Customer churn is the loss of customers who stop purchasing from, subscribing to, or otherwise maintaining a commercial relationship with a business during a defined period. It is commonly expressed as a rate or percentage and is particularly important for subscription, membership, telecommunications, software-as-a-service, financial-services, insurance, media and other recurring-revenue businesses where customers are expected to remain active over time.
A customer can be considered churned when they formally cancel a subscription, allow a contract to expire without renewal, close an account, stop purchasing for a period defined by the business, or become inactive according to an agreed behavioural criterion. The precise definition therefore depends on the business model and customer relationship: a monthly software subscriber who cancels is usually clearly churned, whereas a supermarket customer who has not bought anything for six weeks may not be churned because purchasing is naturally intermittent. Churn is therefore not simply the same thing as a customer failing to buy once; it represents the ending, or operationally defined disappearance, of an expected customer relationship.
Churn is normally discussed in terms of customer churn, revenue churn and sometimes logo churn. Customer churn measures the proportion of customers lost, whereas revenue churn measures the recurring revenue associated with customers who have been lost or have reduced their spending. Logo churn is a common business-to-business term in which each customer organisation is treated as one account or logo, regardless of its revenue contribution. These measures can tell substantially different stories.
A business might lose many small customers and therefore have a high customer churn rate while losing relatively little revenue, or it might lose only a few major accounts and experience a low customer churn rate but a severe revenue impact. Gross revenue churn generally focuses on lost recurring revenue before considering expansion from existing customers, while net revenue retention incorporates contraction, churn and expansion and can therefore show whether the existing customer base is becoming more or less valuable overall. Keeping these measures conceptually separate prevents a business from treating all customer losses as economically identical.
The causes of churn vary considerably and can arise from customer dissatisfaction, poor service, inadequate product functionality, price increases, complicated customer experiences, stronger competitors, changing customer circumstances, business failure, relocation, mergers, budget reductions, changes in technology, regulatory developments or the customer simply no longer needing the product. Some churn is preventable and some is effectively unavoidable.
Voluntary churn occurs when a customer actively decides to leave, while involuntary churn can occur because of failed payments, expired cards, account closure for administrative reasons or other circumstances that result in termination without an explicit customer decision. Businesses often distinguish controllable churn from uncontrollable churn because the appropriate response differs. A product defect may require engineering work, while a customer leaving because their business has permanently closed cannot necessarily be prevented through better service. Understanding causation is therefore more valuable than merely observing the aggregate churn percentage.
Customer churn has a direct relationship with customer lifetime value, recurring revenue, forecasting and the economics of customer acquisition. When customers leave earlier than expected, the business receives less revenue over the relationship's lifetime and may have to acquire replacement customers merely to maintain its existing customer base. If acquisition costs are significant, high churn can make an apparently successful growth strategy economically weak because new customers continuously replace departing customers rather than creating sustained expansion.
Churn also affects recurring-revenue predictability, sales capacity, support requirements and marketing efficiency. A business with a large installed customer base and low churn can often build growth on a relatively stable foundation, whereas a business with high churn may need disproportionately large acquisition volumes to maintain the same customer count. For this reason, churn is frequently analysed alongside acquisition cost, average revenue per customer, retention rate, customer lifetime value and cohort performance.
Churn should be analysed over time rather than interpreted as a single isolated percentage. Cohort analysis groups customers according to a meaningful starting point, such as acquisition month, contract start date, product purchased or customer segment, and then observes how many remain active over subsequent periods.
This can reveal whether newer customers churn faster than older customers, whether a product change affected retention, whether a particular acquisition channel produces customers with weaker retention, or whether churn is concentrated in a specific market segment. Survival analysis and retention curves can provide a more detailed view by showing the probability that a customer remains active after different lengths of time. Businesses can also examine churn by geography, industry, plan, contract length, acquisition source, account size, product usage and customer-support history. Such segmentation is important because an overall churn figure can conceal materially different retention patterns among customer groups.
The timing of churn is also significant. Early-life churn can indicate a poor onboarding experience, misleading acquisition messaging, inadequate product-market fit or a failure to demonstrate value quickly. Churn after a substantial period of use may instead be associated with changing needs, competitive displacement, price sensitivity or declining perceived value. A customer who cancels immediately after a price increase represents a different retention problem from one who leaves after years of declining product usage.
Businesses can therefore use behavioural signals such as declining login frequency, reduced transaction volume, fewer purchases, unresolved support cases, lower feature adoption, declining engagement or failed payments as potential indicators of elevated churn risk. Such indicators do not prove that a customer will leave, because correlation is not causation, but they can help businesses investigate retention problems and intervene appropriately.
Customer churn is often confused with customer attrition, customer inactivity and customer dissatisfaction, although these concepts are related rather than identical. Attrition is a broader term for customers or accounts gradually being lost and may include churn depending on the context. Inactivity describes a lack of recent behaviour but does not necessarily mean the commercial relationship has ended. Dissatisfaction is an attitude or experience, whereas churn is a behavioural or contractual outcome.
A dissatisfied customer may remain because switching is difficult, while a satisfied customer might leave because their circumstances have changed. Conversely, a customer can appear inactive under a simplistic definition while still intending to return. The definition of churn must therefore reflect normal purchasing frequency, contractual terms and the expected customer lifecycle. Businesses should document the precise event or inactivity threshold that constitutes churn so that reported rates remain consistent and comparable.
Churn is commercially significant because retaining an existing customer can affect revenue quality, operational efficiency, customer lifetime value and the predictability of future performance. Reducing churn is not necessarily achieved by attempting to retain every customer at any cost; retention efforts have costs, and a business should consider customer profitability, strategic value, service requirements and the underlying reason for departure. Excessive discounts or incentives can sometimes retain customers temporarily without resolving the fundamental problem and may reduce margins.
Effective churn management therefore combines accurate measurement with diagnosis, customer experience improvements, product development, appropriate pricing, proactive support, effective onboarding and systematic analysis of why customers leave. The most useful interpretation of churn is consequently not simply that a certain percentage of customers disappeared, but that a measurable portion of the customer base ceased the expected commercial relationship and that the pattern, timing, causes and economic consequences of those departures provide information about the health and sustainability of the business.
