The customer who costs too much to win — and why it isn't a marketing problem

Every commercial leadership team has asked the question: how do we attract more customers in a segment we don't serve well yet? The default answer is funnel-shaped — more leads, more visits, more presence. But in nearly every distributor there's one segment that responds to none of those levers, no matter how cheap the lead gets. It isn't a demand problem. It's that winning that customer doesn't pay off — and the company knows it in the math, even when no one ever wrote it down.

The acquisition math no one finishes

Customer acquisition cost usually gets treated as a marketing number: what I spent to generate the lead that became a sale. That number is incomplete. The true cost of winning a customer includes everything you'll spend to serve them for as long as they're a customer — and in B2B distribution, serving isn't cheap.

A R$2,000 order triggers the same operational cycle as a R$200,000 one: someone answers the technical question, someone validates the information, credit gets checked, stock gets confirmed, the confirmation goes out. Spread that across departments and you've spent roughly an hour of work to process two thousand reais. The margin on that order doesn't cover the operation that produced it.

So the small customer isn't expensive because the lead is expensive. They're expensive because the cost to serve them is fixed and the revenue they bring is small. That's the acquisition cost that never makes it onto the spreadsheet — and it's the one that actually decides whether the segment is reachable.

The decision no one makes

No distributor puts "we've decided not to serve small accounts" in the minutes. The decision happens another way, subtler and more dangerous: orders from small accounts get slower, require more validation, get less of the rep's attention. The customer feels the friction. And they leave — for a competitor that managed to process it without friction.

The result is involuntary operational redlining. The company never refused the segment; it just made the segment unwelcome until it gave up. The lost revenue never shows up as a loss, because it was never captured. It's the kind of leak that doesn't hurt because it's invisible: you don't lose the customer you had, you simply never win the one you could have.

Why the small customer wants to buy more often — and why that makes it worse

There's a structural twist. The small customer rarely wants one large, consolidated purchase. They want to buy little and often, as real need shows up — one item today, another Wednesday, another Friday. It isn't a whim: capital tied up in inventory is a cost, and tight cash flow is the reality of running small.

For a distributor built on the traditional model, that's the worst case. Five R$400 orders in a week don't cost a fifth of one R$2,000 order — they cost five times as much to process, because each runs the full operational cycle. The more the small customer behaves like a small customer, the more expensive they get. The segment that looked like "easy revenue" is in fact the most expensive to serve per real billed.

What Tracbel saw when the friction went away

Tracbel is one of Brazil's largest capital-goods distributors — the exclusive distributor of Volvo construction equipment and trucks, 40 branches, 550 people dedicated to aftersales. Five years ago, it orchestrated the entire order flow end to end: catalog and stock in real time, orders flowing straight into the ERP with no re-keying, credit validated on the spot. The effect on the acquisition math was structural — when the operational cost per transaction falls close to marginal, the R$2,000 customer stops costing more than the size of their own order.

And what five years of data show isn't just cost savings. It's behavior. One customer came in during June 2020 with a first order of R$399.76 — a test, someone checking whether they could operate on their own. Twenty-six months later, they'd placed 75 orders and accumulated more than R$75,000. Another, with a seasonal profile, bought sparingly — about 1.4 orders a month — and still always came back, reaching more than R$54,000. A third started with R$867.00 in 2019 and, five years on, had placed 684 orders worth more than R$3.5 million.

Three completely different buying cycles. Three customers who, under the old model, would never have made it past the trial — because every small order would have cost more attention than it was worth. The point isn't the growth multiples. It's that none of them disappeared. When buying becomes frictionless, the small customer isn't an expense: it's an account that starts tiny and grows.

Why this is acquisition, not service

Here's the inversion that changes the strategy. The first R$400 order isn't a small customer. It's a large customer at month zero. What separates one from the other is decided entirely by unit economics: if serving that customer costs more than they pay, they die as a trial; if it costs near zero at the margin, they have room to grow inside your channel for the next five years.

That's why winning the small customer isn't a marketing problem — it's a cost-to-serve problem. You don't need cheaper leads. You need an operation where the cheap lead finally pays off. And when that operation exists, the prize isn't a customer: it's reach itself. Revenue spread across a broader base lowers the risk concentrated in a handful of large accounts. Every tiny customer today is the mid-sized customer in three years and the structured account in seven. Multiplied across the base — in Tracbel's case, thousands of customers and more than R$122 million in digital revenue in 2025 — the segment that was unviable becomes the engine of growth.

Frequently asked questions

Does this mean lowering the price for small customers? No — and that's the most common confusion. The problem with the small customer isn't that they pay little; it's that they're expensive to serve. Cutting price attacks the wrong side of the math and worsens the margin. What has to fall is the operational cost per transaction, not the price. When cost-to-serve trends toward zero at the margin, today's price is already enough to make the customer profitable.

How do we know if we're abandoning a segment without realizing it? The signs are operational, not commercial. Small orders with longer response times than large ones; more validation steps for low-ticket accounts; reps who, understandably, prioritize whoever pays the bigger commission. If your operation implicitly treats the small order as a nuisance, the small customer notices — and the frictionless competitor says thank you.

Isn't it simpler to concentrate on large customers? Simpler and riskier. A base concentrated in a few large accounts is fragile: losing one of them opens a hole. The capillary base is the opposite — distributed risk and compounding growth, because the small customer who operates well in your channel tends to buy more and to stay. Tomorrow's large customer almost always started small somewhere. The question isn't whether the small account is worth serving; it's what it costs to keep them.

The operational data cited here belongs to Tracbel, a CWS Platform customer, used with permission.