Retention and Churn
The most important metric nobody talks about enough. Acquisition gets the headlines; retention determines survival. A startup that acquires 1,000 users/month but retains only 10% is shrinking. A startup that acquires 100 users/month but retains 90% is compounding.
Why Retention > Acquisition
- Compounding: Retained users accumulate. Churned users disappear. Small differences in retention compound into massive differences over time.
- Unit economics: Retained users have higher LTV. Churned users waste the CAC you spent acquiring them.
- Signal: Retention IS product-market-fit. If users stay, you have PMF. If they leave, you don’t. No amount of marketing fixes a retention problem.
- Growth math: Net growth = new users - churned users. If churn is high, you need enormous acquisition just to stand still (the “leaky bucket”).
Altman: “Avoid vanity metrics — focus on retention alongside acquisition.”
Measuring Retention
Cohort Retention Curves
The most important chart in your startup. Group users by when they joined (cohort), then plot what percentage is still active over time.
Good curve: Drops initially, then flattens. Users who survive the first 30 days tend to stay. Bad curve: Drops continuously. Users never stop leaving. Great curve: Drops, flattens, then curves UP (users become more engaged over time — this is rare and incredibly valuable).
Churn Rate
| Metric | Formula | Healthy Range |
|---|---|---|
| Monthly churn | Customers lost ÷ customers at start of month | <5% (B2C), <2% (B2B SaaS) |
| Annual churn | 1 - (1 - monthly churn)^12 | <30% (B2C), <15% (B2B SaaS) |
| Net revenue retention | (Starting MRR + expansion - contraction - churn) ÷ Starting MRR | >100% = growing from existing customers |
| Logo churn | # of customers lost ÷ # at start of period | Higher than revenue churn if small customers leave |
Net revenue retention > 100% means your existing customers are spending MORE over time (expansion > churn). This is the gold standard — companies like Snowflake, Twilio, and Datadog achieve 120-150%+ NRR.
The Rule of 40
A combined health metric: growth rate + profit margin ≥ 40%.
- A company growing 60% with -20% margins = 40 ✓
- A company growing 10% with 30% margins = 40 ✓
- A company growing 20% with 10% margins = 30 ✗
Retention directly impacts both sides: better retention → faster growth AND better margins.
Why Users Churn
Product Reasons
- Never activated: User signed up but never experienced core value (activation failure)
- Lost the habit: Used it once, didn’t return. Product didn’t embed in daily workflow.
- Outgrew the product: Needs evolved beyond what you offer
- Quality degradation: Bugs, slow performance, poor UX changes
Market Reasons
- Competitor offers more: A rival shipped a better version of what you do
- Market shift: The problem you solve became less important
- Price sensitivity: Economic conditions change willingness to pay
Self-Inflicted
- Billing failures: Credit card expires, payment fails, account gets suspended (involuntary churn — often 20-40% of total churn)
- Poor onboarding: Users never learn how to get value from the product
- Ignoring feedback: Users asked for something, you didn’t build it, they left
Reducing Churn: The Playbook
Pre-Churn: Activation
The biggest lever. Users who never activate will always churn.
- Define your “aha moment” — the specific action that predicts retention
- Measure time-to-value: how fast do new users reach the aha moment?
- For PLG products: onboarding IS the product experience
- Vohra’s PMF engine: the “somewhat disappointed” users are your activation opportunity
Active Churn: Re-Engagement
Before a user leaves:
- Identify leading indicators (usage drops, support tickets increase, login frequency decreases)
- Automated re-engagement: emails, in-app messages, personalized recommendations
- Human outreach for high-value accounts: “We noticed you haven’t been using X — can we help?”
- “Win-back” campaigns for recently churned users (easiest to recover within 30 days)
Involuntary Churn: Payment Recovery
Often overlooked, often 20-40% of total churn:
- Dunning emails (automated payment retry + user notification)
- Smart retry logic (retry at optimal times)
- Card updater services (automatically update expired cards)
- Graceful degradation (don’t immediately lock out users — give them time to fix payment)
Structural Retention: Switching Costs
Build switching costs into the product:
- Data accumulation: the more they use it, the more valuable their data becomes
- Workflow integration: embed into daily processes so leaving means relearning
- Team adoption: one user’s decision to leave is constrained by their team’s dependency
- Content creation: user-generated content, configurations, templates become assets they can’t take with them
Retention by Business Model
| Model | Key Retention Lever | Watch For |
|---|---|---|
| SaaS | Workflow embedding, team adoption | Monthly churn >5% |
| Marketplace | Supply quality, liquidity | Seller churn (supply loss is catastrophic) |
| Consumer app | Habit formation, social graph | Day 1 / Day 7 / Day 30 retention curves |
| E-commerce | Repeat purchase rate, loyalty | One-time buyers never returning |
| Freemium | Free-to-paid conversion, then paid retention | Free tier churn is expected; paid churn kills |
Max MRR: Your Revenue Ceiling
Jason Cohen’s most important formula for SaaS:
Max MRR = New MRR ÷ Monthly Churn Rate
| New MRR/month | Churn Rate | Max MRR (ceiling) |
|---|---|---|
| $1,000 | 7% | $14,286 |
| $1,000 | 5% | $20,000 |
| $1,000 | 4% | $25,000 |
| $1,000 | 2% | $50,000 |
Reducing churn from 7% → 4% increases your ceiling by 75%. The same improvement through acquisition alone would require nearly doubling your sales effort. Churn reduction is almost always higher leverage than acquisition growth.
This formula predicts revenue plateaus months before they happen. If you’re adding $1K/mo but churning at 5%, you’ll plateau at $20K MRR regardless of market size. The only exits: improve retention or increase new MRR.
The Churn Death Spiral
When churn exceeds a threshold, it becomes self-reinforcing:
- Users leave → less community/content/value → more users leave
- Revenue drops → less investment in product → quality degrades → more churn
- Best customers leave first (they have the most options) → remaining base is lower quality
- NRR drops below 100% → company shrinks even without losing logo count
The only cure: fix the root cause (usually product or activation), not the symptom (usually more acquisition spend).
See Also
- startup-metrics
- unit-economics
- product-market-fit
- product-led-growth
- product-development
- growth
- moats
Sources
- Startup Playbook — Sam Altman
- Superhuman’s PMF Engine — First Round Review
- The Dynamics of Network Effects — a16z