Cohort attribution, channel performance, pricing signal analysis — written by practitioners who built these analyses for real growth teams.
Your aggregate retention number can look stable while individual cohorts tell a completely different story. Here's how to read both — and why conflating them leads growth teams to wrong conclusions.
Most MRR waterfalls show you expansion revenue as a line item. What they don't show you is which cohorts expanded, when, and what triggered the expansion. The difference is actionable vs. decorative reporting.
CAC is a point-in-time acquisition cost metric. It says nothing about 12-month retention by channel. This is why teams keep funding campaigns that generate customers who churn in the first two quarters.
You raised prices. Aggregate churn looks about the same. Did the change work, or are you seeing a lagged effect? Here's a framework for isolating the pricing cohort to separate signal from noise.
New MRR + expansion - contraction - churn = net new MRR. Simple enough. But decomposing each component into its cohort structure — that's where FP&A teams find the real levers. Here's how.
Your growth data is already in Snowflake. The question isn't whether to use it — it's how to query it for cohort analysis without writing a new dbt model every time someone has a question.
Connect Stripe and see your first cohort breakdown in 48 hours.
