Retention rate

Retention rate is the percentage of customers a subscription business keeps over a given time period, measured against the customers it started that period with and excluding any new customers acquired along the way. It answers a single question: of the people who were already paying you, how many are still paying you at the end.

Retention rate is one of the clearest signals of the health of a recurring revenue business. Because a subscription only earns while the customer stays, the share of customers who remain from one period to the next shapes how predictable revenue is, how much a customer is worth over their lifetime, and how hard new sales have to work just to replace the ones who left. It is usually measured over a consistent window, such as a month, a quarter, or a year, so the figure can be tracked as a trend rather than read once.

Why retention rate matters for subscription businesses

In a subscription model the sale is never really finished, it renews. That makes retention rate a leading indicator of whether the business compounds or leaks. A high retention rate means most of last period's revenue carries forward on its own, so growth stacks on top of a stable base instead of refilling a bucket that keeps draining. A low one means the team has to acquire aggressively just to stand still, and acquisition almost always costs more than keeping an existing customer.

Retention rate feeds directly into the metrics finance and growth teams live by. It is the inverse of Churn rate, it sits underneath Customer lifetime value because a customer who stays longer is worth more, and it shapes MRR and Net revenue retention by determining how much of the recurring base survives into the next period. When retention slips, the effect shows up everywhere downstream: softer lifetime value, weaker expansion, and forecasts that no longer hold. Watching retention rate over time also surfaces problems, such as onboarding gaps or failed payments recovered through Dunning, while they are still fixable.

How to calculate retention rate

Retention rate is calculated over a defined period by comparing the customers you finish with to the customers you started with, after removing anyone you newly acquired during that period so you are measuring only how well you held on to the customers you already had.

Stated in words: take the number of customers at the end of the period, subtract the new customers acquired during the period, divide that by the number of customers at the start of the period, and multiply by 100 to get a percentage.

As a formula:

CRR = ((E - N) / S) x 100

• E is the number of customers at the end of the period.

• N is the number of new customers acquired during the period.

• S is the number of customers at the start of the period.

Worked example

Suppose a business starts the period with 2,000 customers. During the period it acquires 150 new customers, and it finishes the period with 2,070 customers.

Plugging those into the formula, subtract the new customers from the end count to isolate the customers who were retained, which gives 1,920 Divide that by the starting count of 2,000, then multiply by 100. The result is a retention rate of 96%.

Reading the result: it tells you what share of the customers you began with were still customers at the end, independent of how many new ones you added. Keeping the period length consistent from one measurement to the next is what makes the number comparable over time.

Retention rate vs churn rate

Retention rate and churn rate describe the same movement of customers from opposite directions. Retention rate measures the share of customers you kept, while Churn rate measures the share you lost, over the same period. Because every customer either stays or leaves, the two are complements: a customer counted as retained is not counted as churned, and together they account for the starting base.

That relationship makes one easy to derive from the other for a given period. If you know the churn rate, the retention rate is 100 percent minus the churn rate, and vice versa. The practical difference is emphasis. Churn rate spotlights the leak and is the natural frame for a save or recovery effort, while retention rate spotlights the durable base and is the natural frame for lifetime value and forecasting. Teams often watch both so a single number cannot flatter or alarm on its own. One caution: the two only reconcile cleanly when they are measured the same way, over the same window, on the same customer or revenue basis.

Common mistakes with retention rate

• Counting new customers acquired during the period in the end number, which inflates retention rate and hides how many of the original customers actually stayed.

• Changing the length of the measurement window between periods, so a monthly figure gets compared to a quarterly one and the trend becomes meaningless.

• Mixing customer-count retention with revenue-based retention and treating them as the same thing, when a business can keep most of its customers while still losing revenue to downgrades, or the reverse.

• Ignoring involuntary loss from failed payments, so churn that could have been recovered through Dunning is quietly written into the retention number instead of being fixed.

• Reporting a single blended rate across very different segments, which averages away the fact that some cohorts or plans are churning far faster than others.

Benefits and examples

• Gives an early, honest read on recurring revenue health, since a stable retention rate means most of the base carries forward without new sales having to replace it.

• Anchors Customer lifetime value, because a higher retention rate means customers stay longer and each one is worth more over time.

• Makes forecasting more reliable, as a predictable retention rate lets finance project how much of today's base will still be paying in future periods.

• Surfaces product and onboarding problems early, since a segment whose retention rate drops is signaling friction before it shows up as lost revenue.

• Frames the payoff of recovery work, for example reducing involuntary churn through Dunning so more of the starting base is retained each period.

Using propensity modeling to improve retention rate

Retention rate tells you what already happened. Propensity modeling tries to get ahead of it by estimating, for each customer, how likely they are to renew or to churn before they actually do. Instead of treating the whole base the same, a propensity model scores individual customers on their risk of leaving, so retention effort can be aimed at the accounts most likely to lapse rather than spread evenly across everyone.

The model learns from signals the business already collects, such as declining product usage, support tickets, missed logins, payment failures, plan downgrades, and how far a customer got in onboarding. It weighs those signals against the outcomes of past customers to produce a churn-risk or renewal-likelihood score for each active subscriber, which teams can then sort and act on.

Used well, a propensity score turns retention from a reactive exercise into a proactive one. A customer scored as high-risk can be routed to a save offer, a check-in from customer success, or a targeted onboarding nudge while there is still time to change the outcome, and a customer flagged as a likely upgrade can be routed to expansion instead. Because the score isolates involuntary risk too, accounts heading toward a failed renewal can be pushed into Dunning and card-update flows before the payment lapses. The measurable payoff shows up in the retention rate itself: catching at-risk customers earlier keeps more of the starting base intact each period.

Frequently asked questions

What is a retention rate? It is the percentage of customers a business keeps over a defined period, measured against the customers it started with and excluding new customers acquired during that period, so it reflects only how well the business held on to the customers it already had.

How do you calculate retention rate? Take the customers at the end of the period, subtract the new customers acquired during the period, divide by the customers at the start of the period, and multiply by 100. In formula terms, CRR = ((E - N) / S) x 100.

What is the difference between retention rate and churn rate? They are two views of the same movement. Retention rate is the share of customers you kept, and Churn rate is the share you lost, over the same period. For a given period, retention rate equals 100 percent minus the churn rate.

What is a good retention rate? It depends heavily on the business model, price point, and segment, so a healthy figure for one type of subscription can be weak for another. The more useful comparison is against your own trend over time and against benchmarks.

Can you predict which customers are about to churn? To a degree, yes. Propensity modeling scores each customer's likelihood of churning from signals like falling usage, support activity, and payment health, so retention effort can reach the highest-risk accounts before they lapse rather than after.

Why is retention rate important for subscription businesses? Because a subscription only earns while the customer stays, retention rate determines how much revenue carries forward on its own, how much each customer is worth over their lifetime, and how hard acquisition has to work just to replace the customers who leave.