Churn Rate.

Churn Rate, properly understood

Churn rate, calculated as Churn rate = (Customers lost ÷ Customers at start) × 100, measures the share of your customer base that cancels or lapses over a given period — usually calculated monthly, though annual churn is also common for longer-cycle businesses. There are two versions worth distinguishing: customer (logo) churn, which counts accounts lost regardless of size, and revenue churn, which weights the loss by the MRR or ARR those accounts represented — a business can have flat logo churn but rising revenue churn if it's losing its biggest accounts, which is a much worse signal hiding inside a stable-looking headline number. The data comes from the same billing or subscription system used to calculate MRR, filtered to cancellations (not downgrades, which is a separate 'contraction' category) within the period window.

In GCC subscription and app businesses, churn needs to be read carefully around Ramadan and the summer travel season, both of which see temporary usage and payment dips for many categories (fitness, food delivery, daily-use apps) that can look like churn but are actually pauses — some products offer a formal 'pause' rather than cancel option specifically to avoid over-counting seasonal dips as true churn. Payment-method churn is also a distinct, regionally relevant category: a meaningful share of 'voluntary' churn in card-based subscription markets is actually involuntary — expired cards, failed payment retries, or bank fraud blocks on recurring international charges — which needs to be tracked separately from customers who deliberately cancel, since the fix (better payment retry logic, local payment methods) is completely different from a product or pricing fix. Long-cycle B2B relationships common in the region also mean churn events are rarer but far higher-stakes per account, so B2B churn analysis benefits from a qualitative account-level review (why did each account actually leave) rather than pure statistical tracking.

The most common misread is watching only logo churn while revenue churn quietly worsens, or vice versa — both need to be tracked together to see the full picture. Teams also frequently calculate churn on the wrong denominator, mixing trial users or non-paying accounts into the 'customers at start' figure, which dilutes the real churn signal from paying customers. A third pitfall is treating churn as purely a retention-team problem when a large share of it is often set at acquisition — customers acquired through heavy discounting or a mismatched sales pitch churn at much higher rates than those acquired through a good product-market fit, so rising churn is sometimes better diagnosed by looking upstream at who's being acquired, not just what's happening after signup.

Churn rate should always be read next to ARPA and ARR, since revenue churn directly determines how much of your recurring revenue base needs to be replaced just to stay flat, and next to Activation Rate and App Retention, since poor early activation is one of the most reliable predictors of later churn. It's also worth pairing with Contribution Margin — high churn in a low-margin segment is a much bigger problem than the same churn rate in a high-margin one, since there's less room to have already recovered the acquisition cost before the customer leaves.