Profit Insight: From Formula to Decision

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

What Profit Insight Actually Measures

Profit insight is a decision-ready view of what a store actually keeps after realized revenue and the costs relevant to a specific decision. It starts with delivered orders, not every order created in Shopify. Cancelled orders and returned orders should not inflate revenue, while refund value and return behaviour need to remain visible because they reduce what the sale ultimately contributes.

Its scope is broader than revenue minus product cost. Depending on the question, it can include cost of goods, refunds, advertising and other costs that the business has chosen to attribute to the sale. That lets an operator ask a useful question: after the costs attached to this product, collection, customer or acquisition channel, is there enough left to justify scaling it?

The same view can be expressed through gross profit, net profit, contribution after advertising, and their margins. Gross margin focuses on revenue after cost of goods; net margin includes the broader cost scope used for the calculation. Gross margin and net margin answer different questions, so the label and included costs must be explicit.

Profit insight is not automatically the same as accounting net profit. Accounting net profit follows the business’s accounting policy and may include overhead, payroll, depreciation, interest, taxes, accruals and other ledger entries. A decision view may isolate product economics or advertising performance instead. Neither figure is inherently wrong; they serve different decisions. ROAS versus profit explains why an advertising return can look positive while the underlying profit is not.

The Formula, Stated Plainly

Start with realized revenue: the value of delivered orders that remain revenue after cancellations and returns are excluded. Keep discounts, refunds, taxes, and shipping treatment consistent with the question you are answering. The important rule is that the revenue denominator must match the costs and period being measured.

Contribution profit measures what remains after costs that move with the sales activity. The core formula is: Contribution profit = realized revenue − cost of goods sold (COGS) − variable fulfillment costs − payment fees − return-related costs − advertising spend. COGS is the product cost for the units sold. Variable fulfillment includes pick, pack, postage, and other per-order delivery costs. Payment fees, return shipping, and ad spend belong here when they are tied to the orders or acquisition activity being assessed.

You can also show contribution before advertising: realized revenue − COGS − variable fulfillment − payment fees − return costs. Subtracting advertising produces contribution after advertising, which is useful for judging whether acquisition activity leaves money after the costs required to generate it. Contribution margin is the percentage version: contribution profit ÷ realized revenue × 100. A margin of 20% means each $1 of realized revenue leaves $0.20 after the included costs. See the ecommerce contribution margin formula for the distinction between contribution and other profit measures.

Fixed costs—such as salaried staff, rent, software subscriptions, and recurring professional fees—should be included when the question is whether the business or period covers its operating overhead. In that case: net profit = contribution profit after advertising − fixed operating costs. For a short-term decision, exclude fixed costs that will not change if you take or reject the action. Do not force an arbitrary share of those costs into every product unless the allocation method is explicit; it can make a profitable incremental sale appear unprofitable. The appropriate profit insight therefore depends on the decision: contribution for variable economics, net profit for full-period sustainability.

Build the Number From Reliable Inputs

Build the number from records that share one period, currency and scope. Start with the order population: choose an order date or delivery date, then include only orders that belong to that period under the same rule. Gross item sales are reduced by discounts, cancellations and refunds. Returns need a clear treatment: a returned order should not remain in realized revenue, while the associated product cost and fulfilment treatment must follow the same return rule. A return-rate view can help reconcile unit and refund-value differences; see how returns reach profit.

Next, attach costs to the activity that generated the revenue. Use the product cost model for the units in delivered orders, rather than applying one store-wide cost percentage when product costs differ. Add fulfilment costs for those orders, payment fees on the transactions included, and advertising spend for the same marketing period and scope. Include other variable costs—such as transaction-linked packaging or marketplace charges—only when they are both attributable and consistently recorded. Fixed salaries, rent and software belong in a different profitability view unless the decision specifically requires them.

Reconcile each input before calculating. Sales and refunds should agree with the order system; product cost should tie to the quantities and SKUs sold; fulfilment and payment totals should use the same order population; advertising should be labelled as total spend or attributed spend, not treated as interchangeable. MER divides total store revenue by total advertising spend, while attributed ROAS uses attributed revenue, so the two should not be blended; ROAS versus profit explains why.

