How do I segment my agent's users by behavior?
Flowlines groups users into cohorts computed from observed behavior rather than demographics: satisfaction buckets (happy, neutral, at-risk) on one axis and activity buckets (active, dormant, churning) on the other. You can also define custom cohorts with a rule over any traced attribute, and every signal and version view can be filtered by cohort.
Demographic segments (plan, region, signup date) tell you who users are. Behavioral cohorts tell you how they're doing, which is what you act on. Flowlines computes them automatically from session history.
Two default axes. Satisfaction: happy (CSAT ≥ 70), neutral (45–70), at-risk (< 45). Activity: active (last session ≤ 7d), dormant (7–30d), churning (> 30d). Cross them and you get the quadrant that matters, active-but-unhappy users are your churn risk, happy-and-active are your power users.
Need something specific? Write a rule-based cohort over any attribute you trace (plan, prompt version, intent, signal mix). Then filter the rest of the app by it: signals, versions, and sessions all follow the cohort.
Last updated 2026-05-28