1. Scope
Estimates recovered customers and revenue lift when monthly churn improves. It illustrates the sensitivity of LTV to churn; it is not a retention-programme design tool.
2. Inputs and outputs
Inputs
- customers number
- arpu number (currency/mo)
- currentChurn percent (monthly)
- targetChurn percent (monthly)
Outputs
- currentLtv
arpu / currentChurn.
- targetLtv
arpu / targetChurn.
- ltvLift
targetLtv − currentLtv.
- recoveredCustomersYearOne
Customers × (currentChurn − targetChurn) × 12.
Engine source: src/lib/churn-retention-calculator/engine.ts
3. Formula / scoring logic
current_ltv = arpu / current_churn
target_ltv = arpu / target_churn
ltv_lift = target_ltv - current_ltv4. Assumptions
- Churn is memoryless (exponential decay). Real SaaS retention curves are often logarithmic, giving longer tail than this formula suggests.
- ARPU is flat; no expansion-revenue tailwind.
- The recovered-customer figure is a steady-state difference, not a behavioural projection.
5. Data sources
- OpenView SaaS Benchmarks 2024 (churn percentiles) as of 2024
- Paddle SaaS Benchmarks 2024 as of 2024
6. Known limitations
- The widely-cited Reichheld "5% retention lift = 25–95% profit lift" claim is context-dependent and not peer-reviewed. We do not use it as a benchmark. Consult the underlying Harvard Business School working paper directly if needed.
- Treats logo churn and revenue churn as equivalent; they diverge for products with tiered pricing.
7. Reproducibility
Input
customers = 1000, arpu = $30, currentChurn = 6%, targetChurn = 4%.
Expected output
current_ltv = $500, target_ltv = $750, ltv_lift = $250, recoveredCustomers year 1 = 240.
8. Change log
- 2026-04-24 methodology page first published.