alexe-ev Design Retention Playbook
Purpose
Help teams design a structured retention playbook — with defined signals, intervention plays, and escalation paths — that enables CS teams to act proactively rather than reactively.
Skill type
Conceptual skill
Use this skill when
- Churn is being discovered too late to intervene
- CS is handling at-risk accounts inconsistently or without a framework
- A new CS team needs a structured approach to retention
- Churn analysis has identified root causes that now need intervention plays
Do not use this skill when
- Churn root causes haven't been identified yet (use analyze-churn-retention first)
- The goal is health score design (use build-customer-health-score)
Required inputs
- Churn root causes (from analysis or known patterns)
- Customer segments
- CS team structure and capacity
Optional inputs
- Health score system (if exists)
- Current CS intervention approaches
- Escalation path to product or engineering
- Success metrics for retention efforts
Upstream context
Works best when:
- Churn analysis exists
- Health monitoring is in place
- Customer segments are defined
Downstream handoff
Output can feed:
- monitor-adoption-health (playbook interventions triggered by health signals)
- analyze-churn-retention (playbook effectiveness feeds back into churn analysis)
Instructions
- Map churn root causes to intervention types.
- Design a play for each churn cause: trigger signal, intervention action, escalation path, success criteria.
- Design an expansion play: signals for expansion readiness and CS motion.
- Define the intervention timeline: when to intervene relative to renewal or churn signal.
- Define escalation paths: when CS needs product, engineering, or executive involvement.
- Define how playbook effectiveness will be measured.
Output
Provide:
- Retention plays by churn root cause
- Each play: trigger signal, action, owner, timeline, escalation, success criteria
- Expansion play
- Intervention timeline design
- Escalation matrix
- Playbook effectiveness metrics
Risks / caveats
- Playbooks without triggers are just documents — connect every play to a specific signal
- Over-engineered playbooks don't get used — keep plays simple and executable
- Measure playbook effectiveness, not just churn rate — isolate intervention impact