Gemini users are reporting UTM parameters on outgoing links, offering website owners another signal for identifying traffic from the assistant. The observation could improve attribution where a referrer is missing, but it does not establish complete tracking across Gemini’s interfaces or a new default analytics channel.
What has been observed
Search Engine Journal reported the behavior on October 1, 2026, citing a public Reddit discussion in which John Mueller said he could see it too. That corroborates an observation; it is not a published implementation specification.
In the original discussion, one user describes source links carrying utm_source=gemini. Comments differ on when tagging first appeared. Its coverage across responses, devices and Gemini surfaces remains unclear. Claims about a specific mobile-referrer pass-through rate are not supported by a published methodology in the discussion and should not be treated as established measurements.
How the tag can help attribution
Google Analytics documentation explains that UTM values on destination URLs supply campaign information to Analytics. Source and medium can be examined in Traffic acquisition reporting. It also notes that UTM values are omitted from certain path-and-query dimensions and are represented in Page location.
Our practical interpretation is that a surviving source parameter can provide useful evidence when browser referrer information is absent. That does not guarantee collection: redirects, URL handling and the site’s measurement setup can affect what is retained. A source value also does not by itself determine the final channel classification.
Inspect the values before defining a segment
Start with actual incoming URLs and observed acquisition values. Avoid assuming that every Gemini visit has the same complete set of source, medium and campaign parameters. If the reported source value appears, it can support a focused investigation of those sessions, subject to the property’s collection and attribution behavior.
Check a representative click path from the assistant to the destination. Compare the initial URL with the page that loads after redirects. Where server logs record query strings, they may provide another observation of the parameter’s arrival. A logged request is not automatically equivalent to an analytics session or a human conversion.
A measurement change can resemble traffic growth
If previously unidentified visits become attributable, the Gemini row in a report may rise without an equivalent increase in total visits. That possibility matters when interpreting trends around the first observed tagging date.
Keep an annotation for the measurement change and compare overall traffic with the newly identifiable subset. Do not retroactively assign old direct sessions to Gemini merely because the new tag exists. Nor should an increase in tagged sessions be presented as proof that content optimization caused more assistant referrals.
Measure visits and outcomes separately
Once a usable segment is established, examine landing pages, engagement and relevant key events using consistent definitions. Separate the evidence that a visit carried a Gemini tag from the business value associated with that visit.
These observations can help teams assess which content receives referred visitors. They do not measure every brand mention, every citation or the number of people who viewed an answer without clicking. Click attribution is one part of AI visibility measurement.
A useful signal with unresolved coverage
The verified story is a publicly observed tagging behavior, including a reported Gemini source value. The opportunity is clearer attribution for the visits that carry and retain it. Complete coverage, a formal launch date and any pending change to GA4 default channel groups remain unestablished by the evidence reviewed.