The 47-year gap

Nothing changed

In March 2026, Distribution Strategy Group published a report based on a survey of 84 distributors. The title: "Closing the Execution Gap." The tagline could have been written in 1979: "Most distributors measure profitability. Far fewer act on what they learn."

Between those two dates, the technology to measure customer profitability dropped from months of consulting work to something you can do with an ERP export and a laptop. The cost fell by orders of magnitude.

The willingness to act on the results didn't change at all.

And I think I understand why, because the most famous customer profitability case study in business history tells the same story.

The Kanthal story

Kanthal is a Swedish manufacturer of electric resistance heating elements, headquartered in Hallstahammar. In the late 1980s their CEO, Carl-Erik Ridderstråle, saw that profitability was declining even though production costs were stable. The indirect costs (sales, administration, logistics) kept growing and nobody could explain why.

He hired a consulting firm called SAM to analyze the business at the individual customer level. What they found was ugly.

Only 40% of customers were profitable. That 40% generated 250% of the company's total profit. The bottom 10% destroyed 120%. The single largest customer, the one the sales team considered their most important account, was the most unprofitable of all.

Order margins ranged from -179% to +65%. Kanthal had been losing money on specific customers for years. The accounting system couldn't show it because all indirect costs sat in one bucket called "overhead."

Harvard professor Robert Kaplan wrote it up as a case study. It is still one of the most taught cases in the MBA program. The method behind it, Activity-Based Costing, became a field of its own.

Ridderstråle didn't just study the problem. He changed sales compensation from volume to profit. He automated order handling for small accounts. He refused non-stock orders from customers who didn't justify the cost. He ran weekly profitability campaigns inside the company.

It worked. Unprofitable customers became profitable, often without damaging the relationship.

Every textbook ends the story here. What happened next is more interesting.

What happened next

In 2006, two researchers from the Stockholm School of Economics decided to go back and check. Seventeen years had passed since Kaplan's famous case. They visited Kanthal. They interviewed nine people across the organization.

Kanthal had abandoned the method.

Customer profitability was "only partly being measured." The company had built new tools since then (a BI system, a pricing waterfall, a customer classification) but none of them worked properly. In one division, management solved the A-B-C customer classification by classifying every customer as an A-customer. Done. No differentiation needed.

Two employees who had been there for over 25 years explained what happened:

During the good years, we could take any order. We did not need to think because we could always show a positive result. Therefore customer profitability was ignored.— Göran Källström, 25-year Kanthal employee

The Sales Manager, David Öquist, described how profitability data actually got used: "We always try to chase the black sheep. We always try to hit the least profitable customers every year, and when one is improved, there are always ten in line. Somebody always has to be the least profitable."

Then he said something the researchers highlighted: "We have found that it is usually not the customers' fault that they are not as profitable as we would have wished. It is rather our own mistake."

The researchers' conclusion was blunt. The sales team held the real power. Not officially. Not loudly. But they had "the decisive, but unspoken vote in whether a project is going to be successful or not."

Nobody sabotaged anything. The sales team kept doing what they had always done. Without a CEO pushing, the system decayed on its own. The most famous customer profitability case study in the world, the one still taught at Harvard, and the company that originated it couldn't sustain it for two decades.

The numbers

The numbers themselves are not in dispute. Kaplan and Narayanan at Harvard, looking across hundreds of companies: the top 20% of customers generate 150-300% of total profits. The middle 60% roughly break even. The bottom 20% destroy 50-200%.

SPARXiQ, which has analyzed invoice data from hundreds of distributors: most companies sacrifice one-third to one-half of their operating profits to customers they can't even identify as unprofitable.

The data is solid. The data has been solid for 37 years. That is not the problem.

Why they don't act

The problem is that knowing and doing are completely different things, and the distance between them is larger than most people realize.

Start with the fact that the data is usually invisible. Most companies calculate customer profitability as revenue minus direct costs. That misses freight, returns, order processing, service calls. Two customers with identical gross margins can have wildly different real profitability. But if the ERP doesn't show it, the conversation never starts.

Then the incentives. Sales teams are paid for volume. Repricing a customer means a difficult conversation. Losing a customer is visible and gets you blamed. Continuing to serve them at a loss is invisible. Nobody gets fired for quietly losing $50,000 a year on a customer that everyone considers important.

Then the champion problem, which is what killed Kanthal. Ridderstråle built the system. When his attention shifted, nobody picked it up. I have not found a single case study where customer profitability measurement survived the departure of its sponsor. Not one.

And underneath all of this is something simpler: acting is risky in a way that not acting isn't. If you reprice a customer and they walk, that's your decision. If you keep serving them at a loss for another five years, it shows up as gradually declining margins that nobody can attach to any particular cause. The safe choice is to do nothing. So most companies do nothing.

The ones who did

Some companies have broken out of this. What they share is a CEO who decided it was their problem and stayed on it long enough for the organization to absorb the change.

Fastenal closed 40% of its locations and lost 43% of its customer base over 17 years. Sales more than doubled. Profitability hit record levels. But this was not a project. CEO Dan Florness drove it patiently, year after year, until it became how the company operates. 87% of their revenue now comes from 10% of their customers.

A German technical wholesaler discovered that 30% of its revenue came from customers with negative contribution margins. Within six months of acting on that, profitability improved by 1.8 percentage points.

A CPG company found that its easiest customers were quietly subsidizing its most demanding ones. After figuring out the real cost-to-serve, EBITDA went up 12%.

None of them did anything exotic. They looked at the data, then decided to act on it. That second step is where it always breaks down.

The bottleneck

The technology to see this problem has been around since 1989. In 2026 it is essentially free. That was never the bottleneck.

The bottleneck is whether someone inside the company decides this is their problem. Not a consultant. Not a software vendor. Not an analyst. Someone with authority who will stay on it when the sales team pushes back, when other priorities crowd in, when the quarterly numbers look just good enough to put it off for another quarter.

In 1979, Merrifield stood in front of 150 executives who all knew they had the problem. Forty-seven years later, the same charts get shown at the same conferences with better formatting and nicer slide decks.

The pattern that keeps repeating is not that companies lack information. It's that they lack someone willing to act on it.

The few that do act tend to find money they didn't know they were losing. The rest keep attending conferences about it.