GA4 audit: find out how much revenue your analytics is missing

A DTC store came to me because its GA4 purchase reporting did not reconcile with its Shopify order records. The historical case reported on this page describes purchase-event capture improving from 59% to 100%. Purchase counts and revenue amounts remain separate measures; the case is context for the diagnostic method, not a capture guarantee for another store.

The case involved a gap in that store's checkout measurement. It does not establish a universal failure of Shopify's native integration, Shop Pay or mobile checkout. For a new audit, I inspect the installed integration, checkout extensions, pixel configuration and permitted test journeys. Shopify documents that checkout_completed normally fires on the Thank you page, can fire earlier for post-purchase flows, and is absent if its triggering page never loads.

The reported 41% gap concerns purchase-event capture in that engagement. It is not a claim that 41% of revenue was lost or that another store has the same gap. An audit separates missing collection, permitted suppression, duplicates and reporting differences before interpreting campaign performance.

I implemented server-side purchase events and purchase-identifier deduplication for that engagement; the historical case reports 100% capture after the fix and a three-week delivery. Your implementation needs its own accepted comparison period and tests. Destination deduplication rules, consent, missing identifiers and delayed processing can limit what can be reconciled.

Start with the same eligible purchase cohort in the commerce or billing system and GA4. Check transaction identifiers, consent conditions, timing, refunds and duplicate rules before investigating the implementation. A discrepancy is a question to explain, not a universal five- or ten-percent pass/fail test.

Then I inspect the implementation with permitted tools such as Google Tag Assistant, DebugView and checkout network evidence. We agree staging or approved live test journeys in advance, including any transaction that creates a real record. A validation response, destination receipt and processed report are separate observations.

I document every finding with: what is wrong, where the break is, what a correct implementation looks like, and effort to fix (hours, not story points). You get a prioritized list you can hand to a developer or have me implement.

Examples I investigate include missing checkout purchase events, duplicate browser/server collection, missing purchase parameters, incorrect key-event configuration and attribution-setting differences. Their frequency and impact must be established for the reviewed implementation.

Agree the comparison period, transaction definition, currency and treatment of refunds before comparing totals. I use the permitted analytics and order records, documenting which checks were completed and which could not be observed. An absent analytics event is a measurement gap, not evidence that the sale was lost. Findings include the observed evidence and an agreed recheck.

A GA4 audit is often the entry point to a broader marketing attribution engagement. Once GA4 is accurate, you have a foundation for cross-channel attribution - reconciling GA4 data with Meta CAPI, Google Ads enhanced conversions, and your CRM.

If the audit finds a need for server-side implementation, a possible next step is a server-side tracking implementation. If you want a broader picture of all your measurement gaps at once, start with a tracking audit that covers all channels simultaneously.

Not sure GA4 is the problem yet? Run the free growth leak audit first: its public-surface observations can help frame a question, but do not establish private event delivery, destination receipt or revenue loss.