[ ← All guides ]Tracking · GA4 · Google Ads

root@ecombender:~$ cat guides/03-ga4-shopify-ads-numbers-dont-match

Why GA4, Google Ads and Shopify Show Different Sales (and Which to Trust)

Shopify says 100 orders. GA4 says 82. Google Ads says 140. Some of that gap is normal, and some of it means your tracking is broken. Here is how to tell the difference.

Reading time: 13 min · Tools: Shopify Analytics, GA4 Explore, Google Ads Goals

The short version

Each tool answers a different question

Shopify "How many orders?" Counts every paid order on its own server Not affected by ad blockers or consent Use for: revenue GA4 "What did people do?" Counts purchases it can see in the browser Misses some visitors who block or decline tracking Use for: behaviour Google Ads "Which ads helped?" Credits sales to the day of the ad click Includes modelled sales it estimates Use for: bidding
Figure 1: Three tools, three jobs. They were never meant to show the same number.

The four normal reasons numbers differ

1. Different dates for the same sale

Google Ads reports a conversion on the date of the click. Shopify and GA4 report it on the date of the purchase. If someone clicks your ad on October 30 and buys on November 2, Google Ads puts that sale in October.

Oct 30 Oct 31 Nov 1 Nov 2 Ad click Purchase Google Ads reports in October Shopify and GA4 report in November
Figure 2: One sale, two months. Compare longer date ranges, not single days.

2. Attribution windows

Google Ads can credit a sale to any ad click in the window you set, often 30 days and sometimes up to 90. If a shopper clicked an ad, then came back three weeks later from your email, both Google Ads and your email tool may claim the sale. Shopify counts it once.

3. Lost browser events

GA4 mostly relies on code running in the shopper's browser. Ad blockers, privacy settings and declined cookie banners stop some of it. That's why GA4 usually shows fewer purchases than Shopify. Google has been moving Shopify purchase data into GA4 server-to-server for stores using the Google & YouTube app, which closes much of this gap, so check your recent GA4 data after the change before assuming you're losing a lot.

4. Modelled conversions

When Google Ads can't observe a conversion directly, for example because of consent choices, it estimates some of them. Those estimated conversions show in Google Ads but not in Shopify.

What a normal gap looks like

There is no official standard, and sources disagree. Use these as rough ranges, comparing the same 30-day period by purchase date:

GA4 purchases as a % of Shopify orders 0% 70% 80% 100% 200% Tracking broken Worth checking Normal Double counting Under ~70%: missing tags, checkout events not firing, or consent blocking most visitors. Over 100%: the same order is sent more than once. Close to 200% almost always means two purchase tags.
Figure 3: Anything above 100% is a fault, not a rounding difference.

How to find double counting (10 minutes)

Duplicates usually happen when two things send the same purchase. The usual culprits are the Google & YouTube app and a manually added GA4 tag, Google Tag Manager setup or custom pixel.

  1. List every source of a purchase event. In Shopify, check Settings → Customer events for custom pixels, check if the Google & YouTube app is connected to GA4, and check your theme and Google Tag Manager for GA4 tags.
  2. Check for repeated order IDs in GA4. Go to Explore → Free form. Add the dimension Transaction ID and the metric Ecommerce purchases. Sort by purchases, highest first. Any transaction ID with 2 or more is a duplicate.
  3. Check for missing order IDs. If many rows show "(not set)" as the transaction ID, GA4 can't remove duplicates on its own. Every purchase event should send the Shopify order ID as transaction_id.
  4. Keep one source. If the Google & YouTube app already sends purchases, remove or pause the extra purchase tag. Don't run two and hope they cancel out.
Google Analytics › ExploreHomeReportsExploreAdvertisingAdminFree form explorationRows: Transaction ID · Values: Ecommerce purchases · sorted high to lowTRANSACTION IDECOMMERCE PURCHASES10432103921044110421(not set)14Red: counted twiceduplicate purchase tagsAmber: no order IDGA4 can't remove duplicates
Step 2 in practice: any transaction ID with 2 or more purchases is a duplicate. Illustration, not a real screenshot. Your screen may look different.

Check Google Ads isn't counting twice

This one quietly breaks bidding. In Google Ads go to Goals → Conversions → Summary and look at every action in the Purchase category.

✗ Wrong Google Shopping App Purchase Primary GA4 purchase (imported) Primary Each order counted twice. Bidding thinks ads work twice as well. ✓ Right Google Shopping App Purchase Primary GA4 purchase (imported) Secondary One action drives bidding. The other stays for reference only.
Figure 4: Only one purchase action should be set to Primary.

A simple weekly routine

QuestionLook at
How much did we sell?Shopify
Where did shoppers come from and what did they do?GA4
Which campaigns and products should get more budget?Google Ads, checked against Shopify
Is the business profitable after ad spend?Shopify revenue ÷ total ad spend across all platforms

That last row matters most. When every ad platform claims credit, adding up their numbers overstates your sales. Total ad spend against Shopify revenue (often called MER, marketing efficiency ratio) is the number that can't be double counted.

Go deeper

Tracking & Analytics (Server-Side) 2026

Set up clean purchase tracking once, add server-side events, configure enhanced conversions, and build a weekly reporting sheet everyone on your team trusts.

[ Open the course ]