Home / Blog / Note

"A sales pipeline review that changes the forecast"

"Most pipeline reviews ask reps how confident they feel and call the answer a forecast. A three-column review built on buyer evidence produces a smaller number that survives the quarter."

"A sales pipeline review that changes the forecast"
Representational image.

Monday, 9:15. Fourteen deals on the screen, one column called Confidence, and a regional manager going round the table. “Will it close?” “Yes, ninety per cent.” “This one?” “Committed. The buyer loves it.” Forty minutes later the roll-up is ₹3.4 crore and the room feels the quarter is safe. On the last day of the quarter the number is ₹2.1 crore. Nobody lied. The review did exactly what it was built to do: it collected confidence and called it a forecast.

Here is the claim that costs something to disagree with. A pipeline review that asks “will it close?” is not forecasting. It is polling. And the poll is rigged, because the person answering is punished for doubt.

Why the number inflates

The question asks for a verdict on the rep’s certainty, not evidence about the deal. A rep who says “I am not sure” is heard as a rep who is not on top of his territory, so he learns the answer that ends the meeting. The manager needs the number to be large for his own boss, so he accepts it. Both sides prefer the bigger figure, and the CRM stage percentage, entered by the rep, becomes the arithmetic of that shared preference. “Ninety per cent” does not mean a probability. It means “I would like this to close.”

This is not a moral failure, it is a structural one, and it is measurable. XANT Labs analysed 270,912 closed-won opportunities worth over $18.1 billion across 18 companies. Only 28.1 per cent of them closed within 5 per cent of the amount forecast 90 days out. Forty-seven per cent of those 90-day predictions were off by more than half. The direction of the error is the part to keep: when reps were wrong, they overestimated by an average of $91,000 and underestimated by only $47,000. Overestimation ran at nearly twice underestimation, and the larger the deal, the larger the optimism. A forecast built by adding up rep confidence is therefore biased upwards by construction, before anyone tells a lie.

The consequence shows up at the top. Xactly’s 2024 survey of 400 sales, finance and revenue operations leaders found that four in five had missed a quarterly sales forecast in the past year, and over half had missed it two or more times. That is not a tail risk. It is the normal operating condition of a pipeline reviewed by asking people how they feel.

The three columns

Replace the confidence column with three columns for every deal. They are filled before the meeting, in the CRM, in ten minutes, by the rep.

Column What goes in it What does not count
1. What the buyer has done Dated actions that cost the buyer something: shared the budget line, introduced the economic buyer, ran the pilot, sent the legal template “Had a good call”, “they are keen”, a demo you gave
2. What the buyer agreed to do next A next step with a date, confirmed by the buyer in their words: “they will send the PO draft by Friday” A step you proposed and they nodded at, “they’ll get back to us”
3. What we believe Everything else: “the champion is pushing internally”, “budget is approved”, “the competitor is weak here” Nothing is disqualified here, but nothing here enters the forecast on its own

The rule is short. A deal is in commit only when it has evidence in column 1 and a buyer-confirmed next step in column 2 that falls inside the quarter. A deal with only column 3, or a next step dated beyond the quarter, is upside. A deal with no buyer action for 30 days is off the forecast. The forecast is the sum of commit. Upside is reported beside it, never added into it to make the number look healthy.

Column 3 is not a bin for rubbish. It is where real intelligence lives before it hardens into evidence. The discipline is that it stays labelled as belief until the buyer does something that moves it across. Most pipelines are almost entirely column 3, which is why they read as forecasts and behave as wishes.

The one question that does the work

Before any talk of probability, ask one question: “When did the buyer last do something they did not have to do?”

It moves the room’s attention off the rep’s feelings and onto the buyer’s behaviour, and it has a checkable answer. A buyer who has done nothing costly since the demo is not in a deal. He is in a conversation. If the rep cannot name a dated action, the deal is upside, and no amount of “the champion is fighting for us” changes that.

The follow-up, only if needed, is: “What have they agreed to do next, and in whose words?” Take the answer literally. “They said they will revert” is column 3. “They asked us to send the contract so their legal can mark it up by the 20th” is column 2. The difference is not tone. It is whether the buyer has committed an action you can wait for.

