The profitability paradox: is your largest customer your most valuable customer?
The profitability paradox is the possibility that a company’s highest-revenue customer becomes its least profitable, or even loss-making, customer after all service costs are deducted. For FMCG manufacturers, distributors, and companies with recurring B2B sales, this is not a theoretical nuance. It is an error that directly distorts resource allocation: the largest customer receives the best terms even though its net contribution may already have turned negative.
The source of the problem is a simple asymmetry. Revenue appears in one place as a large, clear number. The cost of serving that customer is scattered across different systems as discounts, returns, logistics, collections, and service. Management looks at the visible large number and overlooks the fragmented cost. The customer that “buys the most” is therefore automatically treated as the “most valuable.” That assumption is often wrong, and when it is wrong, it is expensive.
In brief
- According to the “whale curve” distribution derived from Robert Kaplan and V.G. Narayanan’s customer-profitability work (Journal of Cost Management, 2001), the most profitable 20% of customers generate far more than total reported profit, approximately 150–180%; the large middle group roughly breaks even; and the least profitable 20% consumes a significant share of that profit (as summarized by Lake Ridge Bank).
- In a CPG case reported by a consulting firm, special displays, customer-specific pallets, and rebate demands had quietly eroded margin. Once cost to serve became visible by customer, the company improved EBITDA (Plante Moran, July 2025). This is a case report, not an industry average.
- The problem is not the model, but the measurement: revenue is visible, while cost to serve is fragmented and hidden. This is why “large” is assumed to mean “valuable.”
- The correct metric is not revenue, but net profitability after cost to serve. Once that becomes visible, the customer base is reordered.
Why can the largest customer become the least profitable?
Large customers often sit at one of the two extremes of the profitability distribution: either highly profitable or seriously loss-making. In the whale-curve analysis derived from Kaplan and Narayanan’s 2001 customer-profitability research, the most profitable 20% contributes far more than total reported profit; the large middle group roughly breaks even; and the bottom 20% gives back a significant portion of the profit (as summarized by Lake Ridge Bank). A large customer rarely sits comfortably in the middle of this distribution.
The reason is that large volume usually comes with large demands: deep discounts, customized orders, frequent and fragmented deliveries, long payment terms, and intensive service. As one cost-to-serve analysis notes, many companies cannot state with confidence whether their largest customer is more or less expensive to serve than their smallest one. These costs grow with volume, and when they are not recovered in price, margin disappears.
Why does revenue hide cost?
Revenue is attractive because it is visible and consolidated; cost is fragmented. Discounts sit in the sales system, returns in another record, logistics in operations, collection delays in finance, and service burden is often not measured fully anywhere. Because these items are distributed across different systems, they do not appear together in a single report.
Cost to serve is the method that brings this fragmented cost together for a customer, channel, or segment, from the sale through last-mile logistics. The problem is that many companies either never calculate it or do so retrospectively and manually by pulling data from multiple sources. The result is that revenue appears immediately and clearly, while cost appears late and incompletely. What is invisible at the moment of decision is usually ignored.
Same revenue, different profit: where does the difference come from?
Two customers may generate the same revenue and even the same gross margin, yet produce completely opposite net contributions. The difference lies in the layers of service cost beneath revenue.
| Dimension | Higher-cost customer | Lower-cost customer |
|---|---|---|
| Discount and rebate | Deep and heavily negotiated | Close to list price |
| Return rate | High | Low |
| Order structure | Frequent, small, urgent | Less frequent, planned, large |
| Payment behavior | Late and cash-flow intensive | On time |
| Service and special treatment | Intensive and customized | Standard |
| Net profit at the same revenue | Low or negative | Clearly positive |
No single row in the table determines profitability; their accumulation does. In real company data, that accumulation can produce striking results. In one academic cost-to-serve study, 80% of post-service-cost margin came from only 6% of customers. Value is not simply a function of revenue; it is a function of the conditions under which that revenue is generated.
How does the paradox distort decisions?
When the profitability paradox is invisible, the company systematically directs resources in the wrong direction. The largest but unprofitable customer is labeled “strategic” and receives even better terms, deepening the loss. The greatest concession in a price negotiation is given to the customer buying the most, even though that customer may already operate at the lowest margin. The best service resources are allocated there, although the same resources could create much more value with a cleaner customer.
