What Service Levels Are Worth Paying For?

Customers rarely ask for less service. They ask for shorter lead times, higher availability, smaller order quantities, later cutoffs, special packaging, dedicated support, more flexible delivery windows, and faster recovery when something goes wrong. Commercial teams have good reasons to say yes: service protects revenue, strengthens relationships, differentiates the offer, and can be difficult for competitors to match.

Operations experiences the other side of the promise. Higher availability may require more inventory. Shorter lead times can require excess capacity, premium transportation, or more decentralized stock. Smaller orders increase touches and transportation cost. Customization fragments flow. Special handling creates exceptions. The customer sees a service level; the operating system absorbs the complexity required to deliver it.

The executive issue is therefore not whether service matters. It clearly does. The issue is whether the organization understands the full economics of the promise—and whether different customers, products, and channels should receive the same promise at all.

Service Is Not Free—and Neither Is Poor Service

The tension is easy to oversimplify. Cost-focused organizations can protect margins by reducing service, only to lose revenue, loyalty, or strategic accounts. Service-focused organizations can protect customers while quietly destroying margin and tying up cash. The correct answer is rarely "more service" or "less service." It is a deliberate service architecture tied to economic value.

McKinsey documented the problem in a chemical company whose commercial team pushed logistics to reduce delivery time from three days to two. The company achieved the faster promise, but warehouse space, labor, and less efficient truck loading increased distribution costs by 5 percent. Nothing about the two-day service target was inherently wrong. The failure was making the promise before understanding its operating economics.

The opposite mistake is equally real. Service failures create lost sales, penalties, customer defections, and reputational damage. PwC has argued that product availability must be balanced against cost and working-capital requirements and recommends pairing service initiatives with customer- and SKU-level profitability analysis. The executive task is to understand both sides of the curve.

Cost-to-Serve Makes the Economics Visible

Traditional margin reporting can conceal the cost of the service model. Two customers may buy the same product at the same price and appear equally attractive in a standard product-margin report, yet one may place predictable pallet-size orders while the other requires frequent small shipments, expedited transportation, special labeling, technical support, returns processing, disputed invoices, and dedicated inventory.

Activity-based costing was developed precisely to address this kind of distortion. Robin Cooper and Robert Kaplan showed that costs should be traced to the activities that consume resources rather than spread broadly across products or customers through simplistic overhead allocations. Their work later extended directly into product and customer profitability. Kaplan and Steven Anderson's time-driven activity-based costing refined the approach by estimating the cost of capacity and the time required to perform transactions, making it easier to apply in complex operations.

Cost-to-serve uses that logic to answer an executive question that gross margin cannot: what does the company actually spend to fulfill the commercial promise to this customer, product, channel, or service segment?

Those costs can include order processing, pick and pack labor, transport mode, delivery frequency, inventory carrying cost, special packaging, customer-specific production changeovers, technical support, returns, claims, account administration, field inventory, payment terms, and the capacity reserved to protect a service commitment. Once those costs are visible, service stops being an abstract customer virtue and becomes an investable economic choice.

The Customer Profitability Matrix

A useful established lens comes from Benson Shapiro, V. Kasturi Rangan, Rowland Moriarty, and Elliot Ross. Their customer profitability work separates customers by the price received and the cost required to serve them. The resulting matrix makes a simple point with major operating consequences: high revenue does not automatically mean high profit, and the same service model should not be applied indiscriminately across the customer portfolio.

The most profitable quadrant is often the "passive" customer: high price realization and relatively low cost to serve. These relationships can be easy to overlook because they do not generate operational noise. Yet they may deserve disproportionate protection precisely because the economics are attractive.

"Carriage trade" customers pay well but are expensive to serve. Premium service may be entirely rational if the price, strategic value, and lifetime economics compensate for the resources consumed. "Bargain basement" customers pay less but are also inexpensive to serve; a standardized, low-complexity model can make those relationships healthy. The dangerous quadrant is "aggressive": low price realization combined with high service demands. The appropriate response is not automatically to exit. It may be to reprice, redesign the promise, change order behavior, reduce exceptions, or move the account to a different service model.

This is where service-level strategy becomes executive work. The question is not which customers "deserve" better service. It is which promises the company should fund because they create sufficient economic, strategic, or contractual value.

Uniform Service Is Usually a Hidden Subsidy

Many organizations establish enterprise service targets because standardization simplifies planning and creates a clear customer promise. That can be useful, but uniform targets can also hide cross-subsidies. A 98 percent availability target may be economically rational for a high-margin, high-velocity product and irrational for a slow-moving item with uncertain demand. A next-day delivery commitment may protect a strategic customer while destroying economics on a small, price-sensitive account.

The subsidy appears because service cost is pooled. Inventory held for low-volume items is justified by an overall fill-rate target. Premium freight is absorbed in distribution expense. Customer-specific work is treated as overhead. Commercial discounts and operating complexity are evaluated separately. Each decision looks manageable, while the cumulative service architecture consumes margin and cash.

