A payment provider appears as a major referral source in your analytics. Did it bring customers to your store, or did customers simply pass through it during checkout? Before moving marketing budget, inspect the complete journey and the measurement configuration.

For a UAE business using a separate booking, shopping or payment domain, this is a reporting-quality question. Fixing attribution does not create additional sales, and a payment-return page is not proof that a transaction succeeded.

The illustration is a conceptual checkout measurement review, not a real GA4 report or evidence of ITZ client results.

Map the journey before changing settings

List the landing page, storefront, checkout, payment provider and return destination. Identify which domains the business controls, where the approved analytics tag runs and which transitions use links, forms or redirects. Include cancelled and failed payment paths.

Record the affected report, date range and attribution scope. A session-source report and an advertising attribution report need not answer the same question. Keep a copy of the current configuration so later comparisons have a defined baseline.

Distinguish unwanted referrals from cross-domain measurement

Google's unwanted-referral guidance identifies third-party payment processors as a common example of traffic that should not be treated as a new referral source when customers return. The setting changes how matching referrers are handled; it does not install measurement on the payment provider.

Google's cross-domain documentation describes maintaining measurement identifiers as visitors move between appropriately tagged domains. It requires the same tag ID from the same web data stream on the participating pages. Do not add an unrelated payment domain as if the business can configure its tags.

The implementation choice depends on the actual ownership and tagging boundary. Use a narrow rule tied to the observed journey rather than excluding broad sets of domains simply to make a report look cleaner.

Verify the linker through the real transition

For a cross-domain setup, check the documented linker behaviour on the relevant link or form and inspect the destination after redirects. Google notes that the _gl parameter can be removed when a destination redirects or does not support arbitrary query parameters.

Use authorised testing with appropriate consent settings and non-sensitive test data. Do not put personal customer information into campaign parameters or diagnostic screenshots. A successful parameter check is one part of the measurement journey, not proof of correct purchase reporting.

A digital marketing review covering campaign measurement and customer journeys can define the reporting question before configuration changes. Review Your Checkout Attribution with the domains, platform and report symptoms; the initial enquiry does not need analytics credentials or customer records.

Use a controlled acceptance set

The following is a proposed test brief. Agree the expected results with the implementer before running it:

  • A known campaign journey crosses the intended domains and returns to the correct page.
  • A direct visitor follows the same checkout path without receiving an invented campaign label.
  • A cancelled or failed payment does not become a confirmed purchase merely because the visitor returns.
  • A legitimate external referral remains distinguishable from an internal payment step.
  • Repeated return-page visits do not create unexplained duplicate transaction reporting.

Record timestamps, page sequence and the relevant test-order reference using approved handling. Compare the analytics result with the store's actual order/payment status. A CTA click, checkout start or WhatsApp click is not a confirmed lead or sale.

Keep interpretation cautious

Do not assume the setting retroactively repairs historical reporting. Google documents circumstances where a previously excluded source can still appear for returning users because of attribution behaviour. Mark the configuration-change date and explain any limitations in comparisons.

Consent choices, browser restrictions and missing tags can also limit observable journeys. Investigate these alongside referral settings rather than promising perfect attribution. The WhatsApp measurement guide covers a separate contact journey, while isolated ecommerce testing helps plan tests without unwanted customer activity.

Review Your Checkout Attribution to discuss domain ownership, tag coverage, referral rules and a documented acceptance test. The goal is evidence you can interpret responsibly before making marketing decisions.

Official Google documentation reviewed 26 October 2026. Test cases are proposed guidance. No traffic, conversion improvement or sales result is claimed.