# Churn analysis Human Guide

## What This Is For
When the user needs to identify at-risk accounts, understand why customers are leaving, reduce churn rate, build health scores, design save plays, or create win-back campaigns. It gives the agent a clearer input/output frame for go-to-market work: what context to ask for, what decisions to make, and what usable artifact to return.

Use this as a human-readable version of the Churn analysis agent skill. It is meant for marketers, operators, founders, and other non-coders who want the workflow without reading agent-specific implementation instructions.

## When To Use This
- Use this when you need a repeatable process for churn analysis.
- Use this when the task needs judgment, examples, constraints, or a clear output format rather than a one-off prompt.
- Use this when you want to hand an AI assistant enough context to produce a usable marketing artifact.

## When Not To Use This
- Do not use this when you only need a quick factual answer.
- Do not use this when the work depends on private data you cannot share with the assistant.
- Do not use this as a replacement for legal, compliance, financial, or medical review.

## What You Need Before Starting
- The goal or business outcome you want.
- The audience, customer segment, or market context.
- Any source material the assistant should respect, such as notes, briefs, examples, URLs, or brand guidance.
- Constraints such as tone, length, channel, deadline, region, or approval requirements.
- A clear definition of what a good final answer should look like.

## Step-By-Step Workflow
1. State the job clearly: "Use the Churn analysis guide to help me with..."
2. Add context: audience, goal, offer, channel, source material, and constraints.
3. Ask the assistant to identify missing inputs before producing the final output.
4. Have the assistant follow the skill-specific guidance below.
5. Review the result against the final checklist and ask for revisions where needed.

## Skill-Specific Guidance
- **Extract signals** — Analyze four signal categories across every account: support signals, communication signals, usage signals, and commercial signals (see framework below).
- **Score risk** — Build a composite risk score (0-100) for each account using weighted signal categories. Higher score means higher risk.
- **Generate save plays** — For high-risk accounts, produce specific interventions: root cause hypothesis, recommended actions, talk tracks for the CS conversation, and escalation triggers.
- **Build the weekly scorecard** — Compile into a weekly risk report with account-by-account analysis, MRR at risk, trend data, signal distribution, and recommended focus areas.
- **Design interventions** — For each churn driver identified, design the appropriate intervention: product fix, CS outreach, cancel flow save offer, dunning sequence, or win-back campaign.
- **Weekly risk scorecard** — Every account scored and tiered with signal breakdown
- **MRR at risk summary** — Total revenue exposure by risk tier
- **Save play briefs** — For each red/orange account: root cause, recommended action, talk track, escalation trigger
- **Intervention designs** — Cancel flows, dunning sequences, or win-back campaigns as needed
- **Trend analysis** — Signal distribution changes over time
- **Value gap** — Product does not solve the problem well enough
- **Onboarding failure** — Customer never reached the aha moment (churn in first 30-60 days)

## Decision Points And Nuance
The original skill emphasizes: When to Use, Context Required, Workflow, Output Format, Frameworks & Best Practices, Signal Extraction Categories, Risk Scoring Model, Risk Tiers and Response Timelines, The Churn Driver Taxonomy, Cancel Flow Design.

Use these questions to steer the work:
- What is the intended audience or buyer?
- What source material must be preserved?
- What should the assistant optimize for: clarity, persuasion, accuracy, speed, creativity, or conversion?
- What examples represent the desired quality bar?
- What should the assistant avoid?

## Common Mistakes
- **Onboarding failure** — Customer never reached the aha moment (churn in first 30-60 days)
- **Ask why (required).** Present 5-7 reason options matching the taxonomy. Include free-text. This data is essential.

## Copy-And-Paste Prompt
```text
Use the Churn analysis human guide.

My goal:
[Describe the business outcome]

Audience:
[Describe who this is for]

Context and source material:
[Paste notes, examples, links, or existing copy]

Constraints:
[Tone, length, channel, timeline, must-include items, must-avoid items]

Before producing the final output, ask me for any missing information that would materially improve the result.
```

## Final Checklist
- [ ] The output matches the original goal.
- [ ] The audience and context are reflected in the answer.
- [ ] Important constraints and source material were preserved.
- [ ] The assistant made the relevant decisions explicit.
- [ ] The final artifact is ready to use, review, or hand to the next person.

## Source
This guide was generated from the shawnpang/startup-founder-skills skill entry for `churn-analysis`.

## Source Skill Notes
These notes preserve the nuance from the original skill. Use them as supporting reference when the workflow above feels too generic.

# Churn Analysis

## When to Use
Activate when a founder needs to identify at-risk accounts before they churn, diagnose churn drivers, build a customer health scoring system, design cancellation or save flows, recover failed payments, or re-engage lost customers. This includes prompts like "our churn is too high," "which customers are about to leave," "why are customers canceling," "build a customer health score," "set up dunning emails," or "create a win-back campaign." Especially relevant for seed/Series A teams managing customers manually without dedicated CS platforms like Gainsight or ChurnZero.

