Google Gemini: New UTM Parameters Enhance AI Traffic Attribution for SEOs
Google Gemini Adds UTM Parameters to Links, Giving SEOs a Clearer Picture of AI-Driven Traffic
Google's Gemini platform has quietly begun adding UTM parameters to outgoing links — a move that could significantly improve how site owners and SEO professionals measure and attribute traffic arriving from AI-powered search tools.
The update arrives at a critical moment. As AI-driven search tools increasingly intercept user queries before they ever reach a traditional search results page, webmasters have struggled to accurately measure how much traffic platforms like Gemini actually deliver. For many sites, that traffic has been disappearing into analytics black holes — misclassified and invisible.
Why UTM Parameters Matter for AI Traffic Attribution
UTM parameters are small tags appended to URLs that tell analytics platforms like Google Analytics 4 (GA4) exactly where a visitor came from. Without them, traffic from AI chatbots like Gemini can register as "direct" — the default catch-all category used when no referrer or tracking data is present.
Understanding the difference between direct traffic and organic search traffic is already a persistent challenge for SEOs, and AI-sourced visits have made that distinction even harder to draw cleanly.
The problem is more widespread than many SEOs may realize. A Reddit user identified as CrawlWarden first flagged the development in a thread that quickly drew attention from Google's own John Mueller.
CrawlWarden explained the core issue clearly: "There is a referrer, but it mostly survives on desktop web only. Reported pass-through from the Gemini mobile app runs around 9 percent and Android assistant invocations strip it entirely. So a large share of that traffic has been landing in Direct instead of GA4's AI assistant medium. A UTM survives the in-app webview and shows up in raw server logs too, which is why this matters more than it looks even though the referrer technically already existed."
In other words, even when a referrer signal existed, it rarely survived the journey from Gemini's mobile environment to a website's analytics dashboard. The UTM parameter is a more durable solution — one that holds up inside app-based webviews and server-level tracking.
The Scale of the Referrer Problem
To understand why this matters, consider the user journey. When someone interacts with Gemini on a mobile device or through the Android assistant and then clicks through to a website, the referrer header — the signal that tells your analytics platform where that visitor came from — is frequently stripped before it reaches your server. On desktop web browsers, referrer data passes through reasonably intact. On mobile, and particularly within in-app webviews, that signal degrades significantly.
This is not a minor edge case. Mobile accounts for the majority of web traffic globally, which means a substantial proportion of Gemini-referred visits have, until now, been effectively invisible to standard analytics setups. Those visits were not being lost — they were simply being misattributed, most commonly inflating "direct" traffic figures and obscuring the true contribution of AI platforms to site discovery.
Google's Mueller Weighs In
Mueller's response to the Reddit thread confirmed that Google is aware of the referrer problem and is open to addressing it further. After spotting the UTM addition himself, Mueller asked whether Gemini already passed referrer data — and when CrawlWarden explained the limitations, Mueller offered to escalate the issue internally.
"If someone has more information on what happens with the referrer — ideally with something I can reproduce — I'm happy to forward that to the team," Mueller wrote. "It would be great to have these retained in addition to the UTM-tagging."
The exchange is notable because it signals that Google is paying attention to how AI referral data reaches site owners and that further improvements could follow. The Reddit thread itself was posted on or around October 1, 2026, with the original poster noting the UTM behaviour was "very new — 24 hours or so."
How This Compares to ChatGPT's Approach
Google is not the first AI platform to introduce UTM tagging. ChatGPT adds UTM parameters to outgoing links, but only when answers are grounded in live web sources rather than information drawn from training data. That conditional behaviour creates its own measurement gaps — a site referenced from ChatGPT's training data may receive visits that never carry any attribution signal at all.
Google has not yet documented how or when Gemini triggers UTM parameters. It is reasonable to assume UTMs are added when Gemini directly references a website in its response — but until Google publishes official guidance, the exact conditions remain unclear. SEOs relying on this data to inform strategy should treat current figures as directional rather than definitive. Monitoring the Google Search Central documentation for updates on this behaviour is a practical step worth taking now.
What SEOs Can Actually Do With This Information
The addition of UTM parameters to Gemini's outgoing links has immediate practical value for anyone managing a website or running an SEO strategy in the age of AI search.
Improved Attribution Changes the Conversation With Stakeholders
Better referral attribution means site owners can now segment Gemini-referred visitors in GA4 and compare their behaviour against visitors arriving from traditional Google Search or other AI platforms like ChatGPT. That kind of side-by-side comparison has been nearly impossible to do accurately until now.
Attribution also makes the business case easier. When SEOs can point to real numbers tied to a specific AI channel, justifying resource allocation — whether that means creating AI-optimised content or investing in structured data — becomes a much more straightforward conversation with stakeholders. For a broader grounding in how SEO and analytics work together to inform strategy, it is worth revisiting how your current measurement framework is structured before drawing conclusions from this new data stream.
Putting the Data to Work
The lack of documentation around Gemini's UTM implementation is a gap worth watching. Until Google clarifies the conditions under which parameters are applied, incomplete data should not drive hard budget or strategy decisions. With that caveat in mind, there are concrete steps worth taking now:
- Check your GA4 source/medium reports to see if Gemini-tagged traffic is already appearing. Look for UTM parameters referencing Gemini in your acquisition data and compare volume against what previously landed in "direct."
- Set up a dedicated Gemini segment in GA4 so you can track engagement metrics — bounce rate, pages per session, conversions — separately from other AI and organic sources. This will help you understand whether Gemini visitors behave differently from traditional search visitors.
- Monitor Google's documentation channels for an official explanation of when Gemini applies UTM parameters. Until that guidance arrives, avoid drawing firm conclusions from incomplete data.
Structuring Your Analytics Setup for AI-Era Search
As AI platforms become a more significant source of referral traffic, the tools you use to track and interpret that traffic become correspondingly more important. Google's suite of tools for growing your business online — including GA4, Search Console, and Looker Studio — will each play a role in building a coherent picture of AI-driven traffic as measurement standards evolve.
Structured data and clear page-level signals may also influence how frequently Gemini references and links to a given page — though Google has not confirmed this directly. SEOs who have already invested in schema markup and semantic content structure may find themselves better positioned as AI referral tracking matures.
Google's move mirrors a broader shift happening across AI platforms toward greater transparency in referral tracking. As tools like Gemini and ChatGPT become more central to how people discover content online, the ability to measure that discovery is no longer a technical nicety — it is a strategic necessity. The introduction of UTM parameters is a meaningful step forward, but it is the beginning of a longer process of standardisation, not its conclusion.
