Expert analysis · reviewed 6 Oct 2026
Customer value: measuring what a customer really brings in, from acquisition to reactivation.
Key distinctions
Three conditions before calculating.
Margin, not revenue
Every customer value measure is calculated after variable costs, discounts, returns and cost to serve. Revenue ranks customers; it does not say which ones create value.
The cohort, not the average
Customers acquired at different times, through different channels or with different offers do not behave in the same way. A global average hides the gaps that drive the decision.
The incremental effect, not the observed behaviour
The best customers often buy, stay and come back without any action. Only a comparison with a control group or a reference period isolates what an action changes.
Method
Working through customer value in four stages
- 01Measure the entry cost
Calculate a full and incremental acquisition cost by channel, with its payback period on margin.
- 02Project the value
Estimate customer lifetime value on margin, by cohort, with a declared discounting convention.
- 03Correct for real profitability
Deduct the cost to serve customer by customer, then prioritise with an RFM segmentation calculated on margin.
- 04Test every action
Measure retention, cross-selling and reactivation against a control group, and fund only the actions whose incremental margin is positive.
The customer value decision map
Seven decisions in the customer lifecycle, the question each one answers and the trap that most often distorts it.
| Lifecycle stage | Question answered | Most common trap |
|---|---|---|
| Acquire | What does a won customer really cost? | An average CAC that counts customers who had already decided |
| Value | How much will this customer bring in? | A lifetime value calculated on revenue, without discounting |
| Serve | Which customers are really profitable? | An average product margin that ignores the cost to serve |
| Prioritise | Which customers should you work on first? | Mistaking the best RFM score for the best target |
| Retain | Does the programme change behaviour? | Attributing the gap between members and non-members to the programme |
| Grow | Which complementary offer adds margin? | Counting purchases that would have happened without the offer |
| Reactivate | Which inactive customers should you contact? | Measuring on the raw return rate, spontaneous returns included |
The control group is the tool common to all these measurements.
The same customer, from acquisition cost to the next decision
Illustrative example: a customer acquired for an incremental acquisition cost of €150 brings in €120 of margin a year, with retention of 80% and a discount rate of 10%.
Discounted CLV = 120 × 0.80 ÷ (1 + 0.10 − 0.80) = €320. CLV/CAC ratio = 320 ÷ 150 = 2.1. Payback period = 150 ÷ (120 ÷ 12) = 15 months.
The customer is profitable, but ties up fifteen months of margin before becoming so. Any retention, cross-selling or reactivation action is then judged by what it adds to those €320, measured against a control group.
Acceptance conditions
What must be true to act.
- 01
Calculate every customer value measure on contribution margin.
- 02
Compare channels and actions on their incremental effect, never on their observed average.
- 03
Set acquisition cost against the lifetime value and payback period of the same cohort.
- 04
Hold out a control group for every retention, cross-selling or reactivation action.
Limits
What this analysis does not prove.
- Customer value measures describe averages by cohort, not individual certainties.
- Referral and network effects largely escape the measures presented.
- Parameters estimated over one period change with prices, channels and competition.
- A control group measures a local effect, which does not transfer to another population without checking.
References