## Context Required
- **From startup-context:** business model (B2B/B2C, subscription/usage-based), current churn rate (logo and revenue), customer segments, pricing tiers, contract terms, product usage data availability, and current retention tooling.
- **From the user:** available data sources (support tickets, Slack channels, NPS scores, usage logs, email logs, billing data), what "healthy" customer behavior looks like, any historical churn patterns, whether churn is primarily voluntary or involuntary, and the specific churn problem to solve.

Work with whatever data is available. Early-stage companies often lack formal CS systems — the skill works with support inboxes, Slack history, and spreadsheets.

## Workflow
1. **Intake and baseline** — Gather all available customer data: customer lists, support tickets, Slack/communication history, NPS scores, usage data, email logs, and billing records. Establish what "healthy" looks like and identify any known churn patterns.
2. **Extract signals** — Analyze four signal categories across every account: support signals, communication signals, usage signals, and commercial signals (see framework below).
3. **Score risk** — Build a composite risk score (0-100) for each account using weighted signal categories. Higher score means higher risk.
4. **Generate save plays** — For high-risk accounts, produce specific interventions: root cause hypothesis, recommended actions, talk tracks for the CS conversation, and escalation triggers.
5. **Build the weekly scorecard** — Compile into a weekly risk report with account-by-account analysis, MRR at risk, trend data, signal distribution, and recommended focus areas.
6. **Design interventions** — For each churn driver identified, design the appropriate intervention: product fix, CS outreach, cancel flow save offer, dunning sequence, or win-back campaign.

## Output Format
A churn risk report tailored to the specific request. This may include:
1. **Weekly risk scorecard** — Every account scored and tiered with signal breakdown
2. **MRR at risk summary** — Total revenue exposure by risk tier
3. **Save play briefs** — For each red/orange account: root cause, recommended action, talk track, escalation trigger
4. **Intervention designs** — Cancel flows, dunning sequences, or win-back campaigns as needed
5. **Trend analysis** — Signal distribution changes over time

## Frameworks & Best Practices

### Signal Extraction Categories
Analyze every account across these four signal types:

**Support signals:** Ticket volume spikes, unresolved tickets, escalation language ("frustrated," "unacceptable," "cancel"), response time degradation, repeat issues on the same topic.

**Communication signals:** Silent accounts (no contact in 30+ days), frequency decline, sentiment shifts in Slack/email, champion disengagement (the main contact goes quiet), new stakeholder asking basic questions (signals champion departure).

**Usage signals:** Login frequency drops, feature abandonment (stopped using features they previously used regularly), shallow usage (logging in but not completing core workflows), no growth in usage over time, export/data download spikes (preparing to migrate).

**Commercial signals:** Discount requests, downgrade inquiries, payment failures, renewal proximity with no expansion discussion, competitor mentions in any channel.

### Risk Scoring Model
Build a composite score (0-100) by weighting individual signals:

| Signal Severity | Points | Examples |
|----------------|--------|----------|
| **Critical** | 25 | Explicit cancel request, competitor migration started, champion left |
| **High** | 15 | Usage dropped 50%+, 3+ unresolved escalations, payment failed twice |
| **Medium** | 8 | Login frequency declining, support sentiment negative, downgrade inquiry |
| **Low** | 3 | Slight usage dip, delayed renewal conversation, single missed payment |

Multiple signals compound. An account with two high signals (30 points) and three medium signals (24 points) scores 54 — solidly in the Orange tier.

### Risk Tiers and Response Timelines

| Tier | Score | Timeline | Action |
|------|-------|----------|--------|
| **Red** | 70-100 | Action this week | Executive outreach, save offer prepared, root cause identified |
| **Orange** | 40-69 | Action within 2 weeks | CS outreach, intervention plan, monitor daily |
| **Yellow** | 20-39 | Monitor within 30 days | Check-in scheduled, watch for signal escalation |
| **Green** | 0-19 | Routine check-in | Quarterly review, expansion opportunity assessment |

### The Churn Driver Taxonomy
Categorize every churn event into one of these buckets:
1. **Value gap** — Product does not solve the problem well enough
2. **Onboarding failure** — Customer never reached the aha moment (churn in first 30-60 days)
3. **Support failure** — Bad experience getting help
4. **Price sensitivity** — Too expensive relative to perceived value
5. **Champion departure** — Internal champion left the customer's company
6. **Business change** — Customer's needs changed (acquisition, pivot, shutdown)
7. **Involuntary churn** — Payment failure, not a conscious decision to leave

### Cancel Flow Design
1. **Ask why (required).** Present 5-7 reason options matching the taxonomy. Include free-text. This data is essential.
2. **Offer a targeted save** based on stated reason: "too expensive" gets a discount/downgrade, "missing feature" gets the roadmap, "not using it" gets a billing pause.
3. **Confirm with friction.** One extra click showing what they lose. Show value, not guilt.
4. **Offer a pause.** 30-60 day billing pause saves 15-25% of would-be churners in B2C and 10-15% in B2B.
5. **Offboard gracefully.** Confirmation email with data export and a "we'd love to have you back" message.

A well-designed cancel flow saves 10-20% of users who initiate cancellation.