This is the question that makes the rep think, because it gives him something he can go and change. He cannot manufacture confidence on demand, but he can book the next step and ask the buyer for the action. The review stops being a verdict on him and becomes a short list of things to do before next Monday.

How the meeting runs

The order matters more than the template.

Before: the rep fills the three columns. The manager reads column 1 first, not the amount and not the stage. The amount is the least informative number on the screen.

During: twenty-five minutes, five deals at most, biggest first. For each deal, read the buyer action aloud, ask the one question, and place the deal in commit, upside or off. Do not debate probability. If a rep argues a deal up, the answer is not a counter-argument, it is “what will the buyer do this week to prove it?” The deal stays in upside until the buyer does it.

After: the forecast is the commit sum. Send it with upside listed separately and one line for every deal that moved: “moved to upside, no buyer action since 14 October.” Keep the slip reasons. They are the coaching input, not a shame list.

Run it weekly, on the same day, and never in the same meeting as the call review. The call review is about the rep’s skill; this one is about the deal. Combining them turns both into an interrogation, which is the argument in the call review that coaches instead of audits. The weekly cadence is the same point as why sales training decays faster than leadership training: the review is the reinforcement, and the workshop is only the introduction.

A worked example

Illustrative example, not a client result. The figures are constructed to show the shape, not to benchmark anyone.

Eight reps selling industrial components had a quarterly commit of ₹3.2 crore, built the old way, from confidence. The previous quarter had landed at ₹1.9 crore. They switched to the three columns. In the first review, of 41 open deals only 11 had both a dated buyer action and a buyer-confirmed next step inside the quarter. Commit fell to ₹1.6 crore. Upside was ₹1.4 crore. The vice president was angry for a week, because the number looked like a collapse.

It was not a collapse. It was the same pipeline, finally labelled. Over the next six weeks the reps booked next steps to move deals up, and by week six commit was ₹2.0 crore across 19 deals. The quarter landed at ₹2.05 crore. In month one the selling had barely changed; the number had simply stopped lying. Then column 1 exposed the real gap: reps were not asking buyers for next steps at all. That was a skill problem, and it went to the training plan, not back into the spreadsheet.

The honest limit

The first effect of this method is a smaller number, and a smaller number will be read as worse performance. It survives only if a sales leader will defend the reduced commit to finance for one quarter. Where the forecast is a negotiated target rather than an estimate, column 3 will be pressured back into commit within a month, and the method dies. That is a leadership decision, not a process fix.

It does not create pipeline. If commit is small because there are few deals, the review has told you the truth about an empty funnel, and the answer is prospecting, not a better meeting. It does not fix the skill gap it exposes either. A rep who cannot ask a buyer for a next step will not be helped by a column; that belongs in sales capability training and in the manager’s coaching routine. And it is the wrong tool for high-velocity transactional sales with hundreds of small deals a week. It is for considered B2B deals with a buying committee and a cycle measured in weeks, where one conversation can move a crore.

What to do on Monday

Take the five biggest deals in this quarter’s commit. On one sheet, write the three columns for each. Ask the one question. Any deal without a dated buyer action and a buyer-confirmed next step moves to upside. Send the revised commit before the next roll-up, with one line per moved deal, and do not wait for the quarter to end to find out. The number will shrink. That is the review finally working.

Sources

  • https://resources.insidesales.com/blog/sales-forecasting-research (XANT Labs, “The Gap Between Forecasting and Reality”: 270,912 closed-won opportunities, $18.1 billion, 18 companies; 28.1% within 5% of the 90-day forecast; 47% off by more than half; average overestimate $91,000 versus underestimate $47,000)
  • https://www.xactlycorp.com/blog/forecasting/2024-sales-forecasting-benchmarks (Xactly 2024 Sales Forecasting Benchmark Report, survey of 400 sales, finance and revenue operations leaders: 4 in 5 missed a quarterly forecast in the past year, over half missed two or more times)
Keep reading
Measurement

"Measuring training without a control group, honestly"

"You will almost never get a control group in an Indian corporate. Here is the design that still lets you say something true: a waiting cohort, a number the business already counts, and four measurement points."

Consulting

Consulting discovery questions that price the work

Scope creeps because discovery collects symptoms instead of forcing decisions. Five questions, asked in order, turn a vague brief into a one-page scope note you can price and defend.

Want this applied inside your organisation?

Talk to us