Stage-to-stage conversion rate is one of the first metrics any sales team tracks, and in most tools we have audited, it is calculated in a way that quietly hides the exact problem it is meant to surface.
The common calculation
Most CRMs and dashboards calculate stage conversion as deals that reached stage B this period divided by deals that reached stage A this period. This treats the two numbers as though they describe the same cohort of deals, when they almost never do.
Why this hides the real problem
If a team had an unusually strong month of top-of-funnel deals entering stage A, this period's stage A-to-B conversion rate will look artificially low, purely because a large new cohort has not had time to convert yet, not because anything got worse. We have seen sales leaders react to this false signal by changing a process that was not actually broken.
The fix: cohort-based conversion
Track conversion by the cohort that entered stage A, and measure what percentage of that specific cohort reaches stage B, C, and closed-won over time, rather than comparing two different periods' snapshot counts against each other.
A cohort view takes longer to become statistically meaningful for a new metric, but it will not mislead you the way period-over-period snapshots can.
Segment cohorts by lead source at minimum — conversion rate varies enormously by where a deal originated.
Re-run your last two quarters' 'declining conversion' narrative through a cohort lens before acting on it — we have seen this reverse the conclusion more than once.
The snapshot method is not wrong, exactly. It is just answering a different question than the one most sales leaders think they are asking.
