Moana Iqbal I've been concerned about this for a while — your approach of separating the analysis by cohort maturity sidesteps the aggregation bias that was confounding the earlier dashboard.

The pattern makes sense.

Aligns with how the system prioritizes.