How to Tell If Some of Your Customers Are Costing You Money

Key Takeaways

A customer who buys often or places big orders can still contribute very little to your profit. Looking at what each account costs to serve, not just what it pays, shows where the money actually goes.

• Compare customer revenue with the direct and shared costs of serving each account.
• Track staff time, discounts, fulfillment, returns, disputes, and payment delays.
• Use consistent allocation rules and a measurement period that fits your business.
• Compare similar customers to find patterns, not just isolated low-margin accounts.
• Reprice, reset service terms, or reconsider a relationship based on the evidence.

Understanding what customer profitability means

A customer who buys often or places large orders can still contribute little to profit. The answer depends on more than the price of the product or service: it also depends on the work required to win, serve, and retain that account. Picture two customers who spend the same amount each year. One orders on schedule, pays on time, and rarely calls. The other needs custom handling, negotiates new terms every quarter, and pays weeks late. On a revenue report they look identical, but they are very different relationships. The aim isn’t to label customers as good or bad. It’s to understand the economics of each one.

Revenue records what a customer pays. Profit is what remains after the costs tied to that business are counted. A high-volume account can look valuable until discounts, delivery demands, or repeated service interventions come into view, and that gap is easy to miss when reports are organized around total revenue alone. A more useful question is what the account contributes after the costs of meeting its specific needs.

Some of those costs are easy to attach to an account, like the materials used for an order or a dedicated delivery. Others are shared: customer service, account management, warehouse space, and systems all support many customers at once. Include both, but be clear about which figures are measured and which are allocated. Otherwise a cost-to-serve estimate can look precise while quietly shifting overhead from one customer group to another.

It also helps to keep customer profitability separate from lifetime value. Profitability describes the economics of a relationship over a defined period. Lifetime value is a longer-range estimate that builds in expected future purchases, retention, and other assumptions. A low-margin account today might grow into a profitable one, but test that possibility rather than presuming it. Treat expected future value as its own consideration, apart from the profit already earned.

Calculating the profit each customer contributes

The calculation doesn’t have to begin with a sophisticated model. You need a reliable view of customer-level revenue, the costs you can trace directly, and a consistent way to spread shared expenses. Start with the records your teams already keep, then note where missing or estimated data could change the result. Think of it as a decision aid, not a claim of perfect precision.

Bring together sales, credits, discounts, returns, and the direct cost of fulfilling work for each account. Where you can, connect service activity and staff time to the customer too, since a small number of repeated requests can add up to a real workload. Agree on how to treat one-off charges and adjustments before comparing accounts. That gives your teams a shared starting point and makes gaps in the records visible.

Shared expenses need a fair, repeatable allocation rule, even if the data is imperfect. A few common pairings show how this works:

• Customer support: allocate by recorded time or cases, and watch for uneven case complexity.
• Fulfillment: allocate by orders, units, or handling time, and watch for different order requirements.
• Account management: allocate by recorded account hours, and watch for relationship work nobody logs.
• Payment costs: allocate by transaction value or fees, and watch for different payment methods.

The best driver is the one that most closely reflects why the expense occurs and can be applied the same way every time. Write down the rule alongside the result, so nobody mistakes an allocation choice for a measured fact. If the numbers suggest an account is underpriced, test your assumptions before changing a price.

Pick a measurement period that gives a representative view. Too short, and ordinary fluctuations can look like a trend. Too long, and you can hide a recent change in service needs or pricing. Many businesses start with a recent quarter or year, then check whether seasonality or large one-off events distort the picture. When a customer’s needs change substantially, compare periods rather than blending unlike conditions together.

Spotting warning signs and finding where the costs come from

A warning sign is a reason to investigate, not proof that an account is unprofitable. A high level of activity might be costly, but it could also reflect a temporary A warning sign is a reason to investigate, not proof that an account is unprofitable. A high level of activity might be costly, but it could also reflect a temporary launch, a service issue you need to fix, or an intentional investment. Look for repeated patterns and compare them with the account’s revenue and margin. In many cases the clearest signal is several small demands piling up over time.

Frequent support requests are a common one. Calls, escalations, and requests for clarification can absorb more staff time than the account’s revenue supports. Look at both the volume of requests and the effort to resolve them, since ten quick questions are different from ten complex investigations. Check whether the requests point to an avoidable problem, like unclear instructions, before treating them as a permanent cost of the relationship. A short review of service activity brings hidden effort into view, so track the kinds of work that recur rather than relying on a general impression that an account is “demanding”:

• Count support cases and escalations by account.
• Estimate staff time spent resolving each type of request.
• Note repeat issues that may have a preventable cause.
• Compare the effort with the account’s revenue and contribution.

