Skip to main content
FHAVI

Research · Tracking

GA4 purchase revenue does not match Shopify — the five places it breaks

GA4 and the store disagree on revenue by a consistent margin. Five specific causes, in the order they usually appear, and how to identify which one you have.

2 min read

In brief

  • A stable percentage difference is usually tax, shipping or discounts being included on one side and not the other.
  • A step change on one date is a deploy, a currency setting, or a duplicated event — trace it by date, not by average.
  • GA4 revenue and store revenue are not supposed to be identical; they should be explainable by one line of arithmetic.

The five breaks

  1. 1

    Tax and shipping in or out

    The purchase event's value may exclude tax and shipping while the store reports order total including them, or the reverse. Choose one convention and apply it in the event.

  2. 2

    Discounts applied after the event fires

    If value is sent before a coupon is applied, every discounted order is overstated in GA4 by exactly the discount.

  3. 3

    Currency

    GA4 expects the currency parameter with the value. A store selling in one currency and reporting in another produces a systematic difference that looks like a bug and is a mapping.

  4. 4

    Duplicate purchases

    Two events for one order inflates GA4 while the store records one. The tell is a ratio above 1.0 rather than below.

  5. 5

    Refunds and cancellations

    GA4 keeps the purchase until a refund event is sent. A store reporting net revenue and a platform reporting gross will differ by exactly the refunded amount.

Diagnosing in twenty minutes

  • Take one order and find its purchase event. Compare value, currency and transaction id with the order record.
  • Multiply the difference by the number of orders: if it equals total discounts or total shipping, you have found the convention mismatch.
  • Check whether refunds are being sent as refund events at all.
  • Compare a week before and after any checkout deploy. A step change has a date, and the date is the answer.

What 'matching' should mean

Expect a small difference from session attribution and unconsented traffic. What you should not accept is a difference you cannot state in one sentence. 'GA4 runs about 8% below the store because it excludes shipping and we report shipping in the order total' is a working measurement system.

The two numbers are not meant to be equal. They are meant to be reconcilable in one line — and until they are, you are optimising on a figure you cannot defend.

Questions

Asked about this directly

Should I send tax and shipping in the purchase value?
Pick the convention your business reports on and apply it everywhere, then document it. Consistency matters more than which choice you make, because consistency is what makes the difference explainable.
Why is GA4 lower than Shopify most of the time?
Refunds not sent, consent-blocked sessions, ad blockers, and checkout steps that leave the site. Each removes revenue from GA4 without touching the store record.
Do I need server-side GA4?
Not usually. The purchase is normally within the store's own domain, so the browser is reliable there. Fix the value and currency first.

Keep reading

Other pieces in the library

All research

Numbers first, promises second

Send your store and we will tell you which of these failure modes is already happening — before you pay for a fix.

Prefer email? Write to md@fhavi.com