GA4 shows fewer orders than WooCommerce: where they go

Your WooCommerce dashboard says 100 orders this week. GA4 says fewer. That gap is normal, and it has a handful of causes you can check one by one. GA4 records a purchase only when a tag sends it, and in most shops that tag runs in the shopper’s browser on the thank-you page. Anything that stops that page or that request loses the order in GA4, while WooCommerce still has it. Our guide to server-side tracking for online shops has the bigger picture.

WooCommerce is the source of truth for sales. GA4 is where you see which channel brought them. So the goal isn’t to make the two numbers identical; it’s to know why they differ and to close the gaps you can.

Why GA4 misses WooCommerce orders

Check the causes in this order. The first two are the ones a copy of the purchase sent from your server can fix; the third one shouldn’t be fixed at all.

1. The thank-you page never loads

The browser tag fires on the order confirmation page. A shopper who pays through a payment gateway (an online transfer, for example) and doesn’t come back to the shop, or loses the connection after paying, never reaches it. The order is paid in WooCommerce. GA4 never hears about it.

You can spot this one: the missing orders cluster around certain payment methods. Our article on paid orders that never reach the thank-you page covers it in detail.

2. The request is blocked

Ad blockers and privacy filter lists block requests to Google Analytics hosts. So do some browsers’ built-in tracking protections. The page loads, the tag runs, and the request goes nowhere.

With Google Consent Mode, GA4 needs analytics_storage granted to record the purchase as a normal event. If the shopper declined analytics cookies, the order isn’t there as a regular transaction, whatever tag you use. This gap is the shopper’s choice and should stay; nothing on the server side changes it (more in our article on Consent Mode v2 and server-side tracking).

4. The plugin sends the purchase at the wrong moment, or not at all

Some GA4 plugins fire purchase only on the thank-you page, some only for certain order statuses, some not for orders created in the admin panel. Cash on delivery and “on hold” orders often fall through. Place one order with each payment method you offer and watch which ones reach GA4.

5. Duplicates and mismatched order numbers

This one works the other way. GA4 merges two purchases only when they carry the same transaction_id. If one plugin sends the WooCommerce order ID and another sends the order number, GA4 counts two purchases. Fix duplicates first, or they will hide how many orders are really missing.

6. Late payments

This one matters for purchases sent from the server. GA4’s Measurement Protocol records an event at its own time only if that time is less than 72 hours in the past. A transfer paid days after the order still arrives if the server copy carries the time of payment; it doesn’t if it carries the time the order was placed.

How to compare GA4 with WooCommerce

  1. Pick a closed period, for example last week, and export WooCommerce orders with their ID, payment method and status.
  2. In GA4, open an Exploration with the Transaction ID dimension and the purchase event for the same period.
  3. Match the two lists by order ID. Orders missing from GA4 tell you which cause it is: by payment method (cause 1), by device or browser (cause 2), by status (cause 4).
  4. Look for transaction_id values that appear twice or don’t match any order (cause 5).

Give GA4 a day or two before comparing; its reports fill in with a delay.

What sending the purchase from the server fixes

When WooCommerce marks an order as paid, your shop’s server can send the purchase to GA4 through the Measurement Protocol. That copy doesn’t depend on the thank-you page, so it closes cause 1. It closes cause 2 when the blocker stopped only the GA4 request: the server copy still needs the client_id from the _ga cookie that gtag sets, and a blocker that stops gtag entirely leaves no cookie to send.

It does not close cause 3. A purchase from the server still needs the shopper’s analytics consent and that client_id, which ties the order to the visit and its traffic source. Without them the purchase isn’t sent.

Three details decide whether the server copy helps or makes things worse:

  • Use the same transaction_id as the browser tag. GA4 merges the two purchases only on an identical ID. The TallyHost plugin sends the WooCommerce order number, so set your gtag purchase to the order number too.
  • Check the API secret in DebugView. The Measurement Protocol answers with success even when the API secret is wrong, and the data silently disappears. Send one test purchase and confirm it in DebugView.
  • Send only what GA4 needs from the server. Page views and add to cart stay with gtag in the browser, which keeps the session. From the server, the purchase, the refund and the lead are enough; then remove refund and lead from gtag, because GA4 doesn’t merge those two.

The TallyHost plugin for WooCommerce does this: it reports paid orders from your server with the time of payment, TallyHost checks the shopper’s consent before anything goes to GA4, and the debugger shows each event and whether it went to GA4 or was skipped and why (no consent, no client_id). The integrations page lists what GA4 receives and which consent it needs.

If you also send the purchase to Meta from both the browser and the server, read how Meta matches the two copies. It works differently from GA4.

Send your shop's orders from the server

TallyHost reports paid orders with consent checked, to Meta, TikTok, GA4 and Google Ads. Free up to 100,000 events a month.

Start free

Read next