SaaS Churn Calculation: 4 Formulas That Work (2026)

SaaS churn calculation looks simple until two teams produce two different answers from the same customer data. One team counts cancelled accounts. Another counts lost recurring revenue. A third subtracts expansion revenue and calls the result churn. All three numbers can be valid, but they answer different questions.

The fix is to define the population, period, and revenue treatment before you calculate anything. This guide gives you four useful formulas, worked examples, and a reporting checklist you can reuse every month.

The short answer: how to calculate SaaS churn

For basic customer churn, divide customers lost during a period by customers active at the start of that period, then multiply by 100.

Customer churn rate = customers lost during period ÷ customers at start of period × 100

If you started July with 500 paying customers and 20 cancelled before month-end, monthly customer churn was 4%. Do not include customers acquired during July in the denominator. They were not exposed to the full measurement period.

That is the default SaaS churn calculation. Use revenue churn when account values differ, net revenue churn when expansion matters, and annualized churn when you need a yearly planning view.

Before calculating churn, freeze four definitions

Churn disputes are usually definition disputes wearing spreadsheet clothes. Agree on these rules before you open Excel or query your database.

1. Choose the unit

Use customer churn to measure lost logos or accounts. Use revenue churn to measure lost recurring revenue. If one enterprise account is worth as much as 50 small accounts, customer churn alone can hide the financial impact.

2. Choose the time window

Monthly churn is useful for operational reviews. Quarterly churn reduces noise when your customer base is small. Match the period to your billing cycle and keep it consistent across reports.

3. Define when a customer has churned

Pick one event: cancellation date, subscription end date, failed-payment write-off, or deletion. For most subscription businesses, the subscription end date is cleaner because it reflects when access and recurring revenue actually stop.

4. Separate voluntary and involuntary churn

Voluntary churn comes from an active cancellation. Involuntary churn comes from failed payments, expired cards, or billing errors. Combine them for the headline rate, but keep the split underneath because the remedies differ.

Formula 1: customer churn rate

Customer churn tells you what share of the starting customer base left during the period.

Customer churn rate = churned customers ÷ starting customers × 100

Example: You begin August with 800 paying customers. During August, 24 customers reach the end of their subscriptions and do not renew. Customer churn is 24 ÷ 800 × 100 = 3%.

New customers do not reduce this rate. Adding 100 new accounts may grow your total customer count, but it does not undo the fact that 3% of the customers who could churn did churn.

Formula 2: gross revenue churn

Gross revenue churn measures recurring revenue lost from cancellations and downgrades. It ignores expansion, so it shows the full leakage in your existing revenue base.

Gross revenue churn = (churned MRR + downgrade MRR) ÷ starting MRR × 100

Example: You start with $100,000 in monthly recurring revenue. Cancellations remove $4,000 and downgrades remove $1,000. Gross revenue churn is ($4,000 + $1,000) ÷ $100,000 × 100 = 5%.

This is often more useful than customer churn for pricing and segment decisions. Losing ten $20 accounts is not the same commercial problem as losing one $20,000 account.

Formula 3: net revenue churn

Net revenue churn subtracts expansion from lost recurring revenue. Expansion includes upgrades, added seats, and cross-sells from customers who were already active at the start of the period.

Net revenue churn = (churned MRR + downgrade MRR − expansion MRR) ÷ starting MRR × 100

Using the previous example, suppose existing customers also added $6,000 in MRR. Net revenue churn is ($4,000 + $1,000 − $6,000) ÷ $100,000 × 100 = −1%. A negative result means the retained customer base expanded faster than revenue was lost.

Do not put new-customer revenue in expansion. Net revenue churn is designed to measure what happened inside the starting customer cohort.

Formula 4: annualized monthly churn

Multiplying monthly churn by 12 is a rough shortcut. Compounding gives a more accurate annualized result because each month leaves a smaller retained base.

