Analytics & Tracking

eCommerce Tracking Audit: A Checklist for Clean Revenue Data

Broken tracking quietly ruins ad bidding and reporting. Use this tracking audit checklist to find duplicate purchases, missing events and attribution leaks.

TL;DR
  • Start every audit by comparing tracked purchases and revenue with real store orders. The size of the gap tells you how urgent the rest is.
  • Most problems fall into five groups: missing or duplicate purchase events, broken funnel events, self-referrals, consent gaps and messy campaign tagging.
  • Audit after every theme change, checkout change, app install or new ad platform, not just once a year.

Why does a tracking audit matter for profit?

Ad platforms bid on the conversions you send them. GA4 reports and dashboards drive budget decisions. If purchase events are missing, duplicated or credited to the wrong source, you end up:

  • Scaling campaigns that look good only because of double-counted purchases.
  • Cutting channels that look weak because their conversions are being lost.
  • Arguing about numbers instead of making decisions.

A tracking audit is a structured check of whether your data reflects reality. It does not need to be complicated. Most issues show up in a few hours of focused checking.

Step 1: How big is the gap between tracked and real orders?

Pick a recent full week and compare:

Source Orders Revenue
Store backend (Shopify, WooCommerce) Your source of truth Your source of truth
GA4 purchases ? ?
Meta reported purchases (all attribution) ? ?
Google Ads conversions ? ?

Some difference between GA4 and the store is normal (ad blockers, consent choices, cross-device journeys). But a large gap, or more tracked purchases than real orders, means something is broken. The GA4 versus Shopify revenue gap explains the normal causes in detail.

Ad platform numbers will never match the store because they use their own attribution. Compare them with each other over time rather than expecting an exact match.

Step 2: Is the purchase event firing correctly?

The purchase event is the most important event you track. Check:

  • It fires once per order. Look for duplicate transaction IDs in GA4 (an exploration with transaction ID and purchase count shows this). Common causes are thank-you page reloads and two tracking setups running at the same time (for example a native integration and a Google Tag Manager tag).
  • It includes the right values. Transaction ID, value, currency, tax, shipping and items.
  • Value is consistent. Decide whether purchase value includes tax and shipping, and make sure GA4, Meta and Google Ads use the same definition.
  • It fires on all checkout paths. Express checkouts, PayPal, buy now pay later and upsell pages after checkout can skip or duplicate events.

On Shopify, checkout tracking runs through customer events and app pixels. Check that whatever you use is the only source of purchase events, or that duplicates are prevented.

Step 3: Are funnel events complete?

Beyond purchase, the standard eCommerce events tell you where shoppers drop off: view_item, add_to_cart, begin_checkout, add_shipping_info, add_payment_info, purchase. For each one:

  • Does it fire where expected?
  • Does it include item data (item ID, name, price)?
  • Do the item IDs match those in your product feed and purchase events?

Use your browser’s tag debugging tools and GA4 DebugView while you click through the site as a shopper. A clean data layer makes all of this much easier. See GA4 eCommerce tracking for the full event setup.

Step 4: Is credit going to the right source?

Three common leaks steal credit from your real marketing channels.

Self-referrals and payment gateways

If your own domain or payment providers (PayPal, Klarna, Shop Pay and others) show up as referral sources for purchases, sessions are being broken and credit is lost. Fix with GA4’s unwanted referrals list and correct domain configuration.

Cross-domain journeys

If shoppers move between domains (for example a main site and a separate checkout or booking domain), you need cross-domain tracking configured so GA4 keeps the same session.

Campaign tagging

Untagged email, SMS, social and influencer links end up as direct or referral traffic. Check that every owned and paid link uses consistent UTM parameters. Mixed naming (Email, email, e-mail) splits the same channel into several rows. A shared UTM naming convention fixes this.

In regions where you need consent for analytics and advertising cookies, check:

  • The consent banner appears and records choices.
  • Tags respect those choices. Nothing that requires consent fires before consent is given.
  • Consent mode is configured if you use Google tags, so Google can model conversions for users who decline. See Consent Mode v2.

For U.S.-focused stores, requirements vary by state, but checking that your banner and tags behave as intended is good practice everywhere.

Step 6: Is server-side or API tracking set up and deduplicated?

Browser tracking alone loses some events to ad blockers and browser restrictions. Many stores add server-side signals, such as Meta’s Conversions API or a server-side tagging setup.

The key check: when the same event is sent from the browser and the server, it must be deduplicated, usually with a shared event ID. Otherwise purchases are counted twice. For Meta, also check event match quality in Events Manager.

Worked example: finding double-counted purchases

(Illustrative numbers.) A store’s Shopify backend shows 1,000 orders in a week. GA4 shows 1,240 purchases.

The audit finds:

  1. GA4 has 180 duplicate transaction IDs. A thank-you page app fires a second purchase event when customers reload the page.
  2. 60 purchases come from a legacy GTM tag left over from an old theme, which fires in addition to the native integration.

After removing the legacy tag and adding duplicate protection, GA4 shows 940 purchases for the next comparable week, a gap of 6% versus Shopify, which is in the normal range. The paid social campaigns that looked best had been benefiting most from the duplicates, so the team resets their targets based on the corrected data.

The full audit checklist

  • [ ] Tracked purchases and revenue compared with store orders for a full week
  • [ ] No duplicate transaction IDs in GA4
  • [ ] Only one source sends purchase events to each platform (or duplicates are prevented)
  • [ ] Purchase events include transaction ID, value, currency and items
  • [ ] Purchase value definition (tax, shipping) is the same across GA4 and ad platforms
  • [ ] Express checkout, PayPal and post-purchase upsell paths tracked correctly
  • [ ] Funnel events fire with item data and IDs that match the product feed
  • [ ] Payment gateways and your own domain are in GA4’s unwanted referrals list
  • [ ] Cross-domain tracking set up where journeys span multiple domains
  • [ ] All email, SMS, social and creator links use consistent UTM parameters
  • [ ] Consent banner records choices and tags respect them
  • [ ] Browser and server events are deduplicated with a shared event ID
  • [ ] Meta event match quality and Google Ads conversion diagnostics checked
  • [ ] Audit date and findings documented, with a re-check scheduled

How often should you run a tracking audit?

Run a full audit at least twice a year and a quick check (Steps 1 and 2) monthly. Always re-audit after:

  • Theme changes or redesigns.
  • Checkout or payment provider changes.
  • New tracking apps, pixels or tag manager changes.
  • Platform migrations.

These are the moments when tracking breaks, and the longer a problem runs unnoticed, the more bad data ends up in your ad platforms’ bidding models.

FAQ

How close should GA4 purchases be to Shopify orders?

There is no fixed rule, but a gap of a few percent up to around 10% to 15% is common because of ad blockers, consent and cross-device journeys. A larger gap, or more GA4 purchases than real orders, points to a tracking problem.

What is the most common tracking mistake in eCommerce?

Duplicate purchase events. They usually come from two tracking setups running at once or from thank-you pages that fire on reload. They inflate results and mislead bidding.

Do I need server-side tracking to pass an audit?

Not necessarily. Server-side tracking can recover some lost events, but it adds complexity. A clean browser setup with correct deduplication is better than a messy hybrid setup.

Turn this into a plan for your store

The Free Profit Audit scans your store, sizes your biggest profit leaks and builds a 90-day plan.

Free Profit Audit