Reach’s plan to cut 220 editorial jobs, close three online-only brands and create about 60 roles focused on digital revenue is a concrete sign that AI-answer behavior has reached publisher operating decisions. The company says a 46% year-on-year decline in Google traffic, alongside features including Google AI Mode and AI Overviews, has reduced the need for readers to click through to its sites. theguardian.com
For people doing GEO, the important development is not that a publisher has discovered AI search. It is that Reach is changing the size and allocation of its newsroom around a changing distribution model: less dependence on story volume and page views, more investment in original journalism, distinctive brands, subscriptions, longer-form video and “active engaged time.” That is a much harder business response than an experiment with answer-engine formatting. theguardian.com
The evidence should still be read precisely. Reach’s reported traffic decline and its restructuring arrive together, but the supplied reporting does not establish a clean causal measurement showing what share of job losses is attributable specifically to AI summaries. Reach also cites competition from the BBC, while the wider restructuring concerns titles that were not becoming the “dominant publisher” in their regions. theguardian.com
This is a distribution-model change, not just a traffic story
Reach owns the Mirror, Express, Daily Star, Daily Record and regional publications including the Manchester Evening News, Birmingham Mail and Liverpool Echo. Its decision includes closing KentLive, AberdeenLive and GalwayBeo, while it creates about 60 editorial roles intended to drive digital revenue growth. theguardian.com
That combination matters. A publisher is not simply reacting to fewer visits by cutting cost. It is reallocating editorial capacity toward formats and revenue mechanisms it believes can support a direct relationship with an audience. Reach’s chief content officer, David Higgerson, said the company was in a “mammoth shift” in how audiences want content and journalism. His stated newsroom response was “less emphasis on story volume and more on original journalism and distinctive brands.” theguardian.com
For GEO practitioners, that is the central implication. Visibility in an AI-generated answer is not equivalent to a viable publishing outcome. If an answer resolves a user’s immediate need before a click, then conventional search acquisition can lose value even when the underlying publisher’s material remains part of the information ecosystem. Reach explicitly said Google’s AI features can negate the need for readers to click through. theguardian.com
GEO work therefore cannot stop at proving that a domain, brand or passage is visible in AI-mediated search. The commercial question is what happens after that visibility: whether it produces visits, whether those visits are sufficiently engaged, and whether the publisher can convert the relationship into subscription or another revenue stream.
Why “active engaged time” is the more revealing signal
Reach says it will move away from page views as its traditional measure of audience engagement and focus on “active engaged time.” Editors are also expected to make title-specific changes intended to reach audiences through journalism for which their brands will be known and to convert more people into paying subscribers. theguardian.com
This does not mean page views have become irrelevant. The supplied sources do not say that. It does mean Reach is publicly treating raw traffic volume as an inadequate organising metric for its next phase. That is notable because AI answers create a measurement problem for acquisition-led publishing: a search result may expose information, satisfy intent and shape audience perception without producing the publisher-side session that a page-view model captures.
For GEO, engaged time is not a replacement KPI that can be applied mechanically to every site. But Reach’s shift highlights a useful distinction between answer visibility and audience value. A short factual query that ends at an AI summary may create awareness without a visit. A user who needs original reporting, a recognised editorial voice, a deeper video format or ongoing access may still have a reason to form a direct relationship with the publisher.
- Track AI-search visibility separately from referred sessions rather than treating them as the same outcome.
- Assess whether pages that attract search users generate meaningful engaged time once users arrive.
- Identify which information needs can be satisfied in a summary and which require original, distinctive reporting.
- Connect GEO reporting to subscription and other direct-revenue outcomes where those are business priorities.
Those are analytical requirements, not a claim that every publisher should copy Reach’s editorial strategy. The reporting only establishes what Reach intends to do. It does not provide comparative evidence that subscriptions or longer-form video will offset the lost traffic or the job reductions.
The numbers are serious, but the attribution has limits
The Guardian reports that Reach had earlier said its slump in overall digital page views was driven by a 46% year-on-year decline in Google traffic. It also reports digital revenue of £128.9m in the year to 3 March, down almost 1%, and 50,000 paid digital subscribers across six major titles, against a target of 75,000 by the end of the current financial year. theguardian.com
These figures make the distribution pressure measurable. They do not, however, let us calculate the economic effect of AI summaries by themselves. We do not have a breakdown of traffic by Google feature, query type, brand, device or reader cohort. We do not know how many lost Google referrals were previously monetised, how much traffic moved to other channels, or whether a given content category experienced a different outcome.
There is also an inconsistency in the supplied reporting that should not be smoothed over. The Guardian reports a 46% year-on-year decline in Google traffic. theguardian.com A post on daily.dev instead says Reach’s half-year results showed a 55% drop in Google referrals and a 40% decline in website audience. daily.dev The extracts do not explain whether these figures cover different reporting periods or measures, so they should not be treated as interchangeable.
Nor should AI be presented as Reach’s only stated pressure. Higgerson also blamed an “expansionist BBC,” saying it mirrored commercial publishers’ local output “up to 70% of the time.” theguardian.com The closures themselves were attributed to the three brands failing to become dominant publishers in their regions. theguardian.com
What this changes for GEO strategy
The strongest conclusion is not “AI summaries caused 220 job cuts.” The source material does not support that simplification. The stronger, supportable conclusion is that a major publisher has connected reduced Google traffic, readers’ use of AI-generated summaries and a broad redesign of its editorial and revenue priorities. theguardian.com
That moves the GEO discussion beyond rankings, citations and answer inclusion. For publishers, the unit of analysis is increasingly the distribution system: which work earns a click, which work earns attention after the click, and which work can build a paying relationship when search increasingly answers before referral.
Reach’s response is especially clear on one point: it is concentrating on journalism and brands that can be distinctive enough to justify a direct audience relationship. Whether that succeeds remains unresolved. But the workforce decision itself is evidence that AI-answer behavior is being treated as a strategic operating condition, not a speculative future scenario.