For many media organizations, the threat of “Google Zero” is increasingly becoming a reality. Between November 2024 and November 2025, traffic from Google Search to more than 2,500 sites in the Chartbeat network decreased by a third (33%) worldwide and by 38% in the USA. These moves follow similarly precipitous declines in recent years in referral traffic from major social networks like Facebook and X. 

As a result, notes AdExchanger editor Anthony Vargas, “Publishers, typically a tight-lipped crowd, have been surprisingly candid about losing 20%, 30% and in some cases even as much as 90% of their traffic and revenue over the past year.”

A survey of media leaders featured in the Reuters Institute’s latest annual predictions report revealed that publishers anticipate a further decline in traffic from search engines of more than 40% over the next three years. “Not quite ‘Google Zero’, contends author Nic Newman, “but a substantial impact none the less.”

In response, companies need to focus on how to address this challenge. And how to do so quickly, as traditional sources of referral traffic continue to hemorrhage. 

Five core factors to address Google Zero

Here are the five core factors that publishers should incorporate into their strategy and workflows.

1. Grasp the size of the problem

The term Google Zero stems from a question posed by The Verge’s Editor-in-Chief Nilay Patel who asked what would happen to businesses if their Google traffic were to go to zero?

Changes to social media and search algorithms have reduced referral traffic for some publishers. In the AI era, these dynamics are becoming even more pronounced. The impact of AI-snippets at the top of Google search results means that when users ask a question, they find answers at the top of the page and many don’t click through for more detail or scroll down for more options. 

Data from the Pew Research demonstrates the impact of this: when an AI summary appears, users click traditional search results only 8% of the time, compared to 15% without one. 

Meanwhile, for publishers pinning their hopes on Google Discover, it’s worth remembering that most of the growth in this space comes from breaking news, content which is often excluded from Google AI summaries. 

“Google Discover traffic is mostly a mirage,” contends the media analyst Simon Owens, recommending that media companies “avoid optimizing their content operations around it.” “Publishers never owned those audiences and therefore should never have counted on them,” he added.

Lastly, the situation is further exacerbated by content in replacing carefully crafted headlines from publishers with those generated by AI. Initially confined to content in Discover, this is now happening in Search too. As Sean Hollister, senior editor at The Verge, put it, “This is like a bookstore ripping the covers off the books it puts on display and changing their titles.” 

All of this is to say that if you’re still heavily reliant on Google as an engine for traffic, it’s time to think again. 

2. Understand that this is part of a wider shift in user behavior

Much of the coverage to date has focused on the supply side, centering on publishers and platform dynamics. Far less attention has been paid to the demand side and evolving user needs.

Traditional search feels increasingly outdated. In its place, users are turning not just to AI-generated summaries within Google, but to AI tools like Claude, ChatGPT, and Perplexity as they begin their information journey.

A recent study from Eight Oh Two Marketing, surveying 500 active AI users, found that 37% of those sampled now begin their search with these types of AI tools rather than traditional search engines. “Consumers are not choosing AI because it is trendy. They are choosing it [AI] because search has become too noisy, too effortful, and too slow.” 

At the same time, 85% of respondents said they still double-check AI-generated answers using traditional search, using these platforms for verification and deeper exploration.

Publishers need to recognize the implications of these behaviors. Those absent from the first phase, AI-driven discovery, may never be found in the second. Equally, those only visible in the verification phase are absent from the critical entry point of this new information funnel. If they’re not present in both of these environments, then they risk being overlooked and left behind. 

3. Recognize that visibility is increasingly binary

AI environments are far more winner-takes-most than traditional search. Previously, a publisher that ranked fifth on a search query would still earn traffic, as might those even lower. AI-mediated discovery cites a couple sources and if you’re not on the shortlist, you’re unlikely to be discovered.

For media companies, understanding how AI-generated answers are created is critical for ensuring that your content is featured in the places and spaces that LLM’s crawl. The foundations of good SEO, such as authority, clarity, credibility, still matter. Build on your SEO foundation; you don’t need to fully reinvent the wheel. 

Research on AI citation patterns shows that AI systems – like search engines before them – tend to favor sources with strong off-page authority. So, media companies will want to ensure that their content demonstrates the E-E-A-T (Experience, Expertise, Authoritativeness, and Trust) principles. 

