Let me walk you upstream.
Every example in this section is a real pattern I’ve found more than once, anonymized into archetype. In each one, the pain surfaced in one department, the fix got aimed at that department, and the origin sat somewhere else entirely — untouched, patient, generating next year’s version of the same problem.
—
**The sales problem that lives in operations.** Close rates sag. Leadership responds where the pain is: sales training, new comp plan, maybe a new sales leader. But walk the customer journey backward and you find fulfillment friction — deliveries that strain, onboarding that stumbles, small promises that quietly die in the gap between closed-won and delivered. Existing customers feel it, references dry up, expansion revenue stalls, and word moves through the market the way word does. Sales inherits all of it as churn and lengthening cycles. You can train the sales team forever. They’re selling *against their own operations*, and no script fixes that.
**The marketing problem that lives in data architecture.** Lead volume is fine; conversion is dismal. So: new agency, new messaging, new spend. But the actual origin is that “lead” means four things in four systems — the definitions drifted years ago, attribution is guesswork across disconnected tools, and nobody can tell which campaigns produce *revenue* versus which produce activity. Marketing isn’t underperforming. Marketing is flying instruments that lie, optimizing toward the only numbers it can see. The agency inherits the same instruments and fails the same way, at a higher day rate.
**The retention problem that lives in onboarding.** Churn spikes at month eleven, so customer success gets a playbook, a QBR cadence, a save-desk. But go look at week two — the handoff nobody designed, Chapter 8, being experienced by every new customer personally. The promises made in the sales cycle never transferred. The customer explained their goals again to strangers. The first ninety days quietly taught them what this vendor is like, and month eleven is just when the contract lets them act on the lesson. The save-desk is performing CPR at the wrong end of the timeline.
**The forecasting problem that lives in definitions.** The forecast misses, quarter after quarter, always in the same direction. So: forecasting software, pipeline reviews, commit calls. But the origin is that lifecycle stages were never actually defined — “qualified” is a mood, stage three means something different to each rep, and close dates roll forward because rolling them forward is free. The forecast isn’t a math problem. It’s an aggregation of ambiguity, computed to two decimal places. Better software just computes the ambiguity faster.
—
Four stories, one shape. Say it once, cleanly:
**Pain is a location. Cause is a different location. The org chart routes accountability to the pain.**
That’s the trap in a sentence. Symptoms surface wherever the customer or the revenue finally touches the failure — which is almost always downstream, in sales’ numbers or CS’s churn or the forecast’s miss. But revenue moves through handoffs, across systems, in one connected flow. A failure anywhere surfaces at the end of the line, and the end of the line gets the blame, the fix, and eventually the new leadership.
This is why I evaluate all five revenue systems simultaneously, every time, even when the client is certain they have “a sales problem.” Isolated audits produce isolated fixes. Isolated fixes produce recurring problems. The origin is upstream, structural, and — until you know to look — invisible.
Which raises the honest question: if misdiagnosis is this predictable, why do smart companies keep doing it?
Next section. The answer is more sympathetic than you’d think.