Meta ROAS divides revenue attributed to Meta ads by Meta spend. MER divides total business revenue by total marketing spend. Because the scopes differ, the ratios can differ too. Define each report and investigate unexplained discrepancies before using either one for a budget decision.

What each number actually measures

Meta ROAS depends on the attribution window and reporting settings. It records attributed revenue per dollar of Meta spend; it does not directly measure the sales that would have disappeared without those ads.

MER uses total store revenue and the marketing costs included in your definition. Read it alongside contribution margin to assess the overall economics. It cannot identify which ad or channel caused the revenue.

Shopify documents differences between its marketing reports and ad-platform reports, including attribution and data-sync timing. Check those differences when reconciling the numbers.

One month, with the numbers filled in

Say your store does $120,000 in revenue this month. You spent $30,000 on Meta, and about $6,000 more across Klaviyo and the rest of your marketing. This is an illustrative calculation, not a client result.

Meta reports a 4.2 ROAS on 7-day click, 1-day view. Multiply it out: 4.2 times $30,000 is $126,000 of Meta-attributed revenue. Your store did $120,000 in total. Meta is claiming more revenue than the entire business earned, before email or organic take credit for a single order.

Attribution overlap explains why adding channel-reported revenue can exceed store revenue. It does not, by itself, explain why one platform reports more than the store total. Reconcile event values, currencies, conversion definitions, refunds, duplication and reporting dates before accepting that discrepancy. A customer journey can include both a Meta interaction and an email interaction without either dashboard measuring the counterfactual sale.

Blended MER for the month is $120,000 divided by $36,000, which is 3.3.

The scope difference explains why the ratios need separate interpretations. Meta-attributed revenue exceeding total store revenue is an additional reconciliation problem.

What is ROAS actually for?

Inside Meta, attributed ROAS and cost per purchase are useful operational signals. Compare like-for-like objectives, attribution settings, time windows, delivery and customer mix. An ad with higher reported ROAS has a higher attributed return in that comparison; it has not automatically proved greater incremental value. Attribution bias need not be equal across ads.

In-platform ROAS does not include all the costs required to establish business profit.

What is MER actually for?

MER provides a business-level constraint for the budget conversation. Read it alongside contribution margin, product mix, returning-customer revenue and other changes in the business. It does not isolate a channel contribution or estimate the return on the next dollar.

There is no single MER number that is good for every store, but you can derive your own floor in one line. Take your contribution margin before marketing. If it is 40%, every dollar of revenue carries 40 cents to pay for marketing and profit, so a dollar of marketing spend needs to bring back at least $2.50 of revenue just to protect the margin. One divided by 0.40 is 2.5. Below an MER of about 2.5, this store's marketing is eating the profit. At a 30% margin the floor rises to about 3.3, and at 50% it drops to 2.0. Run the arithmetic with your own margin, because the floor moves with it.

At a 40% contribution margin before marketing, $120,000 revenue contributes $48,000. Subtract $36,000 marketing spend and $12,000 remains before any costs excluded from that margin definition. This supports a discussion about available headroom. It does not establish that extra spend will return at the current average.

Average return does not answer the next-dollar question

Suppose an additional $5,000 in spend causes $10,000 in additional revenue. The return on that increment is 2.0. At a 40% contribution margin before advertising, that revenue contributes $4,000 against $5,000 of added spend: a $1,000 shortfall before any further costs. The existing budget could still have a healthy average return.

This is an illustrative scenario with the causal effect assumed. Subtracting one week's revenue from another does not establish that effect. Promotions, seasonality, inventory and other marketing can all change the result. Marginal ROAS concerns return around an additional unit of spend; the finite increment above is a simple approximation. Google's Meridian documentation distinguishes average and marginal return through a modeled response curve.

Line up the dates and define MER before you trust either one

Use matching date ranges and inspect how each report assigns conversions to dates. An interaction-date view and an order-date view can place the same purchase in different periods.

Document which costs belong in MER: platform spend, management fees, email software, affiliate payouts and any other included marketing cost. Keep the definition consistent across periods. Use Shopify's analytics field definitions to confirm what the selected revenue field contains.

Read the platform trend alongside the store

A rising platform ROAS with falling MER warrants investigation. Attribution settings and the customer mix can affect ROAS, while costs and revenue outside Meta affect MER. Retargeting may receive credit for sales that would have happened anyway; estimating that effect requires an appropriate experiment or model.

Bring those trends into a second opinion with the underlying reports and definitions. Divergence identifies a question to resolve, rather than supplying the explanation by itself.

Which number do I optimize to?

Use MER and contribution to check business economics, attributed metrics to monitor delivery, and an appropriate experiment or model when the decision requires an incremental estimate. Read the proposed spend change against its downside and the evidence available. A good average return is not sufficient evidence to keep increasing spend.

Keep the definitions and unresolved discrepancies in the weekly report so the next budget discussion starts from the same numbers.