Why GA4 shows fewer purchases than Shopify, and how to fix it

How big should the gap be?

GA4 will never count every Shopify order. Some orders never happen in a browser, and some visitors decline tracking. The goal isn’t a perfect match. It’s a gap you can explain line by line.

The quickest sanity check: compare the gap with your cookie banner’s decline rate. If about 20% of visitors decline and GA4 misses about 20% of online orders, your setup is probably fine. If GA4 misses 40%, something is broken.

The eight causes

Most gaps come from a combination of these. The first three are the most common in audits.

Reconcile by order ID

Totals hide problems. Matching orders one by one shows exactly which ones are missing and what they have in common.

1. In Shopify, export last month’s orders. Keep the Id, Name, Source, Total and Created at columns.
2. In GA4, list every transaction_id for the same dates, with an Exploration or the BigQuery export.
3. Match on the field your pixel sends. Some setups send the order ID, others the order name like #1042.
4. Group the unmatched orders by source, device, payment method and date. The pattern points to the cause.

With the BigQuery export, this query returns one row per tracked order and flags duplicates:

SELECT
  ecommerce.transaction_id,
  COUNT(*) AS purchase_events,
  MIN(TIMESTAMP_MICROS(event_timestamp)) AS first_seen
FROM `your-project.analytics_123456789.events_*`
WHERE event_name = 'purchase'
  AND _TABLE_SUFFIX BETWEEN '20261001' AND '20261031'
GROUP BY 1
ORDER BY purchase_events DESC

Fix the checkout event

Since Shopify moved checkout tracking to Customer Events, the reliable way to send purchases is a custom pixel subscribed to checkout_completed. Add it in Settings → Customer events → Add custom pixel.

analytics.subscribe("checkout_completed", (event) => {
  const c = event.data.checkout;
  window.dataLayer = window.dataLayer || [];
  window.dataLayer.push({
    event: "purchase",
    ecommerce: {
      transaction_id: c.order?.id,
      value: c.totalPrice?.amount,
      tax: c.totalTax?.amount,
      shipping: c.shippingLine?.price?.amount,
      currency: c.currencyCode,
      items: c.lineItems.map((li) => ({
        item_id: li.variant?.sku || li.variant?.id,
        item_name: li.title,
        item_variant: li.variant?.title,
        price: li.variant?.price?.amount,
        quantity: li.quantity
      }))
    }
  });
});

This pushes to a dataLayer inside the pixel sandbox. Load GTM inside the same custom pixel so it can read it, and connect consent through the Customer Privacy API before any tag fires.

What to do next

• Run the reconciliation on last month’s orders.
• Fix missing or duplicate purchase events before touching anything else.
• Then connect consent, so the remaining gap matches your decline rate.
• Recheck after every theme update or new checkout app.