That means having a presence on high-authority external platforms, citations in credible databases, earned media in leading publications, as well as content across the information ecosystem. This includes your own website, to thought leadership demonstrated by execs on LinkedIn and in publications like Forbes, through to engagement on Reddit forums, and material produced for other platforms such as company podcasts, YouTube videos, and so on.

A recent SEMrush analysis of 325,000 prompts across AI platforms found LinkedIn ranking second only to Reddit as a source for AI chatbot responses, particularly for professional queries. 

4. Build content that AI can’t replicate

Not all content is equally vulnerable to AI summarization. Sites that prioritize original storytelling, exclusive imagery, and strong visuals, can still thrive. 

According to The Digital Bloom, People.com saw a 27.52% year-over-year traffic increase through September 2025 by adopting this approach. Similarly, Substack witnessed a 40% year-over-year growth in July 2025. “This growth reflects users seeking authentic voices and first-hand perspectives,” they wrote. These are “content types that Google’s Liz Reid [VP, Search] specifically identified as gaining traffic in the AI era,” they added, referencing a discussion on wider online behavioral shifts that she had with The Wall Street Journal last year.

Conversely, if your content can be summarized in three bullet points by an AI, it is a commodity. That’s one of the reasons why a lot of evergreen content has been cannibalized and its traffic for many publishers has tanked. 

In response, many media companies need to simply produce less content and ensure that what they are doing is better. Content needs to be more meaningful. More impactful. More distinctive. Can it help to drive a conversion – in the form of a registration, a newsletter signup, or a subscription – so that audiences come to you first, and not their AI platform of choice? (And if not, should you be doing it at all?)

Similarly, media companies need to continue to invest in products that reduce dependency on external discovery channels. The New York Times is perhaps the best known proponents of this. Nearly half of Times digital subscribers now pay for more than one Times product, attracted by a mix of games, cooking, audio, news, opinion, product reviews and sport. Each of these are elements that can help to create habits and drive engagement that is independent of search.

5. Rethink what success looks like

The narrative is not so much that “AI is killing search,” but that AI should force us to rethink what search looks like. Search is no longer just a driver of traffic, it’s a multi-faceted arena covering discovery (AI chat), synthesis (AI chat and snippets), and verification (actually going to your content to dig deeper). 

As a result, publishers have to look more broadly at metrics they measure and value. In the AI-era, software company ClickRank, for example, points to areas such as citation frequency, brand mention rate, share of voice within AI answers, AI-driven referral traffic and sentiment of brand references.

And all of this sits alongside core metrics such as subscriber lifetime value, churn rates, and time on site. Chartbeat’s data shows us that “your most valuable traffic source might already be on your site,” reminding us that while addressing shift in search matters, engagement with existing visitors remains a key area of focus. 

As part of this conversation, at an industry level, we also need to move beyond discussing a reduction in referrals to better understand its impact on revenue. We know that clicks are down, but we know much less about what that means for subscriptions and ad yields. Hopefully publishers are already joining up these dots internally, but a wider industry conversation about this would also be beneficial. 

The bottom line

Worries about Google Zero are well founded, although its impact is uneven. Publishers that are most exposed to this shift are typically those with the heaviest dependence on evergreen, easily summarized content and platform-dependent traffic. Meanwhile, those that are best positioned to navigate these changes, are those who produce content that needs to be seen onsite or in-app, and who already have strong direct and habitual relationships with audiences. Off-platform discovery is part of their playbook, but they are not reliant on it.

There’s no point asking whether AI-powered search will disrupt the traditional referral model. We know the answer. Subsequently, Google Zero encourages us to think about distribution and engagement strategies at a time when search traffic is less predictable and the economics are increasingly hard to quantify.

As Press Gazette describes it, search is not dead, it is fragmenting. What the most successful publishers understand is that Google Zero doesn’t require a single response. It requires a range of them. That includes potential partnerships with tech companies and AI providers, optimizing distinctive content for generative engine optimization (GEO), as well as doubling down on user experience within your properties.  

These media companies recognize that we have moved beyond clicks to a more fragmented and distributed media ecosystem, one where value is defined not just by traffic, but by presence, influence, and direct relationships with audiences.