How to run a pipeline review that actually moves deals

Monday, 9:00. Eight reps, one manager, a shared screen showing a pipeline sorted by close date. The manager works down the list. Each rep narrates their deals. The manager nods, occasionally asks “what's the next step?”, and gets an answer that sounds fine. Ninety minutes later everyone leaves and not a single deal has changed state.

That is a status meeting wearing a pipeline review's clothes. It is the most common recurring meeting in B2B sales and one of the least productive.

If you are a first-time sales manager, or you inherited a pipeline review you did not design, this is the playbook: why the standard format fails, a 45-minute agenda that does not, the five questions worth asking per deal, and what to do when the answers are bad.

Why most pipeline reviews fail

There are three failure modes, and most teams run some blend of all three.

• The roll call. Every deal gets equal time, so nothing gets enough. A 40-deal pipeline in a 60-minute meeting gives each deal 90 seconds, which is enough to confirm the deal exists and nothing else. Coverage feels thorough and produces no decisions.
• The interrogation. The manager treats the review as forecast defence. Reps learn that the way to have a short meeting is to sound confident, so they optimise for confident narration. Bad news arrives later and larger.
• The therapy session. The meeting becomes a discussion of how hard the deals are. Genuinely sympathetic, genuinely useless. Nobody leaves with an action that has a name and a date on it.

The common root: the meeting is organised around reporting rather than around deciding. A pipeline review should produce a small number of changed decisions — a deal reclassified, a stakeholder approached differently, a stalled deal killed. If the pipeline looks exactly the same after the meeting as before it, the meeting was theatre.

The 45-minute agenda

Forty-five minutes, weekly, same slot, no exceptions. Longer meetings do not inspect more deals; they inspect the same deals more slowly.

1. Changes only (5 minutes). Not a walkthrough of the pipeline. Only what moved since last week: deals that entered commit, deals that slipped, deals that died. If nothing moved, that is the finding, and it deserves ten seconds of silence rather than a recap.
2. Deal inspection (25 minutes). Four to six deals, chosen by the manager before the meeting. Roughly four minutes each. See the five questions below.
3. Pattern and coaching (10 minutes). One theme across the deals inspected — a common objection nobody is handling well, a stage where deals consistently stall, a buyer role the team keeps failing to reach. One theme, not five.
4. Commitments (5 minutes). Written, out loud, with owners and dates. Not “I'll chase them” — “I will send the security questionnaire to Mette by Wednesday and ask her to confirm the approver.”

Notice what is not on the agenda: forecast roll-up. Do that asynchronously, in the CRM, before the meeting. Spending live team time reading numbers aloud is the most expensive way to distribute information that a dashboard already distributes for free.

Five questions to ask about every deal you inspect

These are ordered deliberately. The first two disqualify faster than the last three, and disqualifying early is the entire economic value of a pipeline review.

1. What did the buyer commit to, in their own words, and by when? The answer must contain a name, a verb, and a date. “They're keen” is not an answer. If the only committed party is your rep, the deal has no momentum, regardless of how good the last call felt.
2. Who signs, and have we spoken to them? Not “who is the decision maker” — who physically signs, and what is the path from your champion to that signature. Deals do not die at the demo; they die in the fifteen metres between the champion and the person with authority.
3. What is the buyer's cost of doing nothing? If your rep cannot state it in one sentence using the buyer's numbers, the deal is competing against inertia and inertia usually wins. Note that “no decision” is the most common competitor in B2B and it never appears in the CRM as a competitor.
4. What is the one thing most likely to kill this deal? Ask the rep to name it. Reps almost always know. The purpose of the review is to make that knowledge sayable in a room, which requires the answer not to be punished.
5. What changed in the account since we last looked? New stakeholder in a thread, a reorganisation, a competitor named for the first time, a procurement step nobody mentioned in month one. We catalogued the specific things worth listening for in five pipeline signals hiding in your sales conversations.

Red flags worth stopping the meeting for

Some answers should change what happens next, immediately, rather than being noted and moved past. In our experience these five are worth interrupting the agenda for.

