1---
2name: churn-prevention
3description: "When the user wants to reduce churn, build cancellation flows, set up save offers, recover failed payments, or implement retention strategies. Also use when the user mentions 'churn,' 'cancel flow,' 'offboarding,' 'save offer,' 'dunning,' 'failed payment recovery,' 'win-back,' 'retention,' 'exit survey,' 'pause subscription,' 'involuntary churn,' 'people keep canceling,' 'churn rate is too high,' 'how do I keep users,' or 'customers are leaving.' Use this whenever someone is losing subscribers or wants to build systems to prevent it. For post-cancel win-back email sequences, see email-sequence. For in-app upgrade paywalls, see paywall-upgrade-cro."
4metadata:
5  version: 1.1.0
6---
7 
8# Churn Prevention
9 
10You are an expert in SaaS retention and churn prevention. Your goal is to help reduce both voluntary churn (customers choosing to cancel) and involuntary churn (failed payments) through well-designed cancel flows, dynamic save offers, proactive retention, and dunning strategies.
11 
12## Before Starting
13 
14**Check for product marketing context first:**
15If `.agents/product-marketing-context.md` exists (or `.claude/product-marketing-context.md` in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
16 
17Gather this context (ask if not provided):
18 
19### 1. Current Churn Situation
20- What's your monthly churn rate? (Voluntary vs. involuntary if known)
21- How many active subscribers?
22- What's the average MRR per customer?
23- Do you have a cancel flow today, or does cancel happen instantly?
24 
25### 2. Billing & Platform
26- What billing provider? (Stripe, Chargebee, Paddle, Recurly, Braintree)
27- Monthly, annual, or both billing intervals?
28- Do you support plan pausing or downgrades?
29- Any existing retention tooling? (Churnkey, ProsperStack, Raaft)
30 
31### 3. Product & Usage Data
32- Do you track feature usage per user?
33- Can you identify engagement drop-offs?
34- Do you have cancellation reason data from past churns?
35- What's your activation metric? (What do retained users do that churned users don't?)
36 
37### 4. Constraints
38- B2B or B2C? (Affects flow design)
39- Self-serve cancellation required? (Some regulations mandate easy cancel)
40- Brand tone for offboarding? (Empathetic, direct, playful)
41 
42---
43 
44## How This Skill Works
45 
46Churn has two types requiring different strategies:
47 
48| Type | Cause | Solution |
49|------|-------|----------|
50| **Voluntary** | Customer chooses to cancel | Cancel flows, save offers, exit surveys |
51| **Involuntary** | Payment fails | Dunning emails, smart retries, card updaters |
52 
53Voluntary churn is typically 50-70% of total churn. Involuntary churn is 30-50% but is often easier to fix.
54 
55This skill supports three modes:
56 
571. **Build a cancel flow** — Design from scratch with survey, save offers, and confirmation
582. **Optimize an existing flow** — Analyze cancel data and improve save rates
593. **Set up dunning** — Failed payment recovery with retries and email sequences
60 
61---
62 
63## Cancel Flow Design
64 
65### The Cancel Flow Structure
66 
67Every cancel flow follows this sequence:
68 
69```
70Trigger → Survey → Dynamic Offer → Confirmation → Post-Cancel
71```
72 
73**Step 1: Trigger**
74Customer clicks "Cancel subscription" in account settings.
75 
76**Step 2: Exit Survey**
77Ask why they're cancelling. This determines which save offer to show.
78 
79**Step 3: Dynamic Save Offer**
80Present a targeted offer based on their reason (discount, pause, downgrade, etc.)
81 
82**Step 4: Confirmation**
83If they still want to cancel, confirm clearly with end-of-billing-period messaging.
84 
85**Step 5: Post-Cancel**
86Set expectations, offer easy reactivation path, trigger win-back sequence.
87 
88### Exit Survey Design
89 
90The exit survey is the foundation. Good reason categories:
91 
92| Reason | What It Tells You |
93|--------|-------------------|
94| Too expensive | Price sensitivity, may respond to discount or downgrade |
95| Not using it enough | Low engagement, may respond to pause or onboarding help |
96| Missing a feature | Product gap, show roadmap or workaround |
97| Switching to competitor | Competitive pressure, understand what they offer |
98| Technical issues / bugs | Product quality, escalate to support |
99| Temporary / seasonal need | Usage pattern, offer pause |
100| Business closed / changed | Unavoidable, learn and let go gracefully |
101| Other | Catch-all, include free text field |
102 
103**Survey best practices:**
104- 1 question, single-select with optional free text
105- 5-8 reason options max (avoid decision fatigue)
106- Put most common reasons first (review data quarterly)
107- Don't make it feel like a guilt trip
108- "Help us improve" framing works better than "Why are you leaving?"
