Combine revenue, discounts, product costs and cost to serve so sales decisions reflect profitability.
The largest turnover may not produce the best outcome
A major customer may buy heavily but require discounts, frequent deliveries and many service hours. A high-percentage-margin product may sell little or generate returns. Add results after clearly defined costs to revenue rankings. The point is not to reject demanding customers automatically, but to identify which commercial terms or processes need to change.
Define the layers of the result
Start with net revenue after discounts and returns, then subtract the agreed cost of goods sold. Show customer-specific service costs separately: deliveries, commissions, extra work or complaints. Name each layer. A result after selected variable costs is not company profit. Shared and fixed costs must remain visible rather than disappearing between different reports.
Compare two customers on one basis
In a simplified example, customer A produces PLN 30,000 revenue, PLN 21,000 product costs and PLN 7,000 service costs, leaving PLN 2,000. Customer B produces PLN 18,000 revenue, PLN 10,800 product costs and PLN 2,200 service costs, leaving PLN 5,000. These figures precede other shared costs. Lower turnover can therefore contribute more towards the company's remaining expenses.
Allocate service costs sensibly
Not every hour can be measured perfectly. Begin with explicit drivers such as deliveries, specialist time or support requests. Mark estimates and test their effect on the ranking. Do not distribute every cost in proportion to revenue when customers require very different work. At the same time, avoid apparent precision that the underlying data cannot support.
Assess a product within the whole basket
A low-contribution product may support purchases of other items. Before removing it, examine baskets, repeat business, inventory and customer requirements. Do not assume that supporting role without evidence. Compare products using consistent costs and periods, including adjustments. Separate price, volume and sales-mix changes because they imply different commercial actions.
Use the result to discuss terms
Options include minimum order size, delivery frequency, free support scope or discounts. Model the change and its effect on the customer before implementation. Do not assume a price increase leaves volume unchanged. Simplifying service can be better than ending a relationship. Decisions should consider the relationship and potential supported by actual trading history.
Review profitability regularly
Once the data is agreed, create an action list of customers and products, with decision owners and review dates. Check whether revised terms improve the overall outcome, not just one metric. Connect the analysis to sales and marketing plans. Spending more to acquire similar customers becomes sensible only when their economic contribution is understood.
Related articles
- Your company has an ERP but cannot explain its profit. Where does sales data get lost?
- Sales are declining. Where should a business start the analysis?
- Attribution versus incrementality: did the ad create extra sales?
FAQ
Should a low-margin customer be dropped immediately?
No. Check data quality, service costs and options for changing terms. A ranking informs a decision; it is not an automatic verdict.
Is margin percentage enough to compare products?
No. Also consider monetary contribution, volume, returns, service costs and the product's role in the basket.
Sources
Information checked: 25 September 2026. Feature access may depend on the account.