Why Your Meta ROAS Doesn't Match Shopify Revenue

If Meta says your campaigns generated one revenue number while Shopify shows another, that does not automatically mean your Pixel is broken.

Meta and Shopify answer different questions. Shopify records transactions in your store. Meta estimates which conversions should receive credit for exposure to or interaction with your ads. Because of that, identical numbers are not the goal.

Why Meta and Shopify should not match exactly

Some discrepancy can come from normal reporting differences, including:

• attribution windows
• click-through versus view-through attribution
• cross-device customer journeys
• returning customers
• timezone and reporting-period differences
• modeled conversions or platform-specific reporting logic

The objective is not to force the dashboards to agree. It is to understand what each number represents and whether the gap is reasonable for the decision you are making.

When the discrepancy becomes suspicious

I pay closer attention when the gap is large, unstable or inconsistent with the underlying orders in Shopify. Patterns worth validating include:

• purchase events appearing more than once for a single order
• more than one Meta or Google implementation path active at the same time
• browser and server events that may not be deduplicating correctly
• missing or inconsistent transaction value, currency or event identifiers
• GA4 telling a materially different story from both Shopify and Meta
• subscription renewals or returning-customer revenue being interpreted as new acquisition revenue

None of those observations proves a tracking defect by itself. They tell you where to investigate.

If the gap is in purchase count—not just revenue—start with the conversion signal itself. One Shopify order should not become multiple independent Purchase events in Meta.

If GA4 is also reporting a different transaction or revenue total, use the GA4 versus Shopify reconciliation guide to separate collection issues from reporting definitions.

Five things I would validate first

1. Start with the actual Shopify transactions

Confirm the orders, revenue, refunds and time period being analyzed. Before interpreting attribution, establish what actually happened in the store.

2. Compare purchase counts before comparing revenue

If the number of reported purchases is already inconsistent, investigate that first. Revenue differences are harder to interpret when conversion counts are not stable.

3. Map every active implementation path

Identify how Meta and Google measurement are being loaded: Shopify integrations, apps, theme code, tag managers or older implementations. Seeing multiple tags does not automatically mean duplicate conversions; the real question is what each customer action actually sends.

4. Validate the Purchase signal

Check whether a real order sends the expected value, currency and identifiers, and whether the event occurs at the correct point in the transaction.

5. Review browser/server deduplication before blaming CAPI

Using both browser tracking and Meta Conversions API can improve signal resilience, but the two versions of the same purchase need to be recognizable as the same conversion. Simply seeing both browser and server events is not evidence of duplication.

The discrepancy is a symptom—not the diagnosis

A gap between Meta and Shopify tells you where to investigate; it does not tell you what is broken. Before changing code, separate the discrepancy into three categories: attribution rules, implementation behavior and business definitions.

• If purchase counts diverge, validate the conversion signal before debating ROAS.
• If purchase counts are close but revenue differs, inspect values, refunds, reporting periods and which orders each platform is crediting.
• If the implementation is clean but attribution still differs, the dashboards may simply be answering different business questions.

The goal of a forensic review is not to force the dashboards to match. It is to identify which difference is expected, which one is actionable, and whether it is large enough to change a budget decision.

What not to do

• Do not try to make Meta, GA4 and Shopify report exactly the same revenue.
• Do not change tracking simply because two dashboards disagree.
• Do not assume two visible tags automatically mean purchase duplication.
• Do not assume the absence of CAPI is the explanation for every attribution gap.

First establish whether the difference comes from measurement, attribution or business definitions. Then fix the part that is actually affecting the decision.