Your forecast is a fiction

Every Monday, some version of the same ritual plays out across enterprise sales organizations. A rep updates a CRM field. A manager reads the update and forms an opinion. A director rolls up the opinions. A VP presents the rollup to the CRO. The CRO tells the board what revenue will look like this quarter.

At no point in this chain did anyone verify anything. The entire forecast is a narrative, constructed from other narratives, all the way down.

I've been thinking about why this is. It's not that revenue leaders are stupid. Most of them have lived through enough blown forecasts to know the process is fragile. They've all experienced the week-12 surprise, the "committed" deal that evaporates, the 3x pipeline that somehow doesn't cover the number. And yet the process persists, essentially unchanged, quarter after quarter.

The reason, I think, is that narrative-based forecasting serves a social function that has nothing to do with accuracy.

A seller moves a deal to Stage 3. What does that mean? It means the seller decided the deal feels like a Stage 3. Maybe there was a good meeting. Maybe the champion said encouraging things. Maybe the seller needs to show pipeline progression to avoid a difficult conversation with their manager.

The manager reviews the deal in a 1:1. They ask questions, the seller tells a story, and the manager forms a judgment about whether the story is credible. That judgment gets encoded as a forecast category: commit, best case, upside.

Multiply this by fifty reps, five managers, and three layers of leadership. By the time the number reaches the board, it has been filtered through so many subjective interpretations that its relationship to what's actually happening in the deals is tenuous at best.

This is not forecasting. This is collaborative fiction.

The interesting thing is that the raw material for something better already exists. It's in the call recordings, the email threads, the calendar patterns, the document shares, the Slack messages, the proposal redlines. Every interaction between seller and buyer produces a signal about where the deal actually stands. But none of this makes it into the forecast. Instead, we take rich, multi-dimensional interaction data and compress it into a single dropdown that a seller selected three days ago based on vibes.

Then we wonder why forecasts are wrong.

The technology to change this exists today. An AI system can process interaction data and compute a view of deal progress based on evidence rather than narrative. Stage progression becomes a system output, not a seller declaration. Risk signals surface from actual behavior (a champion who stopped responding, a technical evaluation that stalled, procurement not engaging) rather than from a story told in a pipeline review.

This isn't science fiction. The signals are already being captured. Call intelligence tools record conversations. Email systems track engagement. Calendar data shows meeting cadence. The problem isn't data availability. The problem is that nobody has stitched these signals together into a coherent, continuously updated model of deal reality.

But I think there's a deeper reason this hasn't happened yet, and it's not technical.

An evidence-based forecast would be more accurate. It would also be more threatening. When the forecast is narrative-based, everyone has cover. The rep can shade the story, the manager can apply their own interpretation, the VP can add a management buffer. Everyone maintains plausible deniability about what they knew and when they knew it.

An evidence-based system removes that cover. If the system says a deal is weak because the engagement data shows it's weak, nobody gets to argue that their "relationship" with the buyer will save it. If the system shows that a committed deal has zero activity from the economic buyer, that's visible to everyone, instantly.

Most sales cultures are not built for this level of transparency. They're built for narrative management, which is the art of telling a story about a deal that makes everyone comfortable enough to keep moving forward. An evidence-based system would expose the gap between what organizations believe about their pipeline and what's actually there. For many organizations, that gap is enormous, and closing it would be genuinely uncomfortable.

If I were a revenue leader who wanted to move toward evidence-based forecasting, I wouldn't start by replacing the current process. I'd start by running the two systems in parallel. Let the existing narrative-based forecast continue unchanged. Simultaneously, build an evidence layer that computes its own view of deal health and pipeline confidence from interaction signals. Compare the two over a quarter. Where do they agree? Where do they diverge? When they diverge, who turns out to be right?

That comparison produces something more valuable than a better forecast. It produces organizational awareness of how far the current narrative is from reality. And that awareness, uncomfortable as it is, is the prerequisite for change.

Every revenue leader I've talked to agrees, in private, that narrative-based forecasting is broken. The gap between knowing this and actually doing something about it is cultural, not technical. The technology to compute forecasts from evidence is straightforward. The willingness to live with what the evidence shows is the hard part.