Separate churn by failure mode and build targeted interventions
Most churn is a composite number. This framework separates it into distinct, actionable archetypes.
Treating churn as a single metric leads to generic solutions. When monthly churn goes up, teams often improve onboarding, hire CSMs, or offer discounts. Six months later, churn is roughly the same. The effort was real, but it was applied to a composite number containing 4 or 5 different problems—each with a different root cause, signal, and intervention.
Classifying churn into archetypes turns a vague problem into a clear diagnosis. The treatment becomes obvious.
Failed Activation typically accounts for 15-40% of total churn at B2B SaaS companies. A 40% drop in weekly logins predicts churn with 78% accuracy—the strongest individual predictor.
| Archetype | % of Total Churn | Root Cause | Primary Signal |
|---|---|---|---|
| Failed Activation | 15-40% | User never reached first value moment | Low week 1 activity, setup skipped |
| Expectation Gap | 20-30% | Product solves different problem than customer expected | Early engagement, then wall; support tickets about missing features |
| Price Pressure | 15-25% | Value clear, price doesn't align to perceived ROI | Usage high, renewal conversations focus on cost |
| Circumstances Changed | 10-15% | Budget cut, company acquired, champion left | Sudden drop, often with external trigger noted in CRM |
| Involuntary (Billing) | 20-48% | Payment failure (card expired, insufficient funds) | Failed payment event, no prior activity drop |
Analyze your last 50 churned customers. Tally them by archetype. Use notes, support tickets, and payment logs to classify.
Failed Activation: _____ customers (_____%)
Expectation Gap: _____ customers (_____%)
Price Pressure: _____ customers (_____%)
Circumstances Changed: _____ customers (_____%)
Involuntary (Billing): _____ customers (_____%)
Other / Unknown: _____ customers (_____%)
What is your single largest churn archetype? Where should you focus first?
Identify customers who bought the wrong product for their job—before they churn.
An Expectation Gap occurs when a customer signs up expecting a capability that either does not exist, does not work as expected, or requires significantly more configuration than they were led to believe. The source is often the sales conversation or marketing positioning.
This archetype has a distinct fingerprint: the customer is engaged early, asks detailed questions, maybe even completes onboarding—and then hits a wall. They file support tickets expressing confusion or disappointment. They churn feeling misled.
Expectation Gap churn accounts for 20-30% of total churn. The fix is upstream in sales enablement and positioning, not downstream in customer success.
For your last 10 customers who churned citing "not what we expected" or "missing feature," answer the following:
1. What was the promised capability?
2. Where was that promise made? (e.g., sales demo, website feature list, case study)
3. What was the actual product reality?
4. What is the gap between promise and reality? (Scale: 1=Minor, 5=Fundamental)
Use this checklist during sales qualification to identify potential fit gaps before the deal closes.
Based on your audit, list the top 3 expectation gaps to fix in sales/marketing materials:
Spot customers who never get started—and intervene before they quietly leave.
Failed Activation means the customer never reached their first meaningful value moment. They signed up, poked around, and quietly stopped engaging. By the time they cancel, the decision was made weeks or months ago.
This churn often appears at months 2‑4, misleading teams into thinking something went wrong mid‑lifecycle. The problem was in week 1. The fix lives in onboarding.
The week‑over‑week change in total user activity. A drop of 40% predicts churn with 78% accuracy.
Velocity = (events_this_week - events_last_week) / events_last_week * 100
Number of distinct core features used in first 14 days. Customers who use only 1 feature are 3x more likely to churn by month 3.
| Activation Stage | Healthy Signal | Warning Signal | Critical Signal |
|---|---|---|---|
| Week 1 | Completed setup guide; used 2+ core features | Started but did not finish setup; used 1 feature | Logged in once, zero feature usage |
| Week 2-4 | Weekly activity stable or growing | Activity decline 20-50% | Activity decline >50% |
| Month 2 | First value milestone achieved (e.g., report generated, workflow automated) | Milestone not achieved, but some activity continues | No logins for 14+ days |
For a recent cohort of customers (e.g., last 30 signups), calculate these metrics.
% who completed setup: _____%
% who used 2+ core features in week 1: _____%
% who achieved first value milestone by day 30: _____%
Average engagement velocity (week 1 → 2): _____%
% with >50% activity drop by week 3: _____%
Cohort churn rate by month 3: _____%
What is the single biggest drop‑off point in your activation funnel?
Quantify the value delivered—or missing—before price becomes the reason to leave.
Price Pressure churn occurs when the value is clear, but the price doesn't align to the customer's perceived ROI. This is common in the SMB segment. The solution is not automatic discounting—it's demonstrating and quantifying value before the renewal conversation.
Customers who cannot articulate the value they receive are likely to question the cost. Your goal is to make the ROI undeniable.
Price Pressure accounts for 15-25% of total churn. Show ROI before discounting.
Work with your CS team to fill this out for at‑risk accounts 60 days before renewal.
