Customer turnover rate

Customer turnover rate measures how quickly customers end their relationship with an organisation. Because replacing a suitable existing customer generally costs more than retaining one, turnover—also called churn—makes customer loss visible and provides a starting point for diagnosis.

When to use it

• Answer the key performance question: “How well are we retaining customers?”
• Assess this KPI within the Customer perspective.
• Plan data collection, formula use, reporting frequency, and data-source requirements for this KPI.
• Compare results against the targets, benchmarks, examples, or trend guidance available for this KPI.

Origins

Customer turnover began as a practical retention measure in contractual and subscription businesses, then became a major analytical discipline as telecommunications, financial services and digital subscriptions expanded during the late twentieth century. Relationship marketing, explicitly named by Leonard Berry in nineteen eighty-three, shifted attention from individual transactions to relationships sustained over time. Churn is the loss side of that relationship, so its period and eligible population must be defined consistently.

What it is

Perspective: Customer perspective.

Key performance question: How well are we retaining customers?

Customer turnover, customer churn, defection and attrition all describe the loss of clients or customers during a stated period.

The KPI matters particularly in sectors such as banking and telecommunications, where a slightly better competing offer can trigger switching. Identifying vulnerable customers, addressing the causes of defection and designing appropriate win-back activity can materially improve relationship economics. Leaders should monitor both the rate and the operational actions intended to reduce avoidable loss.

How to use it

Measurement

Data collection method

In contractual businesses, classify customers whose contracts renew and those that do not at the end of the agreed term. CRM and business-intelligence systems can perform the same identity and activity tracking across large non-contractual customer bases.

Surveys and focus groups can reveal satisfaction and reasons for intended departure. Behavioural models can mine billing disputes, service failures, support contacts, usage changes and policy complaints associated with attrition. These signals estimate risk rather than prove a customer will leave.

Predictive churn models are most useful when they produce a prioritised, manageable group for retention activity. Test whether contacting that group creates incremental retention after campaign cost, rather than judging a model only by statistical accuracy.

Formula

A simple turnover rate divides the customers who ended the relationship during the period by the relevant customer population at the end of that period.

State the denominator explicitly. An average or opening customer base may be preferable when rapid acquisition makes the ending population misleading.

Frequency

Review monthly or quarterly according to the pace of the market and normal contract or purchase cycle. Fast-moving telecommunications services generally require more frequent monitoring than slower categories.

Source of the data

Contract non-renewals are usually available in sales records and make individual defectors identifiable. Where customers cannot be observed directly, use surveys, transaction patterns or other behavioural evidence and label the result as an estimate.

Cost/effort in collecting the data

Effort depends on customer volume, identity quality and the form of the relationship. A small contract base is easy to measure; a one-to-many retail environment requires stronger identity resolution. Modern CRM tools keep basic measurement costs relatively low, but reliable causal diagnosis still takes analytical work.

Target setting/benchmarks

Industry bodies, specialist consultancies and CRM vendors may provide comparisons. Use them only when customer definitions, observation periods and voluntary-versus-involuntary treatment match the internal metric.

Example

A simple calculation begins as follows:

On 1 November, the organisation has 100 customers.

During the month, it gains 20 customers.

During the month, it loses 5 customers.

The operational value comes from explaining why the losses occurred and testing whether an intervention prevents them.

Verizon Wireless illustrates the potential scale. From 2002 to 2004, its customer base grew from 29.4 million handsets to 43.8 million and it reported $13.7 billion in operating earnings. Over the same period, monthly churn fell from 2.6% to 1.3%, increasing customer equity by $13.9 billion. Only forty per cent of the increase in customer equity came from acquired customers; 60% came from increasing the value of the customer base through stronger retention. From 2002 to 2004, the resulting value creation was about two-thirds of the company’s value at the beginning of 2002.

Top practical tip

Create a clear save and win-back process for customers shown to be vulnerable, but connect every offer to the reason the relationship is at risk. A discount cannot repair recurring service failure, and a retention team cannot compensate indefinitely for a broken product or policy.

Top pitfall

Separate voluntary churn, where a customer chooses to leave, from involuntary churn caused by events such as relocation or death, and from forced churn initiated by the company for reasons such as non-payment. The organisation can influence these categories in very different ways.

Also define when an inactive customer is truly lost. In non-contractual markets, a long pause may reflect an unusually slow purchase cycle or temporary financial circumstances rather than defection.

Further reading

Q&A with Don Peppers and Martha Rogers: Measuring Customer Value Can Be More Important than Measuring Revenue: http://mthink.com/content/qa-don-peppers-and-martha-rogers-measuring-customer-value-can-be-more-important-measuring