Three retention questions that one percentage cannot answer

Customer retention, gross revenue retention and net revenue retention reveal different parts of the relationship. Read them together before deciding whether growth means the customer base is healthy.

A business reports revenue retention above 100%, and the board concludes that customers are staying. That conclusion may be wrong. Remaining customers can spend more while a substantial number of other customers leave.

Retention is a family of questions, not one universal percentage. The CMO needs to distinguish how many relationships remain, how much existing revenue is lost and how expansion changes the total.

Start by defining the opening group and the period. Otherwise, teams can use the same label while calculating different things.

Keep that definition beside the chart and preserve it between reporting cycles. A change in methodology should be visible, not mistaken for a sudden improvement in customer behaviour.

Count retained relationships

For a simple customer-retention measure, identify customers active at the beginning and count how many remain at the end. Divide the retained number by the starting number.

In a hypothetical annual cohort, 100 customers begin the year and 80 of those customers remain at the end. Customer retention is 80%. New customers acquired during the year do not belong in that retained count.

Specify how the business treats pauses, reactivations, merged accounts and customers with several subscriptions. Those conventions can materially affect comparisons, especially when different systems use different definitions of an active customer.

The count describes relationships, not their economic weight. Losing one large account and losing one small account each reduce the retained customer count by one, even though the revenue consequences may differ greatly.

See the revenue that disappeared

ChartMogul’s gross-retention documentation describes recurring revenue retained after churn and contraction, without adding expansion. This view makes losses visible before growth in other accounts offsets them.

Consider a separate hypothetical cohort starting with €100,000 in monthly recurring revenue. During the measurement period, cancellations remove €10,000 and downgrades remove €5,000. Gross revenue retention is 85%: €85,000 divided by €100,000.

Keep the units consistent. This calculation compares recurring-revenue levels, not cash collected during the period or total recognised annual revenue. A business with project-based sales needs a definition appropriate to its own buying pattern.

Gross retention helps expose erosion in the starting base. It does not identify why the losses occurred or whether retaining every customer would have been economically sensible.

Add expansion without hiding the losses

Suppose the same hypothetical cohort also produces €20,000 of expansion in monthly recurring revenue. With no reactivations, net revenue retention is 105%: the €85,000 remaining after losses plus €20,000 of expansion, divided by €100,000.

ChartMogul’s net-retention documentation includes expansion and excludes new business. It also shows why reactivation settings and other report conventions should be stated rather than assumed.

The 105% result is compatible with meaningful churn and contraction. Expansion has more than offset those losses in revenue terms; it has not made them disappear.

Use a revenue bridge that shows each movement. Starting revenue, churn, contraction and expansion tell a clearer story than a single number labelled “strong retention.”

Look inside the average

Segment by relevant differences such as acquisition period, customer size, product or use case. A healthy aggregate can contain a weak group whose losses are temporarily covered by a growing group elsewhere.

Keep segment definitions stable enough to compare. Moving a customer between categories can change segment results without changing the underlying relationship. Document how those movements are treated.

Avoid drawing firm conclusions from tiny groups. A single large cancellation can dominate a segment’s rate. Show the customer count and revenue exposure alongside the percentage.

Compare periods that fit the renewal pattern. A monthly movement may be informative for a monthly subscription but misleading for a portfolio whose largest contracts renew together once a year.

Connect the numbers to customer experience

Retention data tells the team where to investigate. Customer interviews, service records and product-use evidence help explain the mechanism behind the movement.

A downgrade might reflect reduced need, a product problem, budget pressure or a customer moving to a more appropriate package. Treating every downgrade as the same marketing failure produces poorly targeted responses.

Expansion also needs interpretation. Higher prices, additional users and broader product adoption can all increase revenue, but they imply different customer experiences and different future risks.

Keep economics beside retention

Revenue retained at a deep discount or with unusually expensive support may contribute less than the headline suggests. Bring margin, concessions and servicing cost into decisions about retention programmes.

Do not turn a strong retention rate into an automatic instruction to stop acquisition. The balance depends on market opportunity, customer concentration, growth goals and the economics of both activities.

Present the three questions together: which customers stayed, what revenue was lost before expansion, and how the remaining relationships grew. Then explain the most important movements with evidence and a specific next decision.

The aim is a clearer account of the business, not a competition to select whichever retention definition produces the most flattering number.