Keep the opening group fixed
A recurring-services review can become celebratory when net retention exceeds one hundred per cent. The measure suggests that the opening customer group is worth more after losses, reductions and expansion have been considered. That is useful information. It does not say that every relationship strengthened, that the service experience improved, or that the additional contract value has already become cash. I have seen a strong aggregate number shorten precisely the discussion management needed to have about which customers were leaving and why the remaining ones were buying more.
We examined a fixed opening cohort with EUR12 million in annualised contract value. At the comparison date, EUR10.86 million remained before expansion, giving gross retention of 90.5 per cent. Expansion within that same cohort added EUR2.4 million, bringing the closing annualised value to EUR13.26 million and net retention to 110.5 per cent. New customers were excluded from both calculations. The two percentages were arithmetically consistent. Their difference was the point of the exercise: the success of expansion did not remove the loss and contraction that had occurred within the original group.
The value was annualised for a defined management comparison. It was not a claim about the revenue recognised during the year or the cash collected from those contracts. Service start dates, billing schedules and the period covered by each agreement could differ. Finance reconciled the underlying contracts and documented the measurement boundary. That work was necessary before drawing operating conclusions, because a changed cohort or inconsistent treatment of expansion could create an apparent improvement without any corresponding change in the customer relationships. A familiar label is not a substitute for a stable definition.
The commercial team initially preferred to lead with the net figure alone. Its members had worked hard to expand several important accounts and reasonably wanted that achievement recognised. Service leaders were concerned that the combined measure would make lost work appear less urgent. I did not ask either group to abandon its perspective. We presented gross and net retention together, using the same opening population and comparison dates. That allowed the expansion achievement to remain visible while giving lost and reduced contracts their own explanation and management response.
Expansion and loss ask different questions
A customer can buy more because the service has become more useful, because its own business has grown, or because the contract includes a broader scope. Those explanations have different implications for future demand and delivery cost. An increase in annualised value is not automatically an increase in contribution. We therefore asked account leaders to explain what the expansion required from the business: additional people, infrastructure, service coverage or implementation effort. The financial review connected the larger contract to the work accepted in return, rather than treating expansion as a benefit without an operating consequence.
The loss and contraction discussion needed comparable care. A smaller contract might reflect a dissatisfied customer, a deliberate withdrawal from an unprofitable scope, a customer's own retrenchment or an agreed transfer of work. We did not assume that every reduction was avoidable or that every retained euro was desirable. Service and account teams examined the circumstances and the decisions available. Finance helped preserve the comparison and the economic effect. The exercise was not designed to produce one league table of blame; it was intended to identify where management should improve service, renegotiate or accept a smaller relationship.
Concentration also mattered. Several large expansions could offset many modest contractions and leave the company more dependent on fewer relationships. The 110.5 per cent net result did not answer that question. We reviewed the distribution of the movements, while keeping commercially sensitive customer details within the appropriate internal discussion. A board can ask whether growth rests on broad relationship strength or a small number of substantial changes without needing a public list of customers. The narrative should make that uncertainty available, even when the aggregate figure is accurate and favourable.
There is a temptation to turn every reason into a precise category and then treat the resulting chart as objective. In practice, a customer decision may have several causes, and our account of it may be incomplete. We recorded the evidence supporting each explanation and distinguished confirmed customer feedback from the account team's interpretation. That made the review less tidy but more credible. If the reason for a loss was uncertain, an uncertain label was preferable to a confident classification chosen because it fitted the management narrative or protected a departmental result.
Connect the measure to a decision
The revised review produced service and account actions, but the first measurable improvement was in the quality of the discussion, not an immediate change in retention. We did not claim that showing two percentages caused customers to stay. Managers still had to act on the underlying issues and observe later cohorts. Some interventions would take time; some lost work could not be recovered. Retaining that distinction prevented a reporting improvement from being presented as an operating result before the company had evidence that relationships had changed.
I would establish the definitions before the planning cycle begins. If the denominator is debated only after a disappointing result, even a reasonable correction can look like an attempt to change the score. The opening cohort, treatment of partial reductions, comparison period and expansion boundary should be understood by the people who will use the measure. Changes may still be necessary, but they should be explained and, where practical, accompanied by a comparable prior view. Consistency helps the business learn from movement rather than repeatedly argue about which version to believe.
For investment decisions, I want retention evidence beside contribution and capacity. A growing managed-service relationship may require additional coverage before the larger fee is earned. A smaller relationship may release little cost if the supporting team or infrastructure cannot be reduced proportionately. These are operating questions that the retention ratio cannot settle. Finance can help connect them by showing the contract movement, expected delivery requirements and relevant cash timing. The aim is not to burden every commercial meeting with the entire financial model, but to bring the necessary consequences into the decision at hand.
A board should be able to congratulate the team for expansion and still ask why some existing work disappeared. In our example, EUR2.4 million of additional annualised value within the cohort deserved recognition. So did the EUR1.14 million difference between the opening value and the amount retained before expansion deserve investigation. Neither fact invalidated the other. Keeping them together produced a more mature account of recurring growth: one that valued commercial progress while remaining curious about its distribution, cost and durability. That is the standard I want a finance function to support.