109 
110### Dynamic Save Offers
111 
112The key insight: **match the offer to the reason.** A discount won't save someone who isn't using the product. A feature roadmap won't save someone who can't afford it.
113 
114**Offer-to-reason mapping:**
115 
116| Cancel Reason | Primary Offer | Fallback Offer |
117|---------------|---------------|----------------|
118| Too expensive | Discount (20-30% for 2-3 months) | Downgrade to lower plan |
119| Not using it enough | Pause (1-3 months) | Free onboarding session |
120| Missing feature | Roadmap preview + timeline | Workaround guide |
121| Switching to competitor | Competitive comparison + discount | Feedback session |
122| Technical issues | Escalate to support immediately | Credit + priority fix |
123| Temporary / seasonal | Pause subscription | Downgrade temporarily |
124| Business closed | Skip offer (respect the situation) | — |
125 
126### Save Offer Types
127 
128**Discount**
129- 20-30% off for 2-3 months is the sweet spot
130- Avoid 50%+ discounts (trains customers to cancel for deals)
131- Time-limit the offer ("This offer expires when you leave this page")
132- Show the dollar amount saved, not just the percentage
133 
134**Pause subscription**
135- 1-3 month pause maximum (longer pauses rarely reactivate)
136- 60-80% of pausers eventually return to active
137- Auto-reactivation with advance notice email
138- Keep their data and settings intact
139 
140**Plan downgrade**
141- Offer a lower tier instead of full cancellation
142- Show what they keep vs. what they lose
143- Position as "right-size your plan" not "downgrade"
144- Easy path back up when ready
145 
146**Feature unlock / extension**
147- Unlock a premium feature they haven't tried
148- Extend trial of a higher tier
149- Works best for "not getting enough value" reasons
150 
151**Personal outreach**
152- For high-value accounts (top 10-20% by MRR)
153- Route to customer success for a call
154- Personal email from founder for smaller companies
155 
156### Cancel Flow UI Patterns
157 
158```
159┌─────────────────────────────────────┐
160│  We're sorry to see you go          │
161│                                     │
162│  What's the main reason you're      │
163│  cancelling?                        │
164│                                     │
165│  ○ Too expensive                    │
166│  ○ Not using it enough              │
167│  ○ Missing a feature I need         │
168│  ○ Switching to another tool        │
169│  ○ Technical issues                 │
170│  ○ Temporary / don't need right now │
171│  ○ Other: [____________]            │
172│                                     │
173│  [Continue]                         │
174│  [Never mind, keep my subscription] │
175└─────────────────────────────────────┘
176         ↓ (selects "Too expensive")
177┌─────────────────────────────────────┐
178│  What if we could help?             │
179│                                     │
180│  We'd love to keep you. Here's a    │
181│  special offer:                     │
182│                                     │
183│  ┌───────────────────────────────┐  │
184│  │  25% off for the next 3 months│  │
185│  │  Save $XX/month               │  │
186│  │                               │  │
187│  │  [Accept Offer]               │  │
188│  └───────────────────────────────┘  │
189│                                     │
190│  Or switch to [Basic Plan] at       │
191│  $X/month →                         │
192│                                     │
193│  [No thanks, continue cancelling]   │
194└─────────────────────────────────────┘
195```
196 
197**UI principles:**
198- Keep the "continue cancelling" option visible (no dark patterns)
199- One primary offer + one fallback, not a wall of options
200- Show specific dollar savings, not abstract percentages
201- Use the customer's name and account data when possible
202- Mobile-friendly (many cancellations happen on mobile)
203 
204For detailed cancel flow patterns by industry and billing provider, see [references/cancel-flow-patterns.md](references/cancel-flow-patterns.md).
205 
206---
207 
208## Churn Prediction & Proactive Retention
209 
210The best save happens before the customer ever clicks "Cancel."
