Finance for leaders
The Big Customer That Loses Money
Revenue growth can make a business worse. Four questions about margin per customer that any executive can ask without being an accountant.

A company's largest account can be the one that loses it the most money, and the income statement will not say so. It shows sales, costs and a profit at the bottom, perhaps even a healthy EBITDA, the operating profit before interest, taxes, depreciation and amortization. What it does not show is which customers leave that money behind and which ones eat it. Growing with the wrong customers makes the business worse while the total applauds.
What the average hides
Robert S. Kaplan illustrated this in When to Drop an Unprofitable Customer, published in Harvard Business Review in April 2012. It is a fictionalized case, based on a Harvard Business School case about the company Elkay, and its protagonists are characters: Egan, a British supplier of doors and staircases, and Westmid, its customer of 63 years. Egan allocated factory overhead as a percentage of direct labor and corporate expenses as a percentage of sales, so nobody could see what it cost to serve each customer. Timothy J. Jahnke, president and CEO of Elkay Manufacturing, comments in the same article that standard systems allocate costs arbitrarily and are almost certainly wrong. Total margin is an average, and the average hides the customer who loses.
Question oneWhat we are really charging
List price and the price that reaches the till are two different things. In Kaplan's case, Egan had discounts, rebates, promotions and returns bundled into a single line; they added up to 12 percent of sales and were assigned to nobody. Traced back to each customer, the real revenue per account changed shape. The question for a manager is simple: take our largest customer, subtract everything we grant outside the price, and see what is left.
Question twoWhat it costs us to serve them
With activity-based costing, Egan followed every order through its stages, from quoting and buying materials to production, delivery, invoicing and collection, and assigned each customer what it consumed. The result was hard to look at: 1 percent of accounts generated 100 percent of operating profit, the top 10 percent nearly twice that, and the remaining 90 percent were at break-even or subtracting. Westmid had an operating result of minus 28 percent on its sales, and its cumulative losses equaled 40 percent of Egan's profits.
The cause was habit: special orders and custom work priced far below their real cost, rush deliveries in half-empty trucks. The sales manager defended the account on the strength of the relationship and earned commission on volume, so the incentive pushed toward selling more regardless of what was left.
Question threeWhat we get that the number does not show
Jacquelyn S. Thomas, then associate professor at Southern Methodist University's Cox School of Business, warns that no customer should be dropped without weighing the strategic side: brand, market share, referrals. Jahnke confirms it from practice: Elkay sometimes stays with a customer that is not profitable today, to enter a region, for brand presence or because it expects an improvement, but knowing what it costs. Keeping a loss-making customer can be right; keeping one without knowing it is not.
Question fourWhat can change before cutting
Between keeping and cutting there is a lot of room, says Jahnke. He suggests talking frankly with the buyer, raising prices on some products and not stopping there. His company helped a customer switch to products made in a closer plant and shift its mix toward higher-margin ones, and that account stopped losing money. The options, from least to most drastic:
- Price. Charge for special as special: custom orders and rush jobs have a cost.
- Mix. Move purchases toward what carries margin, as Elkay did.
- Service. Thomas suggests finding out which benefits truly matter; perhaps on-time delivery is what counts and the discounts are an extra nobody asked for.
- Incentives. If commission is paid on volume, the team will defend the account even when it loses; paying on margin changes the conversation.
- Orderly exit. If nothing works, the terms change and, if the customer leaves, you help with the transition, because an unhappy customer talks.
An example
Imagine a mid-sized industrial distributor whose largest customer, a construction company, asks for rush deliveries several times a week and negotiates an additional discount every quarter. The general manager asks, for the first time, for that account's margin with everything included: discounts, rush freight, hours spent handling last-minute changes. The number comes out negative.
Her first reaction is to consider dropping them. The second, and better one, is to sit down with the buyer. She finds that the rush orders come from poor planning on the job site, and that the customer cares more about reliability than about the discount. She offers a fixed delivery schedule in exchange for removing the extra discount and charging separately for genuine emergencies.
Closing thought
Asking about margin per customer is a leadership task: knowing who makes money with us, who loses it and which rule allows it. Three questions for the next meeting:
- Do we know what our largest account leaves behind after discounts, rush jobs and returns?
- Who in the company has an incentive to defend it even when it loses?
- What would we change in price, mix or service before thinking about cutting it?
