Your CRM Isn't Lying to You. Your Process Is.

I've carried a number for over a decade. Losing a deal hurts — but not like losing one you should have won. The ones that haunt you aren't the no's. They're the deals that die because you didn't see the risk early enough.

And the second worst feeling? Looking at your pipeline in week seven and realising half the deals in Commit aren't real. They were never real. The CRM just made them look real because nobody asked for proof.

There's a moment in every quarter — usually around week six or seven — when the forecast starts to wobble. Deals that looked solid start slipping. Close dates move. Champions go quiet. And the CRM, which showed a healthy pipeline two weeks ago, suddenly looks like fiction.

The typical response? Tighten CRM hygiene. Mandatory fields. Required updates. Weekly compliance checks.

This rarely works at scale. It doesn't work because it misdiagnoses the problem.

The CRM captures what you tell it to capture

Salesforce, HubSpot, Dynamics — these are record systems. They store what gets entered. That's it. The quality of the data is a direct reflection of what your process asks reps to put in.

When a rep moves a deal from Stage 2 to Stage 3, what does that actually mean? In most organisations, it means the rep decided the deal felt ready. Maybe they had a good meeting. Maybe the prospect said encouraging things. Maybe the close date is approaching and the deal needs to look like it's progressing.

What it doesn't mean — almost ever — is that specific, verifiable evidence exists confirming the deal meets defined criteria for Stage 3.

Because those criteria usually don't exist. Or they exist on a slide from onboarding that nobody has looked at since.

The CRM doesn't lie. It faithfully records the optimism, assumptions, and good intentions that reps feed into it. Nobody requires evidence before those entries are made. That's the problem.

Optimism is the default

This isn't about bad reps. It's about human nature.

When a rep has a positive discovery call, they frame the deal positively. They heard the prospect describe a real problem. The prospect asked about pricing. The conversation flowed well. All of this feels like progress.

But feeling like progress and having evidence of progress are completely different things. Did the prospect confirm a budget? Did they describe the decision process? Is there an event driving urgency?

Without evidence for each of these, the rep is operating on interpretation. And interpretation trends optimistic. Always.

I've seen reps bet the whole deal on one champion — and watch it die the second that person went quiet, got overruled, or left. The rep marks the deal as multi-threaded. It's not. They've spoken to one person. And the deal dies because that champion couldn't sell it internally alone.

Managers feel this as repeat reviews, surprise slippage, and the late-quarter scramble that everyone pretends is normal. It's not normal. It's the predictable outcome of a system that accepts narrative in place of evidence.

Three symptoms you can spot without looking at a single deal

Close dates cluster at quarter-end. When 60% of your pipeline is set to close in the last two weeks, that's not a buying pattern. It's a forecasting pattern. Reps set dates based on when they need the deal to close, not when evidence suggests it will.

Stage progression happens in bursts. Deals sit in Stage 2 for weeks, then jump to Stage 4 in a single update. Real buying processes don't work like that. What actually happened is the rep batch-updated their pipeline before a review.

The same deals sit in Commit for weeks. If a deal has been in Commit for three weeks without closing, either your Commit criteria are meaningless or the deal was moved there on hope. Both are evidence problems.

What "evidence-based" actually looks like

The fix isn't more fields. It isn't more compliance. It isn't another dashboard.

It's requiring evidence before deals progress — and making that evidence visible to everyone who needs it.

In practice, that means stage criteria defined in evidence terms. Not "discovery completed" but "problem confirmed — direct quote from economic buyer captured. Decision process documented with timeline and named participants." That level of specificity.

It means progression requires evidence, not judgement. A deal doesn't advance until the proof exists. No evidence, no advancement — or the manager overrides with a logged reason that's visible to everyone above them.

And it means evidence gaps are visible before reviews. The manager sees what's confirmed and what's missing without having to ask. The review becomes about the gaps, not about reconstructing the deal from scratch.

Stop blaming the CRM

Your CRM is doing what it was designed to do: store records. The problem is those records aren't backed by evidence. They're backed by narrative. And narrative, under the pressure of targets and timelines, always trends optimistic.

The answer isn't a better CRM. It's an evidence layer alongside your CRM — one that requires proof before deals progress, surfaces what's missing before anyone has to ask, and gives leadership a pipeline they can defend with facts, not stories.