SaaS Subscription Metrics: MRR, Churn, ARPA, and the Math That Connects Them
October 21, 2026
Subscription businesses live on a few recurring numbers. Monthly recurring revenue (MRR) is the predictable revenue each month. Average revenue per account (ARPA) is MRR divided by the number of paying customers. Churn is the share of customers or revenue lost each month. Together they determine how fast the business grows and how long customers stay.
The lifetime of a customer is roughly 1 ÷ monthly churn. At 3% monthly churn, that is 33 months; at 5%, 20 months; at 1%, 100 months. This makes churn the most powerful number in the model.
Skip the math and get your answer instantly:
Open the Compound Interest Calculator →Key Definitions
MRR = number of customers × ARPA. If 500 customers pay an average of $100 a month, MRR is $50,000, and annual recurring revenue (ARR) is $600,000. Customer churn rate = customers lost in the month ÷ customers at the start of the month. Revenue churn also counts downgrades, and net revenue retention (NRR) includes expansion from upsells. An NRR above 100% means that existing customers generate more revenue over time even without new sales.
Net new MRR = new MRR + expansion MRR − churned MRR − contraction MRR.
A Monthly Projection
Suppose a company starts with 500 customers, adds 40 per month, and loses 3% of its customers each month. Month 1: lose 15 and add 40, so 525. Month 2: lose 15.75, add 40, so 549. The customer count moves toward a steady state where adds equal churn: 40 ÷ 0.03 = 1,333 customers. After 12 months, the count is about 755, after 24 months about 932, and after 36 months about 1,055, approaching but never quite reaching 1,333.
To grow faster than that ceiling, either increase new customer adds or reduce churn. Cutting churn to 2% raises the ceiling to 2,000 customers.
| Month | Customers (adds 40/mo, 3% churn) | MRR at $100 ARPA |
|---|---|---|
| 0 | 500 | $50,000 |
| 12 | 755 | $75,500 |
| 24 | 932 | $93,200 |
| 36 | 1,055 | $105,500 |
| Steady state | 1,333 | $133,300 |
Why Churn Compounds
A 3% monthly churn loses 31% of customers in a year (1 − 0.97¹² = 30.6%). A 5% monthly churn loses 46%. A 1% monthly churn loses 11.4%. These differences cascade into lifetime value: with $80 of gross profit per account per month, LTV at 3% churn is $2,667, at 5% is $1,600, and at 1% is $8,000.
Improving retention even by a point often delivers more value than a similar proportional increase in acquisition.
Cohorts and Quality of Churn
Aggregate churn can hide the pattern. Cohort analysis tracks groups of customers by signup month and shows how many remain after each month. Many businesses lose a high share in the first months, then stabilize. If new customers churn much faster than older ones, improving onboarding is the lever. Distinguish voluntary churn (customers who choose to leave) from involuntary churn (failed payments), which is often fixable with better billing retries.
Sanity Checks
Compare your metrics with benchmarks for your segment. Consumer subscriptions may see monthly churn of 5% to 10%, small-business software 3% to 7%, and enterprise software under 1% monthly. Growth rates, payback periods, and gross margins should all be consistent: if your model implies customers staying for 10 years but your oldest cohort is 2 years old, treat that with skepticism.
Frequently Asked Questions
What is MRR?
Monthly recurring revenue: predictable revenue from subscriptions each month.
How do I calculate customer lifetime from churn?
Divide 1 by the monthly churn rate. At 3% churn, about 33 months.
What is net revenue retention?
The percentage of revenue retained from existing customers, including upgrades and net of downgrades and cancellations.
What is a good churn rate?
Depends on the segment. Under 1% monthly is common for enterprise software, while consumer apps see far higher churn.
Why is churn so important?
It determines customer lifetime and the maximum size a business can reach with a given acquisition rate.