never said; that remains the manager’s job.
• What forecasting questions should managers ask? Whether the close date is the buyer’s date or ours, whether the buyer has agreed to the path to close, what verified evidence supports the forecast category, and what’s most likely to cause a slip this week.