Blog / Analytics & attribution
A confident wrong number is worse than a missing one
A close rate that read 100% when the truth was 50%, feeding seven reports a sales manager would act on. How it happened, why we fixed it at the source, and what to check in your own dashboards.
A missing number makes a manager ask a question. A wrong number that looks right makes them act. On an AI sales-coaching product we built, we found one of the second kind before the client relied on it.
One win, one hidden loss: 100%
When a customer’s CRM had more than one sales pipeline, deals outside the main one went missing from the close rate. One visible win and one hidden loss reads as 100%, when the truth is 50%. The error surfaced on test data while we were connecting CRM deals to call scoring.
Seven places read the same number
The close rate was not just a dashboard tile. Seven features read it, including the pipeline alerts and the weekly briefing managers receive by email. Patching each one would have left seven chances to get it wrong again.
So we traced the error to the one place it started and fixed it there. All seven reports corrected together. The same week we fixed 17 other issues found on the way.
Check your own numbers
- Test with more than the simple case. Two pipelines, two currencies, two teams: most metric bugs hide in the second of something.
- Check a number against a known answer. Build a small data set where you know the true close rate, and compare.
- Find every reader of a metric. Alerts, emails and exports often compute the same figure in their own way, or read it from one place you can fix once.
- Distrust perfect results. A 100% anything deserves a second look.
How we read data
A confident wrong number is worse than a missing one.
The examples in this article come from real engagements. Client details withheld; every figure comes from the client’s own data.
Read the case study →