GA4 revenue reports can shape decisions on paid media, SEO investment and holiday promotions. But a recent ecommerce audit suggests that, for some stores, browser-based analytics may be a poor substitute for the order system when traffic and sales volumes rise.
In one example cited in Christian Goodrich's Search Engine Journal analysis, GA4 showed a 68% year-over-year revenue decline from Christmas Day 2025 through January 4, 2026. The retailer's store records showed a 13% decline over the same period. The gap does not establish that GA4 will undercount every ecommerce site. It does show why teams should reconcile analytics with their transactional data before cutting campaigns or declaring a channel underperforming.
What the audit found
The analysis examined GA4 order tracking across eight UK ecommerce brands during the 2024-2025 holiday period. Six brands could be matched against store-order records. The other two had issues that prevented a clean comparison, including cookie-banner behaviour masking or misattribution, delayed backfills, double counting, or spikes in bot traffic.
For one retailer, GA4 reportedly captured 81% of web orders in Q4 2024. That figure fell to 41% in Q4 2025 and 32% across the Christmas period. These are observations from a small, agency-connected set of brands, not a market-wide measurement of GA4 accuracy. Goodrich also discloses that his digital agency has served all eight brands, an important consideration when assessing the findings.
Still, the underlying issue is familiar to ecommerce analysts: a completed order in a commerce platform and a purchase event received by GA4 are separate records. A buyer can successfully place an order even if their browser does not send, or GA4 does not process, the associated purchase event.
| Period in the reported retailer example | GA4 share of web orders captured |
|---|---|
| Q4 2024 | 81% |
| Q4 2025 | 41% |
| Christmas period | 32% |
The table reflects a case reported in the audit, rather than a benchmark that other retailers should expect.
Why GA4 and store orders can diverge
GA4 purchase measurement commonly depends on a client-side event firing at, or after, order confirmation. That flow can be interrupted by consent choices, implementation changes, browsers, ad blockers, page failures or customer behaviour. Peak trading adds more moving parts, including promotional landing pages, checkout changes and heavier traffic.
The analysis identifies five patterns worth investigating:
- A stable shortfall can point to consent choices or cookie-banner behaviour that limits analytics collection.
- A sudden loss of tracking after a banner change may mean consent settings are preventing purchase events from being sent.
- Blackouts followed by late data can make day-to-day reporting misleading if purchases arrive after the period being analysed.
- Duplicate transaction IDs or faulty deduplication can produce the opposite problem, where GA4 exceeds actual store orders.
- Non-human traffic can inflate sessions and distort conversion-rate analysis, even when order totals are correct.
These problems have different consequences. Under-recorded purchases can make a profitable campaign appear unprofitable. Duplicate purchases can overstate performance. Bot-driven sessions can make conversion rates look weak and lead a team to blame landing pages or search traffic incorrectly.
Reconcile purchases with the order system
The store platform, order management system or payment records should be the primary reference for completed sales. GA4 remains valuable for analysing acquisition, on-site behaviour and attribution, but its purchase total needs validation before it becomes a budget-setting number.
A practical weekly check during the lead-up to peak season is straightforward:
- Export store orders for matching weeks in the current and prior year.
- Export GA4 purchases using distinct transaction IDs, rather than relying only on a headline revenue card.
- Compare GA4 purchase counts with store-order counts for the same date range and scope.
- Calculate the capture rate: GA4 purchases divided by store orders.
- Investigate abrupt changes, totals above 100%, and differences that widen around a release, consent update or promotion.
The date and order scope must match. Teams should agree how they treat cancellations, refunds, test orders, phone orders, marketplace sales and purchases from apps or other non-web channels. Otherwise, a legitimate scope mismatch can resemble a tracking failure.
Google Search Console can provide an additional warning signal. If organic clicks and impressions remain relatively stable while GA4's reported organic purchases collapse, the discrepancy deserves investigation. Search Console cannot confirm revenue, and stable search demand does not prove that conversion was unchanged. It can, however, help distinguish a possible measurement break from a clear fall in search visibility.
Why the gap affects search and marketing decisions
Ecommerce teams often use GA4 revenue and conversions to evaluate SEO, paid search, shopping campaigns and content. When purchase tracking is incomplete, attribution reports can under-credit the channels that introduced or assisted a sale.
That creates a particular risk during holidays. A retailer may see weaker GA4 revenue, pause a campaign, reduce search investment or change a promotion even though the underlying order system shows a much smaller decline. Conversely, a duplicate-event issue can encourage overspending on a channel whose apparent return is inflated.
The right response is not to abandon GA4. It is to establish a reporting hierarchy. Financial and operational reporting should reconcile to transaction records. GA4 should be assessed for its coverage and used with that known limitation in mind. This also gives agencies and internal teams a shared basis for discussing whether a reported performance change is commercial, technical or both.
Consider more resilient purchase collection
The analysis recommends server-side purchase tracking through the Measurement Protocol or server-side tagging. In principle, connecting a confirmed order to analytics from the server reduces reliance on a browser successfully loading and sending a confirmation-page event.
That does not remove the need for quality assurance. Server-side implementations require careful consent handling, event design and transaction-ID controls. They can also create duplicates if browser and server events are not coordinated. The goal is not simply to send more data, but to create a controlled, auditable connection between a confirmed transaction and analytics reporting.
Before a major trading period, test the full purchase path under the consent scenarios your site supports. Confirm that transaction IDs are unique, orders are counted once, and reporting delays are understood. Recheck after checkout releases, tag-management changes and cookie-banner updates.
If GA4 is central to how you assess search and AI-discovery performance, create a GEO Search Leads account to begin assessing your website's GEO readiness and identify optimisation opportunities. Reliable analytics reconciliation gives those visibility assessments a more useful business context, helping teams separate changes in discovery from gaps in how conversions are recorded.
Frequently Asked Questions
Does this audit prove that GA4 misses most ecommerce orders?
No. The analysis reports findings from eight UK ecommerce brands, with six brands suitable for order matching. Its examples show that significant gaps can occur, but they do not establish a universal GA4 undercount rate.
Which number should be used for revenue reporting?
Completed orders and revenue from the store platform, order system or financial records should be the primary reference for sales reporting. GA4 data should be reconciled against those records before it is used for major budget decisions.
Why can GA4 purchase data be lower than store orders?
Consent and cookie-banner behaviour, blocked or failed browser events, reporting delays and implementation issues can stop a purchase event from reaching or being counted in GA4. A successful order does not guarantee a corresponding analytics event.
Can GA4 also overcount purchases?
Yes. The reported analysis identifies transaction-ID deduplication problems as one potential source of double counting. Comparing distinct GA4 transaction IDs with store orders can help uncover that issue.
Conclusion
The reported examples are a reminder that analytics data is not automatically transaction truth. Ecommerce teams should check GA4 purchase coverage against store records before peak season, then investigate changes in that relationship as carefully as changes in traffic or revenue. That discipline can prevent measurement errors from becoming costly marketing decisions.
Top comments (0)