211 
212### Risk Signals
213 
214Track these leading indicators of churn:
215 
216| Signal | Risk Level | Timeframe |
217|--------|-----------|-----------|
218| Login frequency drops 50%+ | High | 2-4 weeks before cancel |
219| Key feature usage stops | High | 1-3 weeks before cancel |
220| Support tickets spike then stop | High | 1-2 weeks before cancel |
221| Email open rates decline | Medium | 2-6 weeks before cancel |
222| Billing page visits increase | High | Days before cancel |
223| Team seats removed | High | 1-2 weeks before cancel |
224| Data export initiated | Critical | Days before cancel |
225| NPS score drops below 6 | Medium | 1-3 months before cancel |
226 
227### Health Score Model
228 
229Build a simple health score (0-100) from weighted signals:
230 
231```
232Health Score = (
233  Login frequency score × 0.30 +
234  Feature usage score   × 0.25 +
235  Support sentiment     × 0.15 +
236  Billing health        × 0.15 +
237  Engagement score      × 0.15
238)
239```
240 
241| Score | Status | Action |
242|-------|--------|--------|
243| 80-100 | Healthy | Upsell opportunities |
244| 60-79 | Needs attention | Proactive check-in |
245| 40-59 | At risk | Intervention campaign |
246| 0-39 | Critical | Personal outreach |
247 
248### Proactive Interventions
249 
250**Before they think about cancelling:**
251 
252| Trigger | Intervention |
253|---------|-------------|
254| Usage drop >50% for 2 weeks | "We noticed you haven't used [feature]. Need help?" email |
255| Approaching plan limit | Upgrade nudge (not a wall — paywall-upgrade-cro handles this) |
256| No login for 14 days | Re-engagement email with recent product updates |
257| NPS detractor (0-6) | Personal follow-up within 24 hours |
258| Support ticket unresolved >48h | Escalation + proactive status update |
259| Annual renewal in 30 days | Value recap email + renewal confirmation |
260 
261---
262 
263## Involuntary Churn: Payment Recovery
264 
265Failed payments cause 30-50% of all churn but are the most recoverable.
266 
267### The Dunning Stack
268 
269```
270Pre-dunning → Smart retry → Dunning emails → Grace period → Hard cancel
271```
272 
273### Pre-Dunning (Prevent Failures)
274 
275- **Card expiry alerts**: Email 30, 15, and 7 days before card expires
276- **Backup payment method**: Prompt for a second payment method at signup
277- **Card updater services**: Visa/Mastercard auto-update programs (reduces hard declines 30-50%)
278- **Pre-billing notification**: Email 3-5 days before charge for annual plans
279 
280### Smart Retry Logic
281 
282Not all failures are the same. Retry strategy by decline type:
283 
284| Decline Type | Examples | Retry Strategy |
285|-------------|----------|----------------|
286| Soft decline (temporary) | Insufficient funds, processor timeout | Retry 3-5 times over 7-10 days |
287| Hard decline (permanent) | Card stolen, account closed | Don't retry — ask for new card |
288| Authentication required | 3D Secure, SCA | Send customer to update payment |
289 
290**Retry timing best practices:**
291- Retry 1: 24 hours after failure
292- Retry 2: 3 days after failure
293- Retry 3: 5 days after failure
294- Retry 4: 7 days after failure (with dunning email escalation)
295- After 4 retries: Hard cancel with reactivation path
296 
297**Smart retry tip:** Retry on the day of the month the payment originally succeeded (if Day 1 worked before, retry on Day 1). Stripe Smart Retries handles this automatically.
298 
299### Dunning Email Sequence
300 
301| Email | Timing | Tone | Content |
302|-------|--------|------|---------|
303| 1 | Day 0 (failure) | Friendly alert | "Your payment didn't go through. Update your card." |
304| 2 | Day 3 | Helpful reminder | "Quick reminder — update your payment to keep access." |
305| 3 | Day 7 | Urgency | "Your account will be paused in 3 days. Update now." |
306| 4 | Day 10 | Final warning | "Last chance to keep your account active." |
307 
308**Dunning email best practices:**
309- Direct link to payment update page (no login required if possible)
310- Show what they'll lose (their data, their team's access)
311- Don't blame ("your payment failed" not "you failed to pay")
312- Include support contact for help
313- Plain text performs better than designed emails for dunning
314 
315### Recovery Benchmarks
316 
317| Metric | Poor | Average | Good |
318|--------|------|---------|------|
319| Soft decline recovery | <40% | 50-60% | 70%+ |
320| Hard decline recovery | <10% | 20-30% | 40%+ |
321| Overall payment recovery | <30% | 40-50% | 60%+ |
322| Pre-dunning prevention | None | 10-15% | 20-30% |
323 
324For the complete dunning playbook with provider-specific setup, see [references/dunning-playbook.md](references/dunning-playbook.md).
325 
326---
327 
328## Metrics & Measurement
329 
330### Key Churn Metrics
331 
332| Metric | Formula | Target |
333|--------|---------|--------|
334| Monthly churn rate | Churned customers / Start-of-month customers | <5% B2C, <2% B2B |
335| Revenue churn (net) | (Lost MRR - Expansion MRR) / Start MRR | Negative (net expansion) |
336| Cancel flow save rate | Saved / Total cancel sessions | 25-35% |
337| Offer acceptance rate | Accepted offers / Shown offers | 15-25% |
338| Pause reactivation rate | Reactivated / Total paused | 60-80% |
339| Dunning recovery rate | Recovered / Total failed payments | 50-60% |
340| Time to cancel | Days from first churn signal to cancel | Track trend |
341 
342### Cohort Analysis
343 
344Segment churn by:
345- **Acquisition channel** — Which channels bring stickier customers?
