ROAS is a revenue ratio. Check contribution before scaling spend.

Choose a reporting window and state the attribution model and conversion window. Decide whether revenue is before or after discounts, tax, shipping charges, cancellations, and refunds. A platform-reported purchase value may not equal the amount retained by the business. Keep both values if you need to reconcile them.

Record each order once in the economic view, even if several platforms claim it. Joining daily campaign rows to every order in a day multiplies spend. Aggregate at compatible grains and inspect row counts before summing. The spreadsheet merge guide explains why a correct-looking join can still duplicate money.

Suppose a campaign claims $10,000 of net revenue and cost $2,000. Reported ROAS is 5.0. Now subtract $5,000 of product costs, $1,000 of fulfillment, and $400 of payment and other variable costs. Contribution before advertising is $3,600. After the $2,000 ad spend, contribution is $1,600, before shared overhead.

If another $2,000 of returns reduces retained revenue and the cost assumptions are otherwise unchanged, contribution after advertising becomes negative $400. In a real return, some inventory or fulfillment costs may be recoverable, so model the actual treatment rather than assuming all costs remain. The point is to reconcile the economics, not to force one pessimistic scenario.

In the first scenario, pre-ad contribution margin is $3,600/$10,000 = 36%. With a constant 36% margin, one dollar of revenue contributes $0.36 toward advertising. The corresponding break-even ROAS for contribution after ads is 1/0.36, or about 2.78. This calculation covers only the included variable costs and assumes the margin is stable as sales grow.

It is not a universal target. Changes in product mix, discounts, shipping zones, repeat buying, or return rates change the margin. Fixed overhead and the timing of cash payments also affect the business even though they are outside this particular break-even calculation. Label those exclusions beside the number.

A campaign's attributed orders are orders assigned by a tracking rule. They are not necessarily orders caused by the campaign. Returning customers, brand demand, and overlapping channels can make an attributed result look better than the incremental effect. A revenue bridge or spreadsheet alone does not identify what would have happened without the ads.

Use a suitable experiment or other credible causal design when the decision requires incremental return. Meanwhile, compare observed contribution under consistent definitions and flag missing costs. Do not call an attribution comparison a lift study simply because one campaign's ratio is higher.

Keep the order economics and platform spend totals in separate source tabs. Use SUMIFS or a pivot table to summarize reconciled values by campaign and period. Show orders with missing costs rather than treating their margin as 100%. Review the highest-spend and highest-return exceptions before distributing the dashboard.

The Shopify profitability example shows the value of connecting a conclusion to source evidence. Start a free analysis with an export to inspect its findings and limitations. A useful marketing review tells you what the included orders contributed and which assumptions could reverse the conclusion.