Google Analytics today introduced Source Group, a new dimension that consolidates fragmented source values for Facebook, Instagram, and TikTok into single, standardised reporting entries - while simultaneously adding built-in groupings for emerging AI traffic sources including ChatGPT and Perplexity.

The update, announced on June 11, 2026 in the Google Analytics Help Center, addresses one of the most persistent data quality problems in multi-platform measurement: the same traffic source arriving under a dozen different labels. A Facebook visit, depending on how it was tagged or referred, might appear as "facebook", "fb", "Meta-facebook", or any number of variant strings. That fragmentation has long distorted budget analysis and cross-channel attribution for advertisers comparing performance across platforms.

What Source Group actually does

Source Group is a new dimension - distinct from the existing Source Platform field - that groups those variants into a single clean value. According to Google, "a single value for a given source is vital for performance and attribution analysis," and the new field is designed to provide exactly that. The example cited in the documentation is illustrative: instead of "facebook", "fb", and "Meta-facebook" each appearing as separate rows in a report, Source Group consolidates all three under "Facebook."

Source Platform, which has existed in Google Analytics for some time, is being updated in parallel. According to Google, the platform is improving some of its classifications to align with how Source Group now works. Where Source Group produces a high-level brand label - "Instagram" - Source Platform sits one level up, labelling that same traffic as "Meta Ads." The two fields serve related but distinct purposes: Source Group identifies the specific origin, Source Platform identifies the advertising ecosystem.

The scope of standardisation is broad. According to Google Analytics documentation, the update covers third-party platforms including TikTok, Pinterest, and Amazon, applying the same level of granularity that Google's own inventory - YouTube, Search, Maps - has long enjoyed. A messy string like "%instagram%" or "ig" will be automatically bundled into the clean "Instagram" value at the Source Group level, and into "Meta Ads" at the Source Platform level.

AI traffic sources get native groupings

The more forward-looking element of today's update is the inclusion of AI platforms in the Source Group taxonomy. According to the documentation, the new field includes "built-in grouping for emerging traffic sources like ChatGPT (OpenAI) and Perplexity." This marks a further step in Google Analytics' expanding treatment of AI-generated referral traffic as a first-class measurement category.

The context here matters. In May 2026, Google Analytics added a native AI Assistant channel to the Default Channel Group, automatically tagging traffic from chatbots like ChatGPT, Gemini, and Claude with the medium value "ai-assistant" and placing it under a dedicated channel - without requiring any configuration from property owners. That update addressed channel-level classification. Today's Source Group update addresses source-level standardisation: if ChatGPT traffic arrives under variant referrer strings, Source Group will consolidate those into a single clean value.

Before either of those native features existed, Google Analytics was already pointing marketers toward manual workarounds. In August 2025, the platform published guidance for building custom channel groups using regex patterns to capture AI chatbot traffic, marking what was at the time the platform's first official acknowledgment that AI tools needed dedicated measurement treatment. That manual approach placed a meaningful configuration burden on individual properties. Each property owner had to navigate to Admin, access Channel Groups, and maintain regex expressions that could break if a chatbot changed its URL structure. The progression from regex workaround to native channel to native source grouping reflects how quickly AI traffic has moved from edge case to standard measurement requirement.

The scale of that shift is visible in the traffic data. According to Similarweb data published on June 11, 2026, ChatGPT held 52.7% of global generative AI website traffic as of May 26, 2026 - down from 76.4% twelve months earlier. Gemini had risen from 8.9% to 27.3% over the same period. Claude tripled its share from 1.6% to 8.9%. Perplexity and others account for the remainder. The overall AI traffic ecosystem is growing and fragmenting simultaneously, which is precisely the kind of environment where a standardised source grouping framework becomes operationally useful.

Retroactive data and historical analysis

One technically significant aspect of Source Group is that it is populated retroactively. According to Google Analytics documentation, because the field is applied retroactively, analysts can use it as a dimension across historical data to examine patterns that predate the feature's launch. For marketers trying to understand how traffic from a platform like TikTok has changed over a year-long period - and who have struggled with variant source strings corrupting that longitudinal view - this retroactive availability has practical value.

The distinction between retroactive and forward-only data matters in analytics. Some Google Analytics features apply only to data collected after a certain configuration date, which forces analysts to treat pre- and post-feature periods as incomparable. Source Group avoids that problem by applying its grouping logic to the historical session record. The breadth of that retroactive window is not specified in the documentation.

Source Platform updates

The parallel update to Source Platform is described in the documentation as an improvement to classifications in order to align with the Source Group changes. Source Platform is an existing field, not a new one, so the update affects how existing traffic is categorised rather than introducing a new reporting surface. The documentation does not enumerate every specific reclassification, but the intent is that the two fields - Source Group and Source Platform - now form a coherent hierarchy: Source Group identifies the specific platform origin, Source Platform identifies the broader advertising system to which that origin belongs.

For third-party platforms specifically, the documentation notes that the update now provides "the same level of granularity for third-party platforms (TikTok, Pinterest, Amazon) as it does for Google inventory (YouTube, Search, Maps)." That parity framing is meaningful. Historically, Google's own properties have tended to be better-instrumented within Google's own analytics tools. Extending equivalent granularity to competitors' platforms is a practical acknowledgment that advertisers routinely run campaigns across multiple ecosystems and need consistent comparisons.

Relationship to the broader 2026 Analytics roadmap

Today's Source Group update arrives in the context of a sustained period of development inside Google Analytics. January 16, 2026 brought three simultaneous beta launches: cross-channel budgeting, improved web conversion management, and a conversion attribution analysis report. April 29, 2026 introduced Task Assistant, a guided configuration tool designed to help property owners complete the foundational setup steps that those beta features depend on. May 4, 2026 brought cross-channel conversion reporting data to the Data API in alpha. Three days earlier, on June 8, 2026, Google Analytics added a direct integration with Google Business Profiles, pulling seven local interaction metrics - calls, bookings, direction requests, website clicks, messages, menus, and GBP interactions - into the same reporting environment as web and app data.