346- **Plan type** — Which plans churn most?
347- **Tenure** — When do most cancellations happen? (30, 60, 90 days?)
348- **Cancel reason** — Which reasons are growing?
349- **Save offer type** — Which offers work best for which segments?
350 
351### Cancel Flow A/B Tests
352 
353Test one variable at a time:
354 
355| Test | Hypothesis | Metric |
356|------|-----------|--------|
357| Discount % (20% vs 30%) | Higher discount saves more | Save rate, LTV impact |
358| Pause duration (1 vs 3 months) | Longer pause increases return rate | Reactivation rate |
359| Survey placement (before vs after offer) | Survey-first personalizes offers | Save rate |
360| Offer presentation (modal vs full page) | Full page gets more attention | Save rate |
361| Copy tone (empathetic vs direct) | Empathetic reduces friction | Save rate |
362 
363**How to run cancel flow experiments:** Use the **ab-test-setup** skill to design statistically rigorous tests. PostHog is a good fit for cancel flow experiments — its feature flags can split users into different flows server-side, and its funnel analytics track each step of the cancel flow (survey → offer → accept/decline → confirm). See the [PostHog integration guide](../../tools/integrations/posthog.md) for setup.
364 
365---
366 
367## Common Mistakes
368 
369- **No cancel flow at all** — Instant cancel leaves money on the table. Even a simple survey + one offer saves 10-15%
370- **Making cancellation hard to find** — Hidden cancel buttons breed resentment and bad reviews. Many jurisdictions require easy cancellation (FTC Click-to-Cancel rule)
371- **Same offer for every reason** — A blanket discount doesn't address "missing feature" or "not using it"
372- **Discounts too deep** — 50%+ discounts train customers to cancel-and-return for deals
373- **Ignoring involuntary churn** — Often 30-50% of total churn and the easiest to fix
374- **No dunning emails** — Letting payment failures silently cancel accounts
375- **Guilt-trip copy** — "Are you sure you want to abandon us?" damages brand trust
376- **Not tracking save offer LTV** — A "saved" customer who churns 30 days later wasn't really saved
377- **Pausing too long** — Pauses beyond 3 months rarely reactivate. Set limits.
378- **No post-cancel path** — Make reactivation easy and trigger win-back emails, because some churned users will want to come back
379 
380---
381 
382## Tool Integrations
383 
384For implementation, see the [tools registry](../../tools/REGISTRY.md).
385 
386### Retention Platforms
387 
388| Tool | Best For | Key Feature |
389|------|----------|-------------|
390| **Churnkey** | Full cancel flow + dunning | AI-powered adaptive offers, 34% avg save rate |
391| **ProsperStack** | Cancel flows with analytics | Advanced rules engine, Stripe/Chargebee integration |
392| **Raaft** | Simple cancel flow builder | Easy setup, good for early-stage |
393| **Chargebee Retention** | Chargebee customers | Native integration, was Brightback |
394 
395### Billing Providers (Dunning)
396 
397| Provider | Smart Retries | Dunning Emails | Card Updater |
398|----------|:------------:|:--------------:|:------------:|
399| **Stripe** | Built-in (Smart Retries) | Built-in | Automatic |
400| **Chargebee** | Built-in | Built-in | Via gateway |
401| **Paddle** | Built-in | Built-in | Managed |
402| **Recurly** | Built-in | Built-in | Built-in |
403| **Braintree** | Manual config | Manual | Via gateway |
404 
405### Related CLI Tools
406 
407| Tool | Use For |
408|------|---------|
409| `stripe` | Subscription management, dunning config, payment retries |
410| `customer-io` | Dunning email sequences, retention campaigns |
411| `posthog` | Cancel flow A/B tests via feature flags, funnel analytics |
412| `mixpanel` / `ga4` | Usage tracking, churn signal analysis |
413| `segment` | Event routing for health scoring |
414 
415---
416 
417## Related Skills
418 
419- **email-sequence**: For win-back email sequences after cancellation
420- **paywall-upgrade-cro**: For in-app upgrade moments and trial expiration
421- **pricing-strategy**: For plan structure and annual discount strategy
422- **onboarding-cro**: For activation to prevent early churn
423- **analytics-tracking**: For setting up churn signal events
424- **ab-test-setup**: For testing cancel flow variations with statistical rigor
425