How to Read Your Ecommerce Cost Structure

Every example and amount on this page is illustrative. It shows the method and is not a benchmark, a quote or advice.

What Your Cost Structure Actually Measures

An ecommerce cost structure shows how sales revenue is absorbed by the costs of making and selling products, leaving profit. Its four core parts are realized revenue, variable costs, fixed operating costs and profit. Realized revenue should reflect completed sales under the reporting rule you choose; cancelled orders and refunded returns should not quietly remain in the sales base. Variable costs change with order volume or product mix, such as product cost, payment fees, shipping, fulfilment and returns. Fixed operating costs remain broadly stable over a period, such as salaries, software, rent and professional services.

The useful distinction is where each cost sits in the sequence. Revenue less product-level costs produces gross profit. Revenue less the costs that vary with each sale produces a contribution view. Subtract fixed operating costs from the period’s contribution to arrive at operating profit. Advertising may be analysed as a separate layer: contribution after advertising shows what remains after acquiring the sale, while net profit also depends on the other operating costs included in the reporting period. The treatment of COGS deserves a defined policy; see what belongs in ecommerce COGS.

This view is fit for questions such as: Are sales covering the costs required to generate them? Which product or collection has stronger economics? How much fixed-cost coverage is available after variable costs? It is not, by itself, a demand forecast, a price-elasticity estimate or proof that increasing ad spend will improve profit. It also cannot explain performance if returns, inventory costs, ad spend or fixed costs are omitted or mixed between periods. A tool should therefore preserve the layers rather than show revenue alone. ORVX calculates gross and net profit and their margins, and reports profitability by product and collection; its product profitability analysis is designed to keep those cost layers visible.

Put the Calculation on One Line

Put the calculation on one line: realized revenue − COGS − variable selling costs − advertising spend − operating expenses = operating profit. Each subtraction creates a useful layer rather than one undifferentiated cost total.

Start with realized revenue: the sales value you actually keep after cancellations and returns are accounted for. Subtract COGS—the product cost attached to those orders—to get gross profit. COGS should follow a consistent product-cost model; see what belongs in ecommerce COGS. Gross profit answers whether the product can generate money before the costs of selling and running the business.

Subtract variable selling costs such as payment fees, fulfillment, packaging, and other costs that rise with each order. Subtract advertising spend as well to get contribution after advertising. This layer shows what remains to contribute toward payroll, software, rent, agencies, and other operating costs. It is the useful layer for comparing products, collections, channels, and campaigns; product profitability should preserve these deductions rather than rank products by revenue alone.

Subtract operating expenses from contribution after advertising to get operating profit. Operating profit is the broader business result; contribution is the more diagnostic result for deciding whether additional sales are economically useful. Neither measure tells you how to set a price, how much demand a price change will create, or whether a campaign should scale without the relevant volume, cash, and capacity context.

Keep dollars and percentages separate. Cost dollars answer “how much did this cost?” A cost percentage answers “what share of realized revenue did it consume?” Calculate each percentage as that layer’s cost ÷ realized revenue × 100. Gross margin is gross profit ÷ realized revenue; contribution margin is contribution after advertising ÷ realized revenue; operating margin is operating profit ÷ realized revenue. Use the same revenue denominator at every layer, and do not add percentages from different layers.

A Worked Shopify Example

Assume a Shopify store records $100,000 in realized revenue during one month. This is revenue from delivered orders; cancelled orders and returned orders are not counted as sales. Its monthly costs are:

Start with gross profit: $100,000 − $32,000 = $68,000. Gross margin is $68,000 ÷ $100,000 = 68%. The $32,000 product cost is COGS; it should follow the store’s product cost model, not be confused with fulfilment or advertising. See what belongs in ecommerce COGS.

Next subtract variable selling costs. Fulfilment plus payment fees equals $8,000 + $3,000 = $11,000. Contribution before advertising is $68,000 − $11,000 = $57,000, or 57% of revenue.

Subtract the $20,000 marketing spend to measure contribution after advertising: $57,000 − $20,000 = $37,000, or 37%. This is the amount available to cover overhead and produce operating profit. A product-level report should apply the same logic by SKU or collection, as described in product profitability after COGS, returns and ad spend.

Finally, subtract $15,000 of overhead. Operating profit is $37,000 − $15,000 = $22,000, producing a 22% operating margin. This example supports decisions about whether sales volume is covering its direct costs, advertising and overhead. It does not, by itself, explain cash timing, inventory commitments, or whether advertising caused each order.

