Retention fundamentals
How to Calculate Customer Churn Rate
Customer churn rate measures what share of the customers present at the start of a period were lost during it. Keeping the starting cohort fixed prevents new sales from hiding a retention problem behind a stable ending customer count.
What customer churn rate measures
Customer churn rate is the percentage of the starting customer base that leaves during a selected period. If 18 of 300 customers active on the first day are lost by the end, monthly customer churn is 6%.
A lost customer must be defined for the business model. It could mean a canceled paid subscription, a terminated contract, or an account that passes an inactivity threshold. The definition should be observable and applied consistently.
Customer churn is a logo or account measure; it weights every customer equally. Revenue churn answers a different question because losing one large customer can matter more financially than losing several small ones. Track both when account sizes vary.
The formula and each variable
Customer churn rate = Customers lost during period ÷ Customers at start of period × 100
- Customers lost
- Members of the starting customer base who meet the loss definition during the period.
- Customers at start
- Active customers at the exact beginning of the selected period.
- Period
- A stated month, quarter, year, or other consistent interval.
When the starting customer count is zero, churn is undefined for that period rather than 0%.
How to calculate it step by step
Choose the period
Select an interval that fits the buying or renewal cycle. Label the rate as monthly, quarterly, or annual.
Freeze the starting cohort
Record customers active at the first instant of the period. This is the denominator and should not grow as new customers arrive.
Apply a clear loss rule
Count starting customers who canceled, expired, or crossed the chosen inactivity threshold during the period.
Exclude new customers
Customers acquired after the period began are not at risk for the full period and do not belong in this simple cohort calculation.
Divide and investigate
Calculate the rate, then segment by plan, tenure, acquisition source, or reason where the sample supports it.
Worked example
Worked example: a scheduling software company
The company begins April with 500 paying customer accounts. During April, 35 of those starting accounts cancel. Sales adds 50 new accounts, so the company ends with 515.
- Customers on April 1
- 500
- Starting customers lost
- 35
- New customers added
- 50
- Customers on April 30
- 515
- Customer churn = 35 ÷ 500 × 100 = 7%.
- Ending count check = 500 starting − 35 lost + 50 new = 515 ending.
- Simple net subtraction, (500 − 515) ÷ 500, would produce −3% and falsely suggest no churn.
The business grew its total count by 3%, yet still lost 7% of the customers it had at the beginning. Acquisition more than replaced those losses; it did not erase them.
Why the starting cohort is the denominator
Only customers present at the start were exposed to the entire period. Adding new customers to the denominator dilutes the rate and makes fast-growing businesses appear to retain better simply because they acquired more.
For detailed analysis, track acquisition cohorts over equal ages—for example, the share still active after one, three, and six months. That separates lifecycle behavior from calendar growth.
Customer churn and revenue churn are not interchangeable
Customer churn counts accounts. Revenue churn measures recurring revenue lost from the starting base, sometimes adjusted for expansion. If the company loses five large accounts, customer churn may look modest while revenue impact is severe.
Use the metric that matches the decision, and label it. Do not put a revenue number into the customer-count formula or compare a monthly customer rate with an annual revenue rate.
Choose a period that can be interpreted
A monthly rate can be noisy for annual contracts because few renewals are due each month. A yearly rate may hide when a monthly subscription product developed a problem. Align measurement with renewal opportunity and operating cadence.
Do not multiply a monthly churn percentage by 12 and call it annual churn. Retention compounds because each month’s losses apply to a smaller remaining base. Cohort data is usually the clearest path.
The rate identifies a symptom, not the cause
Churn can rise because of product fit, service quality, customer failure, price changes, seasonality, billing failures, or acquisition of poorly matched customers. Cancellation reasons and behavior before loss help interpret the rate.
Retention work should also consider save offers and involuntary churn. A discount that retains an account may reduce gross profit; recovered payment failures may improve retention without changing the product.
What moves customer churn
Can reduce churn
- Clear onboarding that gets customers to useful outcomes sooner.
- Reliable product or service delivery and responsive support.
- Better customer fit at acquisition rather than maximum signup volume.
- Payment recovery and timely renewal communication.
Can increase churn
- Poor fit, weak adoption, service failures, or missing value.
- Price or contract changes without matching perceived value.
- Acquisition campaigns that attract short-lived customers.
- Billing failures counted together with voluntary cancellation.
Common mistakes
Where the calculation goes wrong
Using net customer change
Ending minus starting customers mixes acquisitions with losses and can hide churn completely.
Including new customers in the denominator
They were not present for the full period. Keep the starting cohort fixed.
Leaving “lost” undefined
Cancellation, expiration, and inactivity can produce different counts. Publish the operational rule.
Comparing unlike periods
Monthly and annual churn are not directly comparable, and simple multiplication ignores compounding.
Action checklist
Before you use the result
- Name the exact period and whether it matches renewal behavior.
- Record the active customer count at the start.
- Define cancellation, expiration, and inactivity rules.
- Count only lost customers from the starting cohort.
- Keep new acquisitions outside the simple churn formula.
- Track revenue impact when account values differ.
- Segment and investigate causes without overreading small samples.
FAQ
Questions beyond the basic calculation
Can customer churn exceed 100%?
Not when lost customers are correctly limited to the starting cohort and each is counted once. A result above 100% indicates inconsistent counts or definitions.
Should reactivated customers reduce churn?
Choose and document a policy. Many teams report gross losses and reactivations separately so recovery does not hide the original churn event.
What is involuntary churn?
It is customer loss caused by events such as failed payments rather than an intentional cancellation. Separating it can reveal billing-recovery opportunities.
Is a lower churn rate always better?
Usually retention is valuable, but keeping unprofitable or poor-fit customers at any cost may not be. Read churn with gross profit, support burden, and customer outcomes.
Note: This guide is for general educational use. Churn definitions and cohort methods should be adapted to the contracts, billing cycle, and data quality of your business.