The reverse is also true: when net profitability becomes visible, many decisions improve. In a CPG case reported by Plante Moran, management believed it had negotiated prices effectively, but special displays, customer-specific labels and pallets, detailed reporting, and rebate demands had inflated cost. When cost to serve was made visible by customer across five categories, materials, conversion, warehousing, transport, and service/rebate/discount, the company reset pricing and service levels and materially improved EBITDA.
Should an unprofitable customer be dismissed?
Recognizing the profitability paradox does not mean eliminating every low-profit customer. Confusing the two is a common and expensive mistake. Some low-profit customers are strategically important: they provide volume, spread fixed cost, protect market share, establish presence in a channel, or offer genuine growth potential. Cutting them blindly may shrink the business rather than improve it.
The objective is not to punish the customer, but to decide with a clear view of reality. Once net profitability becomes visible, new options appear: renegotiate conditions with one customer, invest more in another, deliberately carry a third at low profitability, or encourage a fourth to change ordering and delivery behavior. Even if the ultimate decision remains unchanged, it is now informed rather than blind.
Conclusion
Treating the customer that buys the most as the most valuable is intuitive, but often wrong. Revenue is visible; the cost of serving the customer, discounts, returns, logistics, collections, and service, is fragmented and hidden. Once those costs are deducted, the largest customer may prove to be the least profitable.
This is not an accounting subtlety; it is a resource-allocation problem. Until net profitability is visible, the company systematically gives its best terms and best service to the wrong customer. True value lies not in revenue, but in profitability after cost to serve. Once that becomes visible, the customer base is reordered.
How does GDP build it?
- Which data we collect: Revenue, gross margin, discounts/rebates, returns, logistics, payment terms, and service burden by customer.
- Which model we build: A cost-to-serve layer that consolidates fragmented costs at customer, channel, and segment level.
- Which decision we connect it to: A customer-level net-profitability ranking that shows where terms should be renegotiated, where to invest, and which relationships are deliberately maintained at lower profitability.
We build this visibility layer on top of existing sales and finance data through our Commercial Intelligence service.
Frequently asked questions
What is the profitability paradox?
It is the possibility that the highest-revenue customer becomes the least profitable or even loss-making customer after all service costs, discounts, returns, logistics, collections, and service, are deducted. Revenue is large and visible, while cost is fragmented and hidden. The “largest” customer is therefore mistakenly treated as the “most valuable.” True value is measured through net profitability, not revenue.
Why can the largest customer be the least profitable?
Because large volume often comes with large demands: deep discounts, customized orders, frequent and fragmented deliveries, long payment terms, and intensive service. Kaplan and Narayanan’s research suggests that large customers often sit at one of the two extremes of the profitability distribution, highly profitable or loss-making. If growing service cost is not recovered through price, margin erodes.
What does cost to serve include?
It includes every cost beneath gross margin associated with serving a customer, channel, or segment: discounts and rebates, returns, warehousing, transport and last-mile logistics, order processing, collection and financing burden, and service or support. These costs are spread across systems and therefore need to be consolidated deliberately.
Should I immediately stop serving an unprofitable customer?
Usually not. Some low-profit customers are strategic: they provide volume, spread fixed cost, protect market share, or offer growth potential. Renegotiating price, order structure, delivery patterns, or service levels is often a better first step. The decision should be based on actual value, not used to punish the customer.
Where should we start to make customer profitability visible?
Begin with one customer slice. Start from revenue and gross margin, then deduct discounts, returns, logistics, collections, and service costs by customer. The goal is not perfect cost accounting, but a sufficiently accurate net-profit view to change the ranking. Most of the required data already exists in sales, finance, and operations systems.
Sources
Academic: Robert Kaplan and V.G. Narayanan, “Measuring and Managing Customer Profitability” (Journal of Cost Management, 2001), the source of the whale-curve distribution; summarized by Lake Ridge Bank. Guerreiro et al., cost-to-serve measurement and customer-profitability analysis: 80% of post-service-cost margin generated by 6% of customers.
Consulting and case evidence: Plante Moran (July 2025), cost-to-serve visibility and EBITDA improvement in a CPG case, case evidence, not an industry average. West Monroe, the frequent lack of clarity regarding the cost of serving large customers. Coupa, definition of cost to serve.
Last reviewed: July 2026.