A segmented service model makes those choices explicit. Different customers or products can receive different availability targets, replenishment rules, lead times, order minimums, support levels, or recovery commitments. The segmentation should not be arbitrary. It should reflect the economics of the relationship and the strategic consequence of failure.

The Decision Is Larger Than Customer Profitability

Cost-to-serve is necessary, but it is not sufficient. A customer can be expensive today and still be strategically important. A low-volume spare part can appear unattractive on a narrow SKU P and L while being essential to an installed-base service contract. A new channel can carry poor current economics but create future scale, data, or market access. An enterprise account may impose unusual requirements while influencing a much larger portfolio of business.

This is why the executive question cannot be reduced to a simple profitability ranking. Cost-to-serve should reveal the economics; strategy must determine whether those economics are worth carrying. The discipline is to make the subsidy conscious rather than accidental.

In practical terms, leadership should distinguish among at least four sources of value: current contribution, future economic potential, strategic importance, and the consequence of service failure. Higher service is justified when one or more of those sources of value exceed the incremental cost and capital required to support the promise. When they do not, the service model should be challenged.

A Practical Example: The 24-Hour Promise

Consider an illustrative industrial distributor that has built its brand around rapid availability. Over time, the promise has expanded. Strategic accounts receive next-day delivery, but so do hundreds of smaller customers. Sales teams can request expedited shipments with limited approval. Slow-moving parts are stocked across multiple locations to protect local fill rates. The published service metric looks strong, yet inventory and premium freight continue to rise.

A cost-to-serve analysis changes the conversation. The largest strategic accounts generate attractive contribution despite high service costs because their pricing, volume, and broader relationship justify the investment. A second group of customers is profitable primarily because their ordering behavior is predictable and inexpensive to support. A third group buys at highly negotiated prices while generating frequent small orders, expedited requests, special handling, and disproportionately high returns.

The company has been giving all three groups versions of the same promise, but the economics are entirely different. The answer is not to abandon rapid service. It is to protect it where it creates value and redesign it where the company is funding customer convenience without an adequate return.

For the high-value segment, the company may preserve the 24-hour commitment and strategically position inventory to support it. For predictable low-cost accounts, it may maintain competitive service through standard replenishment and consolidated delivery. For high-cost, low-price relationships, it may introduce order minimums, scheduled delivery days, expedited-service charges, different stocking rules, or commercial repricing.

The operational outcome can be significant because service policies drive network design. Change the promise and the required inventory, capacity, transportation, labor, and exception handling can change with it. That is why service-level decisions should not be delegated entirely to sales or supply chain. They shape the economics of the operating model.

The Counterargument: Superior Service Can Be the Strategy

There is a legitimate risk in applying cost-to-serve too aggressively. Companies can rationalize away the very service attributes that differentiate them. Amazon, industrial aftermarket businesses, healthcare supply networks, and premium B2B models can create substantial strategic value precisely because customers trust availability, speed, or recovery performance that competitors cannot easily match.

Some service investments also create scale benefits that are difficult to assign cleanly to individual customers. Network density can lower unit transportation cost. Higher inventory availability can support multiple revenue streams. A superior service reputation can improve retention and pricing power. Cost models are estimates, and they can produce false precision if executives treat every allocated dollar as economically avoidable.

The answer is not to weaken the service proposition. It is to know which parts of the proposition create differentiated value and which have accumulated through historical precedent, commercial negotiation, or unmanaged exception. Cost-to-serve should challenge the promise, not automatically shrink it.

Service Policy Is a Portfolio Decision

The most important shift is conceptual. Service level should not be treated as a universal operating KPI detached from the economics of the customer or product. It is a portfolio of promises, each consuming a different amount of inventory, capacity, labor, transportation, and management attention.

Executives should therefore ask whether the company can see the economic cost of each major promise, whether customers and products are meaningfully segmented, whether premium service is priced or strategically justified, and whether poor economics are being addressed through pricing, behavior change, process redesign, or service differentiation.

The objective is not to minimize cost-to-serve. A company that does that perfectly may become exceptionally efficient at disappointing customers. The objective is to allocate service investment where it produces the greatest enterprise value.

The strongest service strategy is not the one that promises the most. It is the one that knows exactly which promises are worth paying for—and has designed the operation to deliver those promises reliably.

Selected Sources

Shapiro, B. P., Rangan, V. K., Moriarty, R. T., and Ross, E. B.. Manage Customers for Profits (Not Just Sales). Harvard Business Review, September 1987. Source

Cooper, R., and Kaplan, R. S.. Profit Priorities from Activity-Based Costing. Harvard Business Review, May–June 1991. Source

Kaplan, R. S., and Anderson, S. R.. Time-Driven Activity-Based Costing. Harvard Business Review, November 2004. Source

PwC. A New Roadmap for CPG Pricing Strategies and Growth. 2024. Source

McKinsey and Company. Is Your Top Team Undermining Your Supply Chain?. 2010. Source

McKinsey and Company. Turn Slow-Moving Inventory into Fast Profits. November 2019. Source