Annualized churn = 1 − (1 − monthly churn)^12

At 3% monthly churn, the compounded annualized churn is 1 − (0.97)^12 = about 30.6%, not 36%. Use this annualized figure for planning only when monthly churn is reasonably stable.

For an actual annual report, calculate churn directly from the customers or recurring revenue present at the start of the year. Do not annualize a noisy month and present it as observed performance.

A worked SaaS churn calculation

Assume your subscription database shows 1,000 active customers and $150,000 MRR on September 1. By September 30, 35 starting customers have churned, taking $7,500 MRR with them. Downgrades remove another $1,500, while upgrades among starting customers add $4,000.

Customer churn: 35 ÷ 1,000 = 3.5%. Gross revenue churn: ($7,500 + $1,500) ÷ $150,000 = 6%. Net revenue churn: ($7,500 + $1,500 − $4,000) ÷ $150,000 = 3.33%.

These numbers tell a coherent story. Account loss is moderate, but churned accounts carry above-average revenue. Expansion softens the damage, yet it does not offset it. The next useful step is to segment churned MRR by plan, tenure, acquisition channel, and cancellation reason.

Five SaaS churn calculation mistakes

Using the ending customer count as the denominator

The denominator should be the customer population that could churn during the period: customers active at the start. Ending count mixes acquisition with retention.

Mixing account churn with user inactivity

A user who stops logging in is at risk, not necessarily churned. Keep product inactivity, cancellation, and revenue loss as separate events.

Counting reactivations inconsistently

Decide whether a reactivated customer is a new customer or a returned customer. Store reactivation as its own movement so your acquisition and retention reports remain explainable.

Letting failed payments linger forever

Define a grace period, such as the end of your dunning sequence. Mark involuntary churn only when recovery attempts finish, then backfill recoveries consistently if the account returns.

Averaging percentages across segments

Do not average plan-level churn rates unless every plan has the same starting base. Add the underlying churned and starting counts, then calculate the combined rate.

How to calculate churn from your database

Your source tables usually include customers or accounts, subscriptions, invoices, plans, and subscription events. Create a start-of-period snapshot, classify each recurring-revenue movement, and preserve the customer ID behind every aggregate.

A practical monthly output has these fields: starting customers, churned customers, starting MRR, churned MRR, downgrade MRR, expansion MRR, voluntary churn, involuntary churn, and the four calculated rates.

With AI for Database, you can ask for this output in plain English, save it as a self-refreshing dashboard, and trigger an email or Slack alert when churn crosses a threshold. For example: “Compare monthly customer churn and gross revenue churn by plan for the last 12 months, excluding test accounts.” You can then inspect the generated result without asking an engineer for a new SQL report every month.

Keep the underlying database read-only and review the first result against a small hand-checked sample. Plain-English querying removes the SQL bottleneck; it does not remove the need for clear metric definitions.

Monthly churn reporting checklist

Use the same snapshot time and timezone every month. Exclude internal, test, and fraudulent accounts. Reconcile starting MRR to the prior period. Separate cancellations, downgrades, expansion, and new business. Break out voluntary and involuntary churn. Compare customer churn with gross and net revenue churn. Investigate segments, not just the blended rate. Record definition changes next to the chart.

The headline number matters less than consistency. A stable 3% calculated the same way every month is actionable. A prettier 2.4% created by changing the denominator is theatre.

Questions SaaS teams ask about churn calculation

The quick answers below cover the choices that most often create conflicting churn reports.

Turn the formula into an operating signal

Calculate customer, gross revenue, and net revenue churn together. Then segment the movement by plan, tenure, account value, and churn type. That combination tells you how many customers left, how much money left, and whether expansion compensated for the loss.

If the calculation still lives in a monthly spreadsheet, connect the source database to AI for Database, validate the first report, and save it as a live dashboard. The goal is not another metric document. It is a consistent signal your team can act on before the next month closes.