NEWS · NOVEMBER 14, 2026 · SEO

Google publishes a new help page on identifying Google Search traffic in analytics and server logs

On November 12, 2026, Google added a page to its Search Central documentation. It explains how visits from Google Search can be recognized in web analytics or server logs by reading the HTTP Referer header, and why those counts may not line up with clicks in Search Console. It is a documentation addition, not a product change.

01 · WHAT HAPPENED?

Google added an official write-up on spotting its own traffic

Google's Search Central updates log has an entry dated November 12, 2026: the company added a new page on spotting Google Search traffic via HTTP Referer headers. The stated reason is to show site owners how to find and follow that traffic in their own analytics.

The page is presented as a complement to Search Console. Search Engine Roundtable covered it the same day and Search Engine Journal the day after. This article rests on Google's page and those two reports.

02 · DETAILS

The Referer carries only a scheme and a host

Per Google, Search uses the origin referrer policy, so the header holds only the scheme and hostname of the page that sent the visitor, with no path and no query parameters. That also means the phrase a user typed never appears in it. To see which queries bring visits, the page points to the Performance report in Search Console.

The value depends on where the click happened. On the web, Google Search sends https://www.google.com/ or a country-specific Google domain, for example google.co.uk or google.co.jp. The Google app on iOS uses the same values. The Android app can send those too, and when it opens a link in a browser tab it may send a value shaped like android-app://com.google.android.googlequicksearchbox/https/www.google.com.

Other Google services served from the same origins, Google Images for example, send the same values. Discover visits carry whichever values belong to the surface on which the reader saw Discover. Search Engine Journal draws the practical conclusion: a google.com referrer shows that Google sent the visit, but it does not by itself tell Search, Discover and Images apart, and the Discover report in Search Console is the way to split Discover clicks out.

Sometimes there is no header at all. According to Google, private browsing and anti-fingerprinting protections, results opened in a new tab in some browsers, a jump from a mobile search widget to an outside browser, and privacy extensions or strict browser settings can all cause it.

03 · WHY IT MATTERS

Google now lists why your numbers may differ

The page says plainly that spotting Search traffic through the Referer is not exact, and that analytics or log figures often fail to match Search Console clicks one to one. The main points it covers, across its matching and discrepancy sections:

Custom channel rules, referral exclusions or filters can move visits into another channel such as referral or direct. Browsers also send a Referer when fetching images, CSS and JavaScript, but in that case the header points to your own page's address, so counting should cover landing-page requests only. Client-side redirects can swap in your own page URL, and redirects that do not keep the query string drop parameters such as gclid, widely used for ad tagging. If a redirect applies a stricter Referrer-Policy header, the Referer can disappear entirely.

The measurement logic differs too. Search Console counts a click at the moment a user picks a link, while analytics tools record a visit only after the page loads and the tracking code runs. That second step can be stopped by ad blockers, consent choices and privacy settings. Finally, Search Console reports organic clicks from Google Search only and leaves out Google Ads clicks, whereas analytics totals may include paid traffic depending on setup.

Search Engine Journal adds that the page does not deal with separating AI Mode and AI Overviews clicks. SEJ says the standard Performance report counts both but offers no search appearance type that isolates them, citing its own November 2 reporting.

04 · TÜRKİYE

What it means for businesses in Türkiye

This section is our reading, not a fact from the sources. Because the mechanism works on the browser and analytics side, any site that gets visitors from Google, including those in Türkiye, may find it relevant. The page gives local Google domains as examples; we would expect domains like google.com.tr to show up for searches from Türkiye, but that comes from our own inference, not from the page.

The practical value is in client reporting. A gap between organic sessions in GA4 and clicks in Search Console is a common point of debate for businesses, and Google now lists the reasons on an official page. Checking the landing-page filter, redirect chains and the effect of the consent banner in a report may help explain the gap; that is our suggestion.

The UNALSOFT view

We turn this story into a short checklist for site owners: do your redirects keep the query string, do your analytics channel rules put Google traffic in the wrong bucket, and how much of a measurement gap does your consent banner leave? In our web design work we review site structure, redirects and measurement setup with these questions. The Google page promises no measurement outcome; it only explains why you may see different numbers.

Why don't your analytics numbers match Search Console?

A short conversation is enough to review the redirect and measurement setup of your site together.

Message on WhatsApp