Discounts and special terms deserve the same scrutiny. Rebates, extended payment terms, and exceptions can turn a healthy-looking price into a weak net return, so review the terms actually applied, not only the list price or the original contract. Unmanaged concessions are a common form of price leakage, especially when older terms stay in place or approvals are handled inconsistently. Returns and credits reduce realized sales, while disputes and late payment consume time and make it costlier to collect what you’re owed. Consider the frequency and reason for each event rather than treating all adjustments alike. A return caused by a fulfillment error may call for an operational fix, while repeated disputes over agreed charges may call for clearer documentation or terms.

Once an account looks less profitable than expected, the next step is finding the source. Look at how often orders arrive, how they’re handled, and what special delivery or support they need. In distribution, that might be order size, delivery frequency, and location. In other settings it might be implementation hours, onsite visits, or the number of service handoffs. Small exceptions can also become routine: extra reporting, repeated revisions, bespoke packaging, or requests that fall outside the original agreement. Record the work and who is doing it, then compare the pattern with the account’s scope and price. Some customization is worth it strategically, but it should be a choice rather than invisible labor. Finally, decide which acquisition and payment-processing costs belong in the customer view, and apply the same rule to comparable accounts so you aren’t mixing unlike definitions.

Comparing profitability across customer groups

A single account can reveal an issue, but comparisons show whether it’s an exception or a broader pattern. Group customers in ways that affect the work involved, such as product mix, sales channel, order profile, or service needs, and compare like with like where possible. Avoid groups so broad that their differences disappear. A channel may carry different selling costs, a product category may need more handling, and one service tier may involve far more support. Keep the groups practical enough that your teams can act on what you find.

Once accounts are grouped, check whether high costs cluster around common behaviors rather than customer labels. Perhaps small, frequent orders demand more handling, or one type of custom request drives staff time. Patterns like these can point toward a process change or a better-defined offer that helps more than one account. Don’t assume the customer is the cause when an internal process may be creating the expense.

Low current profit isn’t always bad news. A customer who is new, growing, or expected to adopt more services may be worth the investment, but that expectation should have a clear rationale, a time horizon, and signs of progress. Compare the actual trend with the original expectation, and include the cost of serving the account during that period. If the future value is still speculative, keep it separate from the profitability you’ve already observed.

Deciding how to respond to unprofitable accounts

A low margin doesn’t dictate one response. The right choice depends on why the account is costly, whether that cost can change, and how the relationship fits your broader priorities. Make sure the analysis is strong enough to support a conversation, and that the people closest to the account can explain the facts. Then choose a remedy that addresses the cause, not just the symptom.

If the work is valuable to the customer but the price doesn’t cover its cost, a price or package change may be the answer. Base the discussion on the specific service requirements and the value provided, not on an arbitrary surcharge, and consider whether a different service level would fit the customer better. When you’re ready to raise a price, our guide to communicating a price increase to customers covers the timing and messaging.

Some costs shrink once the terms are clearer. Spell out what’s included, what counts as extra work, what order requirements apply, and how exceptions will be handled. The aim isn’t to make service difficult. It’s to align expectations so your teams can deliver consistently and customers understand the arrangement. Afterward, check whether the new terms actually reduced the effort or risk that prompted the change.

If a relationship stays uneconomic after reasonable changes, consider whether a lower service level or an orderly exit is the sounder choice. Weigh the consequences first: the revenue effect, contractual commitments, your capacity, and how strategic the customer is. Document the rationale and bring in the right decision-makers before acting. A deliberate decision beats letting an account keep consuming resources by default.

Building an ongoing customer profitability review

Customer economics shift as prices, volumes, service needs, and internal costs change. A one-time analysis gives you a useful baseline, but it won’t tell you whether an earlier fix worked or whether a once-attractive account has become more demanding. Set up a review your teams can maintain and tie it to decisions they already make. Keep the process consistent without pretending every relationship is the same.

Choose a small set of measures that make the economics easy to read. Net revenue, contribution margin, service hours, fulfillment activity, discount levels, and payment performance are good starting points. Use definitions that different teams apply the same way, and add notes where the data is estimated. A measure earns its place when it helps someone make or review a decision, not just because it fits on a dashboard. Pair the financial results with the operational drivers behind them.

Set thresholds that flag an account for review when margin falls, service activity rises, or discounts exceed an agreed range. A flag should prompt a look, not an automatic penalty, since a temporary change may have a perfectly reasonable explanation. Set the trigger early enough that your teams can act before an issue becomes entrenched, and be clear about who investigates, what evidence they need, and how decisions are recorded. Revisit an account’s economics, too, when its order pattern, product mix, service expectations, or contract terms change materially. Compare the new situation with the assumptions behind the current price and service model, then decide whether the relationship still works as designed or needs a different arrangement.

Conclusion

Spotting unprofitable customers gets easier once you look at revenue alongside the work and costs required to earn it. Track those costs consistently, investigate patterns before drawing conclusions, and respond with changes that fit the account and the business. Some relationships will need new terms, and others will simply show you a process worth improving.