Life after SEO: How newsrooms are adapting to the AI search shift

Search referral traffic to publishers is collapsing, and newsrooms across the industry are rethinking how journalism gets discovered as Google and AI chatbots increasingly answer questions directly rather than sending readers to source sites. According to Chartbeat's annual publisher playbook, referrals from Google Search to hundreds of publishers fell 40% from July 2025 to July 2026, double the decline seen in the prior 12 months. The shift has intensified fears of "Google Zero," the scenario in which Google stops functioning as a gateway to the web and becomes an answer engine instead.

For two decades, publishers built SEO teams, trained journalists to write search-friendly headlines, and structured stories around query patterns. That playbook is losing relevance as audiences get answers from Google's AI Overviews, Gemini, ChatGPT and Perplexity without clicking through. Google currently excludes news queries from AI Overviews, but news-adjacent searches, such as biographical or historical context, already surface AI-generated summaries that compete directly with reported journalism.

As a senior editor at a trade publication covering the industry for people who work in it, I've been tracking how newsroom leaders are responding, and the consensus so far is that there is no settled playbook for what's being called GEO, or generative engine optimization.

Ezra Eeman, Director of AI in Media at WAN-IFRA, describes the citation landscape as volatile: sources favored by AI platforms shift monthly, from Wikipedia to Reddit to YouTube, and citation patterns differ across ChatGPT, Gemini and Perplexity. Shelby Blackley, Director of Global SEO & AI Search at The Athletic, says the industry lacks sufficient data to know reliably what drives citations. Clara Soteras, an SEO and audience consultant, argues GEO isn't a total black box since trial and error and monitoring tools can reveal patterns, but she stresses that newsroom KPIs built around unique users and page views need to change, since AI citations often generate no clicks at all.

Some publishers are testing anyway. Axios, according to Ben Berkowitz who oversees newsroom AI operations, is leaning on its Smart Brevity format, short paragraphs and bullet points, which he says LLMs process well. Axios tracks how often it appears in citations for likely user queries, though results vary even when the same question is asked repeatedly. Berkowitz frames the payoff as brand authority rather than traffic, betting that Axios's newsletters, events and direct reader relationships will retain audiences even as referral traffic declines. Jessie Willms, SEO Editor at The Guardian, argues GEO should sit on top of solid SEO fundamentals rather than replace them, treating AI visibility as an ongoing signal of how discovery is shifting.

A Pew Research Center study of over 68,000 searches by 900 US Google users in March 2025 found only 1% resulted in a click on a source cited in an AI Overview, underscoring the disconnect between visibility and traffic. Eeman argues visibility still matters for brand recognition, subscriber growth and negotiating leverage with AI companies, pointing to citation concentration among outlets like Reuters, the BBC, the New York Times and the Guardian. He suggests smaller publishers may find an edge in hyperlocal or specialized coverage where they are among the only authoritative sources.

Publishers are also wrestling with whether to block AI crawlers. Research cited in the piece found 48% of major news sites across ten countries were blocking OpenAI's crawlers by the end of 2023, driven by concerns over uncompensated training use and inaccurate attribution. Eeman now advises a selective approach, protecting premium content while opening free content to crawlers as a brand introduction. Willms notes that licensing content to AI companies is an alternative, though revenue from such deals may be short-term or uncertain.

Impact

Referral traffic has dropped 40% year over year for publishers tracked by Chartbeat, double the prior year's decline. Generative AI use is rising fast, with weekly usage across six countries surveyed by the Reuters Institute nearly doubling from 18% to 34% between 2024 and 2025, though only 10% of a global sample use chatbots specifically for news, per the Digital News Report 2026. Citation visibility remains concentrated among a small number of established brands, leaving smaller publishers to compete on niche or local authority. Blocking decisions on AI crawlers carry tradeoffs between compensation concerns and discoverability, with no universal strategy across the industry.

What to watch

Newsroom leaders should watch for how AI platforms evolve their citation behavior, since patterns differ by platform and shift over time with no stable formula yet established. Expect continued experimentation with monitoring tools tracking brand mentions and prompt performance, alongside pressure to redefine success metrics beyond page views and unique users. Crawler policy will likely keep evolving as publishers weigh selective blocking against licensing deals of uncertain value. Most importantly, watch whether direct audience relationships, through newsletters, events and subscriptions, prove durable enough to offset further declines in search and social referral traffic.

Some newsrooms have experimented with an auto backlink tool to boost referral traffic, but many now question whether link-building still matters when AI summaries answer readers’ questions directly.

Many newsrooms are now weighing whether a seo backlinks builder still deserves a place in their workflow as AI-generated answers reshape how readers find stories.

Newsrooms that once relied on search traffic now have to work harder to get backlinks through original reporting, expert sourcing, and direct relationships with readers.

Source: reutersinstitute.politics.ox.ac.uk

Read our previous article: 8 Link Building Price Index Resources SEOs Should Check in 2026