A practical input schedule looks like this:

A Worked Example With Real Numbers

Take one delivered order with $120 of realized revenue. The customer received the order, and there was no cancellation or return. The order-level costs are $42 for the product, $8 for shipping, $3.60 in payment fees, $6 for packaging and fulfilment, and $18 in advertising. The arithmetic is shown below.

Before fixed costs, the order produces $42.40: $120.00 − $77.60. Its profit margin is 35.3%, calculated as $42.40 ÷ $120.00 × 100. This is the useful view when deciding whether the order or its associated advertising generates enough contribution to justify further variable spend. It is not yet the store’s fully loaded net profit.

Now allocate fixed costs. Suppose the store has $1,200 of monthly software, rent and salaried payroll costs, and 100 delivered orders in the month. A simple allocation assigns $12 to each order: $1,200 ÷ 100. The order’s fully loaded result becomes $30.40, or a 25.3% margin. The allocation does not change the cash cost of fulfilling this particular order; it changes how much of the store’s fixed cost the order is made to carry.

The right version depends on the decision. Use the pre-allocation figure to assess product and advertising economics at order level. Use the allocated figure to assess whether the current order volume and pricing can support the wider business. For a broader treatment of the distinction between gross and net views, see gross margin vs net margin in ecommerce.

The Mistakes That Make the Number Wrong

A profit number can be mathematically correct and still answer the wrong question when its inputs are incomplete. The most common omission is returns: counting the original sale while leaving out the refund, return shipping, processing fee, and inventory impact makes a returned order look profitable. A 20% return rate is not a 20% reduction in revenue if the returned products also cost $40 to ship and inspect. Separate unit returns from refund value, as explained in how returns reach profit.

Tax and shipping need a consistent treatment. Sales tax collected for a tax authority is generally not store revenue, while a shipping charge paid by the customer may be revenue and the carrier charge is a cost. Excluding customer-paid shipping but including the carrier bill understates profit; including tax as revenue without the related obligation overstates it. State the treatment before comparing periods.

Advertising creates a separate attribution trap. Platform-attributed revenue is not the same as total store revenue, so adding attributed ROAS to a store-wide margin can assign the same sale to multiple channels. Use attributed ROAS to assess channel reporting and MER to compare total advertising spend with total store revenue; ROAS versus profit explains why they are not interchangeable.

Blended averages hide the decision. A $30 average contribution can combine one product earning $50 with another losing $10. Product and collection-level results expose that mix. Timing can distort the result too: orders recorded in June, ad spend paid in June, and refunds processed in July do not belong to the same view unless the period rule is explicit. Finally, check for double counting. Do not subtract landed product cost once inside cost of goods and again as a separate fulfillment cost, or subtract ad spend both in a channel result and again in the total. A reconciliation to order revenue, cost categories, and the chosen period catches these errors before the number drives action.

What to do with the answer

Use the answer to choose the next question, not to label the month as simply good or bad. If net profit is positive, identify which products, collections, customers and channels created it. If it is negative, locate the gap: product cost, fulfilment, returns, advertising or overhead. Gross profit and net profit support different decisions, so use this gross margin vs net margin comparison before changing pricing or assortment.

For marketing decisions, compare contribution after advertising with the cost of acquiring the sales. A campaign can show a 3.0 ROAS on $10,000 of attributed revenue from $3,333 in ad spend, yet leave little contribution after product, fulfilment and returns. Review ROAS vs profit and compare attributed ROAS with MER, which relates total store revenue to total advertising spend rather than assigning revenue to specific ads.

Then move from the store total to the level where an action is possible. Product and collection profitability can show what to scale, reduce, test or stop; customer profitability can show whether repeat behaviour improves the economics. Return rate by units and refund value can identify revenue that requires investigation. Inventory position, dead stock and remaining stock value add the cash and assortment context to a weak product result.

The answer does not prove that one change caused the result, or that repeating a profitable period will produce the same outcome. It cannot estimate price elasticity or predict how demand will respond to a price change. Treat any recommended action as a prioritised reading of the store's own numbers, with a stated reason and confidence—not as an instruction to change prices, bids, stock or customer communications. For the mechanics behind the calculation, see how to calculate ecommerce profit.

General information only: tax, duty and import rules change and depend on your situation, so check with an accountant or the official source before acting.