Where Cost Structure Reports Go Wrong

A report is incomplete when it captures Shopify orders but not the costs of the channels that generated them. For illustration only, $80,000 of store revenue plus $20,000 from a marketplace can look like $80,000 of sales if the marketplace feed is missing. If the report also includes the full $15,000 advertising bill, the resulting margin is understated. Channel revenue, fees and ad spend need the same scope before the comparison means anything.

Classification errors create a different problem. Product purchase invoices are not automatically COGS for the month they are paid; COGS should follow the units sold and the product cost model. Expensing $30,000 of inventory purchases when only $18,000 was sold understates that month's profit by $12,000. The remaining $12,000 is inventory, although it can still become a cost if it turns into dead stock. Define what belongs in ecommerce COGS before comparing periods.

Returns and discounts can make reported sales look healthier than the economics. Suppose orders total $100,000, discounts are $10,000 and refunds are $8,000. Realized revenue is $82,000, not $100,000. A report using gross orders overstates the base against which COGS, fees and advertising are measured. Taxes collected from customers should also be separated from revenue and profit; treating a $2,000 sales-tax liability as sales inflates both.

Finally, a blended average can hide the decision that matters. Product A may generate $100 of realized revenue, $20 of COGS and $30 of advertising, leaving $50 after advertising. Product B may generate the same revenue, but with $70 of COGS and $5 of advertising, leaving $25. The blended result is $37.50 per product, which describes neither one. Check product, collection and channel-level results, including return value and discount treatment, as well as the total. A useful product profitability report preserves those differences instead of averaging them away.

Choose a Tool That Preserves the Economics

A trustworthy calculation starts with complete connections. Verify that the tool imports delivered, cancelled and returned order status from Shopify, product data, customer data and the product cost model—not just gross order value. It should also bring advertising spend beside store revenue, with each merchant’s trading currency kept separate. ORVX connects to Shopify and advertising accounts, and calculates realized revenue from delivered orders only.

Check how costs are assigned before accepting the result. Product costs should follow the product or variant sold, while ad spend should be allocated using a stated rule, such as campaign, channel, product or collection. If a $10,000 campaign promotes three collections, a report that assigns all $10,000 to the highest-revenue collection can distort contribution. A defensible tool shows the rule and lets you trace the amount back to its source. For the underlying treatment of inventory costs, see Ecommerce COGS.

Refresh speed matters when spend and returns change during the month. Ask how often orders, refunds and advertising data sync, and whether late-arriving costs revise earlier periods. Then test the drill-down: a $2,400 advertising expense should open to the campaigns and products behind it; a product margin should reconcile to revenue, COGS, returns and ad spend. Product and collection profitability is more useful than a revenue-only dashboard, and customer-level profitability can reveal whether repeat buyers offset expensive acquisition.

Finally, test scenario controls without confusing them with forecasts. You should be able to change an assumption—such as ad spend, selling price or return rate—and see which line changes, while keeping the reported actual separate. A scenario is arithmetic, not a promise that demand will respond. Confirm that the tool can show contribution after advertising and margins under the revised assumption, and that its actual result remains auditable in product profitability reporting.

Turn the Result Into a Decision

Use the structure to set a limit before you buy more traffic. If an order produces $50 of realized revenue, carries $18 of COGS and $12 of other variable selling costs, it contributes $20 before advertising. To preserve $8 for operating costs and profit, the acquisition cost ceiling is $12 per order. That is a decision limit, not a promise that spending $12 will produce the same order volume.

Use product and collection contribution to find pricing and assortment actions. A product selling for $40 with $16 of COGS, $10 of variable selling costs and $9 of advertising contributes $5. A price increase may improve that contribution, but the calculation does not estimate how demand will respond; validate the change with a controlled test. For products with weak contribution, check returns, discounting, fulfilment costs and product profitability before simply cutting ad spend.

For channel decisions, compare the cost structure of the sales generated with the spend that generated it. MER divides total advertising spend by realized store revenue; attributed ROAS assigns revenue to particular campaigns. They answer different questions, so neither ratio alone proves that a channel is profitable. Keep the channel when its contribution supports the required margin, and investigate when attribution looks strong but total economics deteriorate.

Finally, use contribution to test fixed-cost capacity. If each order contributes $8 after advertising and monthly fixed costs are $24,000, the business needs 3,000 such orders to cover them. This is break-even arithmetic, not a complete forecast: order volume, price response, returns and mix can change the inputs. Review dead stock cost separately, because inventory tying up cash is waste even when the overall ratio appears acceptable.

A useful ecommerce cost structure is more than a percentage of sales: it is a consistent map of where each revenue dollar goes. Calculate it on comparable data, inspect the major cost layers, and use the result to change a specific pricing, marketing, fulfilment or operating decision.