Account: _________________________________
Annual Contract Value: $_____
Primary Use Case: _________________________________
Quantifiable Value Metrics
Total Monthly Value: $_____
Monthly Cost: $_____
ROI Ratio (Value/Cost): _____:1
If the ROI ratio is below 3:1, what additional value could this customer capture before renewal?
Use this in renewal preparation.
Systematically monitor for early warnings across all churn archetypes.
A health score combines behavioral, commercial, and sentiment signals into a single number that ranks accounts by churn risk. The best scores use 3–5 signals, not 20. They should update weekly, not daily, to smooth noise while catching meaningful trends.
The score is only valuable if it triggers action. A score of 35 means nothing without a playbook: "At 35, the CSM schedules a value review within 48 hours."
Weight: 25%. Week‑over‑week change in total activity.
> -20%: Healthy | -20% to -50%: Warning | < -50%: Critical
Weight: 20%. Number of distinct core features used last 30 days.
<2: Critical | 2-3: Warning | >3: Healthy
Weight: 20%. Ticket frequency & sentiment; spike in negative tickets.
Weight: 35%. Failed payment, seat reduction, contract end within 60 days.
For each at‑risk account (health score < 50), complete this checklist.
Based on this audit, what is the most likely churn archetype for this account?
What is the prescribed intervention? (Refer to Section 6: Intervention Design Matrix)
Match the right action to the diagnosed churn type.
Generic interventions waste resources. A billing failure requires a dunning sequence, not a CSM call. An expectation gap requires sales enablement, not an onboarding tweak. This matrix maps archetypes to specific, measured interventions.
| Churn Archetype | Primary Intervention | Owner | Success Metric |
|---|---|---|---|
| Failed Activation | Week 1 onboarding sequence; automated nudges for incomplete setup; milestone celebration. | Product / Marketing | % achieving first value milestone |
| Expectation Gap | Sales enablement on actual capabilities; update marketing materials; pre‑sales qualification checklist. | Sales / Marketing | Churn rate from customers with documented use‑case fit |
| Price Pressure | ROI report 60 days pre‑renewal; value‑expansion plan; tier adjustment. | Customer Success | Renewal rate among accounts receiving ROI report |
| Circumstances Changed | Identify early (champion departure tracking); offer pause/hold option; archive and re‑engage later. | Customer Success | % successfully paused vs. lost |
| Involuntary (Billing) | Dunning system: failure taxonomy, retry logic, card updater, pre‑expiry notifications. | Finance / Operations | Recovery rate (median is 47.6%) |
Select your top churn archetype from Section 1. Design a 90‑day intervention plan.
Target Archetype: _________________________________
Current % of total churn: _____%
Target reduction (in 90 days): _____%
Key Actions (Who, What, By When):
Success Metrics:
Track the impact of your interventions over time with cohort‑based measurement.
Cohort analysis isolates the effect of a change by comparing groups of customers who signed up before and after an intervention. Aggregate churn rates can be misleading—a new marketing campaign might bring in lower‑fit customers, raising churn even if your product improvements are working. Cohort analysis reveals the truth.
The recovery rate metric: If 70%+ of customers who hit friction resume meaningful usage within 48 hours, your product relationship is healthy. Below 60%, customers are building workarounds or mentally cataloging reasons to leave.
Define your intervention cohort and track its performance against a baseline.
Intervention: _________________________________
Start Date: _________________________________
Cohort Definition: All customers who signed up on or after start date.
| Metric | Baseline Cohort (Pre‑Intervention) | Intervention Cohort | Delta |
|---|---|---|---|
| Day 7 Activation Rate | _____% | _____% | _____ pp |
| Month 1 Retention | _____% | _____% | _____ pp |
| Month 3 Churn Rate | _____% | _____% | _____ pp |
| Average Revenue Per Account (ARPA) | $_____ | $_____ | _____% |
Based on this data, is the intervention working? What will you do next?
A step‑by‑step guide to prioritizing your churn reduction efforts for maximum impact.
You've diagnosed your churn archetypes. You have a list of potential interventions. Where do you start? This decision tree helps you prioritize based on impact, effort, and speed of results.
Rule: Fix involuntary churn first. It's the fastest, cheapest win. Then address the largest voluntary archetype where you have the most control.
Yes → Implement a dunning system. Median recovery rate is 47.6%. This is a 2‑week infrastructure fix. Do this first.
No → Proceed to step 2.
Refer to your classification from Section 1. Rank them by percentage.
Yes → Proceed to step 4.
No → Go back to Sections 2‑5 to diagnose signals and design an intervention.
Yes → Run a cohort test (Section 7).
No → Break the intervention into a smaller pilot. Start with a single segment or feature.
Use the decision tree and your earlier worksheets to complete this plan.
Quarter: _________________________________
Primary Goal: Reduce total churn from _____% to _____%.
Priority 1 (Weeks 1‑4):
Priority 2 (Weeks 5‑8):
Priority 3 (Weeks 9‑12):
From guessing to knowing.
productquant.dev