Source Group fits within that pattern. Each of these updates extends the range of data that Google Analytics can ingest and standardise. Cross-channel budgeting tools are more reliable when the underlying source data is consistent. Attribution models produce cleaner output when the same platform is not split across fifteen variant source strings. The retroactive availability of Source Group means that the cross-channel budgeting features launched in January - which depend on accurate medium-level and source-level data to model how channels contribute to conversion paths - may now produce more coherent historical analysis than was possible when those features launched.

Why consistent source data is a prerequisite for measurement

The fragmentation problem that Source Group addresses is not a minor inconvenience. When a single platform appears under multiple source strings, budget reporting breaks in compounding ways. A media buyer comparing Facebook and Instagram performance against Google Search - a routine task for any multi-channel advertiser - is working with corrupted data if "facebook", "fb", and "Meta-facebook" are appearing as three separate rows rather than one. The platform's budget share looks artificially smaller. Optimisation decisions made against that data are correspondingly unreliable.

The problem has become harder as the number of traffic sources has grown. In 2020, most web traffic came from a manageable set of sources with relatively consistent tagging. By mid-2026, the list includes major social platforms, multiple search engines, retail media networks, and a growing set of AI assistants - each with its own referrer patterns, UTM conventions, and tagging inconsistencies. According to Google's documentation, Source Group is designed precisely to absorb that complexity: it "automatically bundles disparate messy source values" into clean, high-level categories.

For advertisers running campaigns on TikTok, Pinterest, and Amazon alongside Google Search and YouTube, the standardisation means those platforms can now sit in comparable rows in a cross-channel report without manual dimension mapping. That parity has been a gap. As covered by PPC Land, Google Analytics had introduced manual traffic source dimensions as far back as February 2024, but those required advertisers to implement and maintain consistent UTM parameters themselves. Source Group shifts part of that normalisation burden from the advertiser to the platform.

The AI measurement dimension

The inclusion of ChatGPT and Perplexity in the Source Group taxonomy is not incidental. It reflects the structural reality that AI assistants are now a measurable, if still modest, component of referral traffic for many websites. The measurement challenge is that AI platforms are not fully analogous to social networks or search engines in how they generate referrals. Some AI assistants pass referrer strings; others do not. Some traffic that originates from an AI assistant arrives without any UTM parameters. Some arrives as what appears to be direct traffic. Standardising the source values that do arrive is a necessary first step, but it does not solve the attribution gap for sessions that arrive without identifiable origin signals.

The scale context is useful here. Research published by Ahrefs in February 2026 found that ChatGPT sends approximately 190 times less traffic to websites than Google, despite handling roughly 12% of Google's query volume. That gap reflects the fundamental difference between a search engine that surfaces links as its primary output and a conversational assistant that aims to answer questions directly. Most ChatGPT sessions end without an outbound click. For the sessions that do result in a referral, having a clean Source Group value - "ChatGPT" or "Perplexity" - rather than a string of variant referrer fragments makes subsequent analysis more tractable.

Advertising section improvements

According to Google's documentation, one of the explicit goals of the Source Group update is to make "cross-channel reporting in the advertising section more simple and actionable." The advertising section of Google Analytics - which houses the conversion attribution analysis report, the cross-channel budgeting tools, and the Conversion performance report now accessible via the Data API - depends on consistent traffic source data to function as intended. If the source dimensions feeding into that section are fragmented, the cross-channel comparisons those tools produce are distorted.

The documentation frames the update in terms of enabling advertisers to "identify performance and optimization opportunities from both first and third-party platforms and inventory sources." In practice, that means a media buyer running campaigns on Meta, TikTok, Pinterest, Google Search, and YouTube can, after Source Group is fully populated, view those channels in a single consistent dimension without needing to clean the data externally or create custom groupings to collapse variant strings.

Technical availability

The update was published in the Google Analytics Help Center on June 11, 2026. Source Group is a new dimension, distinct from the existing source, medium, and Source Platform fields. It is automatically populated - no implementation steps or tag changes are required from property owners. The retroactive population means historical reports will reflect the new grouping without any date-based break in the data series.

Source Platform updates are also automatic, applying to the existing field rather than creating a new reporting surface. Properties that have existing reports or explorations built on the Source Platform dimension will see the updated classification values, which may affect historical comparisons for analysts who have built reporting logic against specific Source Platform values.

Timeline

Summary

Who: Google Analytics, the web analytics platform used by a large share of the global advertising industry.

What: The launch of Source Group, a new dimension that consolidates variant source strings - including "facebook", "fb", and "Meta-facebook" - into single clean values for platforms including Facebook, Instagram, TikTok, Pinterest, Amazon, YouTube, Google Search, Google Maps, ChatGPT, and Perplexity. The existing Source Platform field received parallel classification improvements. Both updates are automatically applied with no implementation steps required from property owners. Source Group is populated retroactively across historical data.

When: June 11, 2026, published in the Google Analytics Help Center.

Where: Available across Google Analytics properties globally, surfaced as a new dimension in standard reports, explorations, and the advertising section.

Why: Traffic from a single platform frequently arrives under multiple variant source strings, distorting cross-channel budget analysis and attribution. Source Group normalises those variants into a single reportable value at the platform level, enabling more reliable comparisons across Google inventory, third-party social platforms, retail media networks, and AI assistant traffic sources. The update also future-proofs measurement infrastructure as AI-generated referral traffic continues to grow and fragment across multiple competing platforms.