Most people run a GA4 audit tool, get a clean report, and stop looking. That is the mistake. A green checkmark from an automated checker means your tags fired in a way the tool recognized. It says nothing about whether the revenue number sitting in your GA4 property this morning matches what your store actually made last night.
That distinction is not pedantic. It is the entire reason ecommerce teams keep finding six-figure reporting gaps in properties that technically "pass" every automated check they have ever run.
What a GA4 audit tool is actually built to check
Every automated GA4 audit tool on the market, GA4Check, GA4 Auditor, Elevar's Shopify-specific version, Conversios, runs the same basic category of check: presence and syntax.
Did the GA4 tag load. Did the recommended ecommerce events fire, view_item, add_to_cart, begin_checkout, purchase, in the sequence Google's own Purchase journey report is built around. Did Consent Mode get installed. Do key events carry the parameters Google expects, currency and value on a purchase, for instance.
Google's own troubleshooting documentation for GA4 tag setup describes exactly this scope: confirming a tag is sending hits and that those hits contain the expected structure. GA4's DebugView and GTM's Preview and debug mode exist to answer "is this firing correctly," live, hit by hit. That is a genuinely useful question. It is also a narrower question than most people assume they are asking when they run a GA4 audit tool.
None of these tools, free or paid, open your store's order export and compare it to what GA4 reported for the same date range. None of them know what your actual purchase volume was last Tuesday. They check the pipe, not the water running through it.
Three things that pass every automated check and are still wrong
Double-counting that looks like two correct events
A purchase event firing from a manually installed GTM tag and again from a platform's native sales channel integration will each individually pass a presence check. The checker sees two well-formed purchase events with valid parameters and reports nothing wrong, because nothing is wrong at the level it inspects. Your reported revenue, meanwhile, is roughly double what the store actually made.
This is one of the more common patterns we see in delivered tracking audits: a native platform pixel re-running the same tags inside a sandboxed iframe alongside an existing GTM container, each stream sending its own page_view and purchase hit to the same GA4 property under two different visitor identifiers. Every individual tag passes an automated check. The property as a whole does not reflect reality.
Vanity events wearing a key-event label
GA4 lets anyone mark any event as a key event (the GA4 term for what used to be called a conversion) directly in the Admin UI. An automated checker confirms the key event fires. It has no way of knowing whether the event behind that label is actually a purchase, or whether it is session_start or user_engagement, two events GA4 fires automatically on nearly every visit, mislabeled as the business's primary conversion. We have seen a property reporting roughly 526,000 "conversions" in 90 days against $0 in tracked revenue for exactly this reason, while every one of the account's real Google Ads conversion actions read zero. A tool would have shown that property as healthy.
Consent-driven undercounting that only shows up in the numbers, not the config
A checker confirms Consent Mode is technically installed. It cannot tell you how much traffic that consent setup is quietly dropping, because the gap only shows up when you compare GA4's totals against an independent source of truth. Orbit Media's research across 33 GA4 accounts found GA4 underreporting traffic by an average of 11% on sites without a cookie consent banner, and 20% on sites that had one, with individual accounts ranging as high as 30%. Every one of those accounts almost certainly had Consent Mode "installed" in the sense an automated tool checks for. The underreporting still happened.
Where the two approaches actually fit together
Run the free tool first. It is fast, it costs nothing, and it will legitimately catch dumb failures, a tag that never deployed, Consent Mode that was never turned on, an event with a missing required parameter. There is no reason to skip that step.
Treat its "all clear" result as clearance for the easy stuff only. The harder question, whether the revenue your GA4 property reports actually matches what your store made, requires pulling the real GTM container and GA4 property and checking the numbers against a source of truth your business already trusts: Shopify's own order export, your payment processor, your finance team's revenue figure.
That reconciliation step is most of what a manual GA4 audit checklist walks through beyond the basics any tool can see, and it is the same gap a GTM container audit closes on the tag-management side: finding the duplicate tags and orphaned triggers that an automated checker has no way to flag because, individually, each one still fires.
What to do with a clean automated report
A clean result from GA4Check, GA4 Auditor, or any Shopify-specific checker should raise one question, not close it: does the number match reality.
- Run a free checker first and fix anything it flags. It is a legitimate first filter, not a wasted step.
- Pull your store's order total for a fixed date range and compare it directly against GA4's reported purchase revenue for the same window. Any gap over a few percentage points needs an explanation, not a shrug.
- Check whether more than one source could be sending the same event. Two tag sources, a legacy pixel left active alongside a newer one, a platform's native integration running next to a custom GTM setup.
- Confirm every key event in Admin is actually the business event it claims to be, not a default GA4 event relabeled.
- If any of the above turns up a gap you cannot explain yourself, that is the point to bring in a fixed-scope GA4 and GTM tracking audit rather than a second automated scan. A deeper look is a day or two of work, not a rebuild, and it is built to answer exactly the reconciliation question no checker can.
If you want someone to run that reconciliation for you at no cost first, Anlyto's free analytics audit pulls your real GA4 and GTM setup, checks it against your actual numbers, and sends back the three highest-impact fixes ranked by revenue effect, the step every automated tool stops just short of.
