1. Context
The company sold a compliance and risk platform to mid-market financial services firms. Their Series B deck had projected 40% growth in Year 1 post-raise. Eighteen months later, they were tracking at 14%. The product was solid — existing customers renewed at 94% and expanded. The problem was entirely in new business.
The CRO had been in the role for eight months. She had built out a proper RevOps function, implemented a new forecasting cadence, and was running weekly pipeline reviews. The team was working hard. And yet every Monday, the forecast for the quarter changed. Sometimes by 10%. Sometimes by 30%. The board had started requiring bi-weekly updates because they'd stopped trusting the numbers.
2. The Misdiagnosis
The CRO's working theory was a process problem. Reps weren't updating Salesforce consistently. Some deal information was living in email threads and personal spreadsheets rather than the CRM. If she could get clean, complete data, the forecasting would improve.
She had already spent two months working on Salesforce hygiene — new required fields, weekly data audits, a points system for completeness. Data quality had improved measurably. Forecast accuracy hadn't moved.
The board's theory was different: they thought the problem was the CRO herself. They'd seen this pattern before in portfolio companies. New CRO comes in, improves process on paper, but doesn't have the pattern recognition yet to call the quarter accurately. Some of them were quietly suggesting it might be time to find someone more experienced.
3. Why It Seemed Right
The data quality theory had obvious logic behind it. You can't forecast accurately from incomplete information. The CRO had diagnosed a real problem and attacked it methodically. The fact that she'd improved data completeness from 61% to 87% was a real achievement.
The board's theory also wasn't unreasonable. Forecast accuracy is partly a skill — experienced sales leaders develop intuition for which deals are real and which are reps telling them what they want to hear. An eight-month CRO, however capable, hasn't yet built the company-specific pattern recognition that produces reliable calls.
What both theories missed was a more fundamental issue: there was no shared definition of what a "committed" deal meant. Each rep had their own mental model. The CRO was synthesising subjective assessments without a common language underneath them.
4. What the Diagnostic Revealed
We asked one question that changed everything: "For your most recently closed-won deal, at what point did you mark it as Commit?" We asked every rep individually. The answers ranged from "when I felt good about the relationship" to "when they told me they were definitely buying" to "when procurement got involved" to "a week before quarter end because I needed to make my number."
Eight reps. Eight completely different definitions of the same stage.
The CRO was aggregating Commit figures from reps using incompatible criteria and reporting them to the board as a single number. The number was precise — it had a decimal point. It just had no meaning, because the inputs weren't measuring the same thing.
We then ran the historical data. Deals marked Commit by different reps had wildly different close rates. Rep A's Commit deals closed at 74%. Rep C's closed at 31%. They were using the same word to describe very different situations, and the forecast was averaging them as if they were equivalent.
5. Structural Changes
Commit was redefined with hard, observable criteria. A deal could only enter Commit if: the economic buyer had confirmed a purchase decision in writing (email counted), a contract had been requested or sent, and a close date had been agreed with the buyer — not set by the rep. If any of those three things hadn't happened, the deal went into Best Case, not Commit.
Best Case was also redefined. It required a verbal close-date commitment from someone with signing authority, active legal or procurement involvement, and no unresolved blocking issues identified in the last review. Deals that didn't meet that bar went into Pipeline.
The weekly review format changed. Instead of "what's in your Commit this week," the CRO now asked "walk me through what the buyer did since last week that supports this being in Commit." Reps had to evidence buyer movement, not assert deal health.
A rolling accuracy score was tracked per rep. How often did deals in their Commit bucket actually close in the quarter? This became a coaching input, not a performance metric — but it gave the CRO a calibration layer she hadn't had before.
6. The Outcome
In the first quarter after the new definitions went live, forecast accuracy moved from 58% to 79%. The CRO called 91% of actual revenue — within $180K on a $6.2M quarter. The board meeting that quarter was the calmest in two years.
By the following quarter, accuracy was at 84%. The bi-weekly board updates were dropped back to monthly.
The broader effect was on pipeline management. Because Commit now meant something specific, reps stopped gaming the stage to show progress. Deals in Best Case and Pipeline were worked differently — more urgency on Best Case to find the missing criteria, more qualification rigour on Pipeline to avoid deals sitting inert.
Growth accelerated to 38% in the twelve months following the engagement, not because the team changed but because they stopped spending cycles on deals that looked like pipeline but weren't going anywhere.
7. Why This Matters
Forecast accuracy problems are almost never data problems or people problems at their root. They are definition problems. When the word "committed" means something different to eight different reps, you don't have a forecast — you have a weighted average of eight different people's optimism levels.
The fix is uncomfortable because it requires accepting that your current Commit number is probably inflated, and the quarter will look worse before it looks better. But the alternative — continuing to report numbers the board has stopped believing — is more expensive in the long run, both for the company and for the CRO's tenure.
The CRO in this engagement kept her job. More than that, she became one of the most credible voices in her board meetings because she was the person whose forecasts the board had learned to trust.