Expert analysis · reviewed 6 Oct 2026

Customer profitability: measuring the cost to serve before rewarding large customers.

SUPPORTED DECISIONFor each loss-making customer or segment, choose between serving differently, renegotiating the terms or reducing effort, after checking that the costs assigned are genuinely avoidable.

Key distinctions

Three conditions before calculating.

01

Cost to serve, not average margin

Product margin assumes that every customer costs the same to serve. Deliveries, returns, support and payment terms create gaps that only a per-customer view reveals.

02

Avoidable costs, not allocated costs

Spreading fixed overheads in proportion to revenue manufactures loss-making customers on paper. Only the costs that would disappear with the customer or the activity count for the decision.

03

Serve differently before serving less

A loss-making customer can become less so through a service offer, a minimum order or a cheaper channel. Dropping the customer remains the last option, and it has its own effects.

Method

Measuring customer profitability in four steps

  1. 01Link costs to activities

    Identify the activities each customer consumes — orders, deliveries, returns, calls, visits and credit; an activity without a measurable driver stays out of the calculation.

  2. 02Value each driver

    Assign a unit cost to each activity, for example an urgent delivery or an hour of support, and document the source of that cost.

  3. 03Build the cumulative curve

    Rank customers from most to least profitable, accumulate their contribution and locate the peak of the curve as well as the customers who pull it down.

  4. 04Choose the action customer by customer

    For each loss-making customer, compare serving differently, renegotiating, reducing effort or accepting the loss for a documented reason.

ORIGINAL ASSET

The customer profitability whale curve

Four customer groups ranked from most to least profitable. Cumulative contribution rises above total profit, then falls back under the effect of loss-making customers.

Customer groupShare of customersGross marginCost to serveContributionCumulative contribution
Most profitable20%€900k€300k€600k150%
Profitable30%€500k€380k€120k180%
Break-even30%€400k€420k−€20k175%
Loss-making20%€300k€600k−€300k100%
20% of customers deliver 150% of profit; the least profitable 20% destroy 75% of it.

Cumulative contribution is expressed as a share of total profit. Before deciding, check which costs to serve would actually disappear.

ILLUSTRATIVE EXAMPLE · SIMULATED DATA

A customer at 30% gross margin who leaves only 12.5%

01 · SITUATION

Illustrative example: a distributor generates €200,000 in revenue at a 30% gross margin, with 48 urgent deliveries, €8,000 of year-end rebates, €5,000 of returns, 120 hours of support and 90-day payment terms.

02 · CALCULATION

Gross margin = 200,000 × 30% = €60,000. Costs to serve = 48 × €250 + €8,000 + €5,000 + 120 × €50 + 200,000 × 90 / 365 × 8% = €34,945. Contribution = 60,000 − 34,945 = €25,055, i.e. 12.5% of revenue.

03 · DECISION

The customer remains profitable, but two and a half times less than its gross margin suggested. Replacing 36 urgent deliveries with planned deliveries would save €9,000 and raise its contribution to €34,055, without touching the price.

Acceptance conditions

What must be true to act.

  1. 01

    Calculate each customer's contribution after costs to serve, not only after the cost of goods.

  2. 02

    Assign only costs whose driver is measured per customer or per order.

  3. 03

    Check that a cost is avoidable before declaring a customer loss-making.

  4. 04

    Test a new service rule on one segment before rolling it out.

Limits

What this analysis does not prove.

  • Unit costs per activity are estimates that must be reviewed regularly.
  • Past profitability says nothing about the future value of a growing customer.
  • A customer's indirect effects, such as referrals or the volume negotiated with suppliers, escape the calculation.
  • Dropping a customer can leave fixed costs unchanged and shift the loss onto other customers.

References

Works cited.

  1. Niraj, Gupta & Narasimhan (2001), Customer Profitability in a Supply Chain (opens in a new tab)
  2. Reinartz & Kumar (2000), On the Profitability of Long-Life Customers in a Noncontractual Setting (opens in a new tab)
  3. Kaplan & Anderson (2004), Time-Driven Activity-Based Costing (opens in a new tab)