How to Run a Forecast Call That Is Not a Radio Show

Here is the test. Sit in on your team's next forecast call and count two things: minutes of uninterrupted seller narration, and numbers that change as a result of the call.

If each seller talks for ten minutes and no number changes, you don't have a forecast call. You have a radio show — a weekly broadcast where deals are narrated, optimism is performed, the leader nods along, and everyone leaves with exactly the beliefs they arrived with. Radio shows are pleasant. They are also how quarters get blown, because every blown quarter I have ever autopsied traced back to the same deal: the one everybody privately doubted and nobody formally downgraded. The radio show is where that downgrade goes to not happen.

A forecast is a promise backed by evidence. The call exists to inspect the evidence — not the optimism, not the narrative, not the seller's tone of voice. Here is how to rebuild it.

Fix the definitions first

Forecast categories fail when they're vibes with labels. Make them structural:

Commit — committed to close this quarter. All three closure paths — commercial, business, technical — are secured. You would bet your credibility on it.
Strong Upside — could close this quarter; one or two of the three paths secured.
Weak Upside — could close this quarter; zero paths secured but some underway (mid-evaluation, for instance).
Omitted — in pipeline but not ready to be forecasted.

The three-closure-path language is what makes the definitions inspectable. Technical: has the proof been evaluated and the readout held — not "trending well," held? Business: has the Economic Buyer given a conditional commitment on record? Commercial: are procurement, security, and legal mapped with customer-confirmed dates? Each path is a factual question with a yes or no answer. "It feels strong" answers none of them.

And notice that the Commit definition is deliberately personal — you would bet your credibility on it — because forecast integrity is a culture before it is a spreadsheet. A seller who commits a deal is signing it, not describing it.

The moment that teaches the definition

The best forecast-culture story I know happened on an ordinary Thursday. Mid-cycle, week 15 of a marquee deal, evaluation criteria trending green, the AE moved the deal to Commit. His leader moved it back — on the call, in front of the team, with the category definition on screen:

"Technical path: trending, not secured — the readout hasn't happened. Business path: secured — conditional commitment on record. Commercial path: zero — sourcing hasn't returned a single MAP date. That's one of three. Strong Upside."

It cost the seller nothing but pride, and it taught the team the definition better than any enablement session ever could. The deal entered Commit four weeks later — when all three paths were evidenced — and closed in week 22, exactly as forecast. The forecast was boring. That is the point of a forecast. A forecast call that produces drama has failed; the drama was supposed to be caught in week 15, at the category boundary, where it's cheap.

Inspect a one-pager, not a monologue

The structural fix for narration is to change what's on screen. Before the call, every seller completes a Deal Review One-Pager for each material deal — eight fields, no essay questions:

Forecast category and close date
Deal-health score and the lowest MEDDPICC letter
Cost of delay, one sentence, their number
Last Economic Buyer touch / next EB touch — dates
The champion's most recent test — what they did for us, and when
The competitor's best move this month
The one thing that kills this deal
This week's single highest-leverage action

The review discusses the evidence in the fields, not the story around them. And every field is a trap for a specific fiction. "Last EB touch: September" in a December-close Commit deal is a finding. A champion whose most recent test is "returns my emails quickly" is a coach wearing a champion's title. A blank in field 6 doesn't mean the competitor is idle; it means you don't know what they're doing. An empty field is a finding, not a formality — if the seller can't fill it, the deal has just told everyone where the work is.

Field 2 deserves emphasis. MEDDPICC letters score 0–3 on evidence — 0 unknown, 1 assumed, 2 confirmed by one source, 3 confirmed by multiple stakeholders with evidence — and the lowest letter is the deal's real score. A deal with seven 3s and a Paper Process of 0 is not 87% healthy. It is a deal that will die in vendor registration in January. The lowest letter is also the week's work assignment, which is how the forecast call stops being an audit and starts being coaching.

Pair the one-pager with a single roll-up score per deal on screen — I use a 0–100 Deal Velocity Index, weighted across MEDDPICC evidence, the clarity of the customer's three Whys, EB engagement, confirmed MAP dates, and champion strength, with a hard floor of 75 for Commit. The specific instrument matters less than the property: the score must be built from evidence a leader can demand to see.

Four questions that change numbers

Run the call on questions with factual answers:

"Which of the three closure paths is secured, with what evidence?" — the category question.
"What changed since last Thursday?" — the momentum question. A deal where the honest answer is "nothing" three weeks running is not a deal; it's an entry.
"Which leading indicator is red this week, and what's Monday's plan against it?" — because revenue is a lagging indicator; the inputs (new meetings, real evaluations in flight, business cases completed) are what you can actually manage.
"Show me the deals that haven't moved stage in 30 days — advance them, regress them, or park them today." — the anti-zombie question. Today, not "let's watch it."

And normalize the stage regression. In a healthy operating rhythm, at least one deal gets regressed — stated in exit-criteria language, not as punishment — whenever one deserves it. A pipeline where scores and stages only ever rise isn't a strong pipeline; it's a team that has learned bad news is unwelcome. The whole apparatus exists to make bad news cheap on Thursday instead of catastrophic in week 13.

Publish accuracy, close the loop

One quarterly practice completes the system: publish forecast accuracy by person. Not to shame — because sellers systematically over- or under-call in personal, correctable patterns. The seller who sandbags and the seller who dreams both destroy planning; they just do it in opposite directions, and neither can correct a pattern nobody has shown them.

I learned all of this the uncomfortable way. When COVID gutted our Latin America operation and I was left rebuilding the region essentially alone — carrying the technical work, the selling, and the number simultaneously — I had no slack for deals that were secretly weak. The Thursday inspection wasn't a management ritual; it was survival. Every deal had to argue for itself with evidence, because there was no headcount to absorb a fiction. The discipline that emerged grew the region's ARR roughly six-fold in two years — and it works even better with a full team, because the radio show scales linearly while the inspection compounds.

Frameworks do not run themselves. They run on a rhythm, and the rhythm is the leader's actual product: forecast and notes updated by Thursday, the team call on evidence, scores on screen, three-closure-path language mandatory, one regression when one is warranted. Do that for a quarter and something strange happens to your forecast.

It gets boring. Which was always the point.

This essay is adapted from Chapter 11 and the Epilogue of The Value Engine: How Elite Enterprise Sales Teams Turn Buyer Pain into Forecastable Revenue by Rudy M. Celekli. The book lays out the full weekly, monthly, and quarterly operating rhythm on one running deal. The free Field Toolkit includes the Deal Review One-Pager (Template 10) and the DVI Scorecard, ready to put on screen this Thursday. Get the book and the Toolkit at the link in the footer.