How did monthly recurring revenue change this period, and where did the change come from?
№ 002 · MRR waterfall
Definition
MetricNet MRR movement in period: opening MRR + new + expansion - contraction - churn = closing MRR
Unitcurrency per month; ratio of gains to losses
Bars walk from opening MRR through additions (new, expansion, reactivation) and subtractions (contraction, churn) to closing MRR. Build by tagging every billing event and summing dollars per category. Largest negative bar is your biggest leak. Example: Start $487K, +$42K new, +$18K expansion, -$19K churn, End $523K.
- MRR
- Monthly recurring revenue. Sum of every active subscription normalised to a monthly value, excluding one-off fees.
Benchmarks
| Bottom 30% | Median | Top 30% |
|---|---|---|
| Gains under 1.5x losses; waterfall periodically closes below opening MRR | Gains roughly 1.8-2.5x losses; the industry-wide ratio of adds to losses was ~1.82 in March 2024 on ChartMogul data, down from 2.55 at the 2021 peak | Gains (new + expansion) at least 4x losses (churn + contraction), the classic quick-ratio-4 bar; churn bar under 1% of opening MRR |
Private SaaS, all ARR bands, billing-data derived, 2021-2024 period; gains-to-losses definition per Mamoon Hamid quick ratio (new + expansion MRR over churned + contraction MRR) · The 4x bar is an investor heuristic from 2015, set when growth was more new-logo driven; scaled companies at 2.5-3.5x with strong GRR are often healthier than a small company above 4x. The ratio benchmark is the same family as card 016; this card reads it inside the waterfall.
Category split omitted: Waterfall-component ratios by segment or ACV are not published by two independent Tier 1-2 sources.
When it looks bad
The downward churn and contraction steps grow period over period until they rival the new-business step, leaving a closing bar barely above opening despite visible acquisition activity.
Opening MRR $500k, new $40k, expansion $10k, contraction $15k, churn $30k: closing $505k. Gains-to-losses ratio of 1.1x means the company replaced 90% of what it sold.
What to do about it
- Attack the contraction step separately from churn: contraction is usually seat reductions or plan downgrades, best countered with usage-based or per-outcome pricing components that let value re-expand; OpenView survey data showed usage-based companies at 125% NRR vs 115% for subscription peers.
- Recover involuntary churn inside the churn step: card updater plus tuned retries moves recovery from ~53% to ~71% of failed payments (Recurly), and 90% of what will be recovered lands in the first 10 days, so front-load the sequence.
- If the expansion step is thin, price the second unit of value: add-on modules, extra seats, or usage tiers; expansion ARR costs roughly $1.40 of S&M per dollar vs $2.00 for new-logo ARR (Benchmarkit 2025).
- Rebuild the waterfall quarterly by cohort to see whether losses come from a specific vintage or price plan rather than treating churn as one undifferentiated bar.
Sources
- ChartMogul Quick ratio of 4+ signals efficient growth (Hamid's investment bar); 2-4 is average; below 2 churn is eating growth chartmogul.com ↗
- The Founder's Corner (reporting ChartMogul data) Industry-wide SaaS quick ratio peaked ~2.55 in Sep 2021 and declined to ~1.82 by Mar 2024 the-founders-corner.com ↗
- SaaS Capital Moving NRR from the 90-100% band to 100-110% improves growth rate by ~5 percentage points saas-capital.com ↗
- Recurly Research Monthly churn by price tier: 4.29% total at $10-25/mo down to 2.87% at $100-250/mo; involuntary share shrinks with price recurly.com ↗
- Recurly Optimized retry strategies lifted failed-payment recovery from ~53% to ~71%; 90% of recoveries happen within 10 days of failure recurly.com ↗
- Tomasz Tunguz (Theory Ventures) Quick ratio restated as (growth rate + churn rate) / churn rate; a 4x ratio at 15% monthly growth implies tolerating 5% monthly churn, so high growth can mask high churn tomtunguz.com ↗