The mandate
Organize customer success around observable value, not account activity.
Designs onboarding, adoption, support, retention, expansion, and voice-of-customer loops.
The Customer Success Agent plans onboarding, value milestones, support, health signals, retention reviews, feedback routing, and escalation. It uses cohort and customer evidence rather than intuition disguised as a score.
What to bring
Bring the customer segment, first-value milestone, friction, and evidence of behavior.
The brief should include customer commitments and support themes as well as usage signals. Missing data must remain different from healthy behavior.
- 01
The product, plan, customer segment, and lifecycle stage in scope.
- 02
The observable outcome that tells the customer and company value has started.
- 03
Onboarding friction, support themes, adoption gaps, churn reasons, and customer commitments.
- 04
Cohort data, qualitative feedback, support evidence, usage signals, and known limitations.
- 05
The onboarding, adoption, support, retention, expansion, or feedback outcome to improve.
What you receive
A journey, health model, and feedback loop tied to customer outcomes.
The outputs connect onboarding and adoption signals to owned interventions, escalation boundaries, and product decisions that can be closed with customers.
- 01
Customer onboarding journey
success/onboarding.mdReady when
Milestones, owners, friction, and time-to-value signals are explicit.
- 02
Customer health and escalation model
success/health-model.jsonReady when
Signals lead to defined interventions and owners.
- 03
Feedback-to-product workflow
success/feedback-loop.mdReady when
Evidence is categorized, routed, and closed with customers.
The method
Map value, detect risk carefully, and close the customer feedback loop.
The agent follows the customer from promise to outcome and makes every intervention evidence-based and owned.
- 1
Map value
Define lifecycle stages, observable milestones, responsibilities, and time-to-value.
- 2
Detect risk
Combine cohort, usage, support, and qualitative evidence without turning missing data into a score.
- 3
Intervene
Assign playbooks, owners, service levels, escalation, and success evidence.
- 4
Route feedback
Send product, sales, legal, and support evidence to owners and close the loop with customers.
Signals & guardrails
A health score is useful only when it predicts an owned intervention.
Exposed customer data, unsupported commitments, unexplained scoring, or feedback that disappears without a response block the system.
Signs the work is useful
- Time to first value
- Retention by cohort
- Support resolution time
Reasons to pause
- Health scores lack predictive evidence
- Customer commitments bypass product and legal review
Approval boundary
This entrypoint declares no open-world effect. Applying its todos still requires an authenticated, scoped project run.
Technical appendix
The exact contract behind the profile.
Useful for operators who need to inspect the immutable Kit version, typed boundary, and verification surface.
- Kit
- company-suite@0.1.0
- Entrypoint
- plan-customer-success
- Function
- plan-customer-success
- Runtime
- python@3.12
- Input
- specialist-brief.schema.json
- Output
- specialist-plan.schema.json
Verification
specialist-plan-quality · specialist-plan-schema
Connected systems
No external connector required