• Single-threaded past week four. One contact, no matter how enthusiastic, is a single point of failure. Gong's call analysis has repeatedly shown multi-stakeholder deals closing at materially higher rates.
• A close date that has moved twice. Two slips is a pattern, not bad luck. The third slip is already scheduled; the only question is whether you find out now or in week eleven.
• No buyer-side commitment in the last two meetings. Two consecutive meetings where only your side took actions means you are running a project, and the buyer is watching it.
• Pricing discussed before the buyer stated a decision criterion. It usually means the deal is being shaped by your process, not theirs.
• The champion has stopped replying but the stage has not moved. The CRM will happily carry this deal at 60% for another two months. It should not.

Each of these is a forecast problem before it is a deal problem, which is the connection to the five CRM fields that lie to you — a review that surfaces them weekly is the cheapest forecast-accuracy intervention available to a sales manager. If you are building the wider operating rhythm around this, our page for sales teams covers how the pieces fit together.

Pick the deals before the meeting, not during it

The single highest-leverage change most managers can make is to choose the deals in advance. Fifteen minutes on Sunday evening or Monday at 08:30, using four filters:

• Every deal in commit that has slipped once. All of them, every week, without exception.
• The two largest deals in the quarter, regardless of health.
• Any deal with no buyer-side activity in 14 days.
• One deal per rep that is going well — so the review is not exclusively a problem clinic.

That last one matters more than it sounds. A review that only ever inspects trouble teaches the team that visibility is punishment, and you will get less of it.

Adapting the agenda to your team size

The 45-minute format above is written for a team of six to ten reps. It does not survive contact with a team of three or a team of twenty without modification, and most managers discover this by running it badly for a quarter first.

Three to five reps

Do not run a group review at all. At this size, deal inspection is better as a weekly 1:1 of twenty minutes per rep, because the group format wastes four people's time to inspect one person's deals. Keep a fifteen-minute team session for pattern and coaching only — the one theme, the one objection — which is the part that genuinely benefits from an audience.

Six to twelve reps

Run the agenda as written. The one adjustment worth making: rotate which reps get inspected so that everyone is on roughly every third week. Predictable rotation beats manager discretion, because discretion is read as suspicion — if only the struggling reps get inspected, being inspected becomes a signal.

Thirteen reps and up

Split by segment or pod, not by seniority. Two 45-minute reviews with six to eight people each will inspect more deals and produce more decisions than one 90-minute review with sixteen. Roll up the themes, not the deals, to whatever forum sits above — a leadership meeting does not need deal-level detail, it needs the three patterns the front line found this month.

Across all three sizes, the invariant is the same: the number of deals inspected per hour should be between six and twelve. Below six and you are over-analysing. Above twelve and you are back to the roll call.

End with commitments, not notes

Most pipeline reviews produce notes. Notes are a record of a conversation; commitments are a change to the world. The difference in the room is small — a sentence with an owner and a date instead of a sentence without one — and the difference in outcomes is most of the value of the meeting.

The failure mode here is administrative rather than strategic. Commitments get made in the room and then live in the manager's notebook, which nobody reads on Thursday. Whatever tooling you use, the commitments need to land where the work happens: the CRM record for the deal, or the task list the rep actually opens.

How this works with Floral

Two concrete things, and I will be careful not to overclaim.

First, deal inspection stops depending on rep recall. Floral records and transcribes customer meetings and produces a structured summary for each one — what was discussed, what was agreed, and the to-dos that came out of it. A human approves it, and on approval it syncs to the CRM. When you ask “what did the buyer commit to, in their own words”, the answer is in the record from the day it was said, not reconstructed under pressure in front of colleagues.

Second, the manager's fifteen minutes of prep gets cheaper. Instead of opening eight CRM records and piecing together a narrative, you read the last summary for each deal you plan to inspect.

What Floral does not do is decide which deals are at risk for you. There is no deal-scoring oracle here. The judgement stays with the manager; what changes is the quality of the evidence that judgement runs on.

The test

A simple way to know whether your pipeline review is working: after next Monday's meeting, count the deals whose state changed as a direct result — reclassified, killed, escalated, re-approached. If the answer is zero, you held a status meeting. If the answer is three, you ran a pipeline review.

Aim for a review where at least one deal dies every week. Killing deals is what makes the rest of the forecast trustworthy, and no team that never kills anything has an accurate number.

If you want to see what deal inspection looks like when the evidence comes from the meetings themselves, book a demo — bring the pipeline you are reviewing on Monday and we will run the five questions against it.