Conversion data drives budget decisions, and it is the data most likely to be wrong. Duplicated purchases flatter revenue, broken chains starve channels of credit, and both produce charts that look normal. Eight checks to verify every conversion counts once, done by hand or automatically in two minutes.
Tracking Auditor scores conversion integrity across your live GA4 and GTM: duplicates, missing links, value quality. Your first audit is free.
Run a free audit →Start from the business, not the tool: purchases, leads, sign-ups, quote requests. Compare that list against GA4's key events. Gaps mean invisible conversions. Extras, like page_view or scroll marked as key events, mean inflated ones. Both corrupt any bidding fed by the data.
Perform the real action and follow it through the chain: user action → dataLayer push → GTM trigger → tag fire → GA4 event → key event in reports. Use GTM Preview and GA4 DebugView side by side. A break at any link means under-counting, with no error anywhere.
Refresh the thank-you page: does the purchase fire again? Without a transaction ID, GA4 can't de-duplicate, and every refresh, back-button and email-link revisit inflates revenue. This is the single most common conversion finding on ecommerce sites.
A purchase event that fires with no value, the wrong currency, or net-vs-gross inconsistency 'works' in every debugging tool and still corrupts ROAS, LTV and any value-based bidding downstream. Check value, currency, tax and shipping handling against what the order system records.
Pull the same date range from GA4 and from the order system or CRM. Within ~5–10% is normal (consent opt-outs, ad blockers). Beyond that is a tracking problem: duplicates if GA4 is high, a broken chain or consent misconfiguration if GA4 is low. The direction of the gap is the diagnosis.
If conversions import from GA4, or fire via separate Ads tags, every upstream error becomes a bidding error with budget attached. Check which conversion actions Ads actually optimises toward, whether they're deduplicated against each other, and whether Enhanced Conversions is configured or just enabled.
Payment redirects that skip the thank-you page (PayPal, Klarna, 3DS challenges), purchases completed in another tab, single-page-app checkouts that never 'navigate'. Conversions lost at the edges look like a conversion-rate problem and are actually a measurement problem.
Redirects that strip UTM parameters, consent setups that block attribution cookies on the landing page, and cross-domain checkout flows without linking all funnel conversions into 'direct'. The conversion counts, but the channel that earned it gets nothing, and budget follows the misattribution.
No transaction-ID de-duplication, so reported revenue runs 10-30% above actual, and nobody questions good news.
The checkout changed, the dataLayer push moved, the trigger never fired again. Discovered months later.
session_start marked as a conversion: reports full of 'conversions' that mean nothing and Ads bidding toward them.
The purchase fires, the value is 0 or USD-labelled GBP. Every ROAS number downstream is fiction.
The redirect skips the thank-you page; a third of transactions never reach GA4.
A consent or redirect issue strips attribution. Channels look dead while conversions keep arriving.
Conversion integrity is one of five dimensions in a complete tracking audit. The full process is in the Google Analytics audit guide. Many conversion failures are really consent failures (consent mode audit) or container failures (GTM audit) upstream. For the condensed to-do-list version of everything, use the GA4 audit checklist. Debugging a specific failure right now? Start with GA4 purchase event not firing.
A conversion tracking audit verifies that every conversion a business relies on (purchases, leads, sign-ups) is captured once, with the right value, attributed to the right source, and consistent with the order system or CRM. It traces each conversion end to end and reconciles the totals against a source of truth.
Compare a full week or month of GA4 conversions against your order system or CRM for the same window. Within ~5–10% is normal. A larger gap is a tracking defect: GA4 running high usually means duplicates; running low usually means a broken tag chain, consent blocking, or funnel edge cases like payment redirects.
Almost always duplicate purchase events: the thank-you page fires the event on every load and there's no transaction ID to de-duplicate refreshes and revisits. Check whether reloading the confirmation page fires another purchase in GA4 DebugView.
They measure differently: attribution models, conversion windows, and view-through conversions all differ. Some disagreement is expected; a 2x disagreement means one side's tracking is broken, and an audit that traces both chains finds which.
Yes. Tracking Auditor cross-checks GA4 key events against the GTM tags meant to send them, detects duplicate-purchase risk and missing transaction IDs, and scores conversion integrity as one of five dimensions in a full tracking audit.
Connect GA4 and GTM read-only and Tracking Auditor checks every key event has a matching tag, flags duplicate-purchase risk, and scores conversion integrity alongside consent, governance, event quality and cookies. Your first audit is free, no card required.
Run your free audit