{"id":90615,"date":"2026-06-30T09:15:00","date_gmt":"2026-06-30T07:15:00","guid":{"rendered":"https:\/\/www.humanlevel.com\/?p=90615"},"modified":"2026-06-29T12:27:58","modified_gmt":"2026-06-29T10:27:58","slug":"brand-reputation-seo-geo","status":"publish","type":"post","link":"https:\/\/www.humanlevel.com\/en\/blog\/digital-pr\/brand-reputation-seo-geo","title":{"rendered":"How brand reputation influences SEO\/GEO"},"content":{"rendered":"\n

Imagine you\u2019re in a meeting with colleagues from different departments. On one side of the table, the SEO team talks about crawling, indexing, domain authority and search rankings. On the other, the communications team talks about media coverage, key messages, spokespeople and corporate reputation. Both work for the same company and yet it can feel like they operate in parallel universes: different metrics, different tools, goals that rarely overlap in the same conversation. <\/p>\n\n\n\n

This is how these two worlds have historically worked in most organisations, and it makes perfect sense. But with the changes being driven by artificial intelligence, that separation now comes at a cost it didn\u2019t before.<\/strong> <\/p>\n\n\n\n

Two worlds that never spoke to each other<\/h2>\n\n\n\n

Traditional PR had its own logic: building institutional reputation, managing relationships with media and journalists, protecting the brand image in times of crisis, and projecting corporate messages coherently and consistently. Its reference metrics were share of voice, impressions and advertising value equivalency. SEO, meanwhile, lived inside the web: structure, content, load speed, link architecture. They were disciplines with different goals<\/strong>, different profiles and budgets that rarely overlapped. <\/p>\n\n\n\n

That separation was perfectly consistent with how the digital ecosystem worked until relatively recently. Search rankings depended almost exclusively on what happened within your own domain, while brand reputation was built in the media, press conferences, corporate events, industry awards and public opinion. Two different territories, two different teams, two different ways of measuring success that didn\u2019t need to coordinate to work. <\/p>\n\n\n\n\"\"\n\n\n\n

The moment paths cross<\/h2>\n\n\n\n

Google has been incorporating external reputation signals into its algorithms for years through the concept of EEAT<\/a>. What used to be SEO\u2019s exclusive territory\u2014ranking well in search results\u2014started to depend in part on what happens off-site: media mentions, verified user reviews, presence in recognised authoritative sources. <\/p>\n\n\n\n

And those signals have been fully cemented with the arrival of generative AI. According to Muck Rack\u2019s Generative Pulse report<\/a>, conducted in December 2025 and based on more than one million links cited by ChatGPT, Claude, Gemini and Perplexity, around 94% of the citations generated by language models come from unpaid sources, and close to 25% specifically from earned media<\/a>. When ChatGPT, Gemini or Perplexity answer a question about a company or a product category, almost everything they say is drawn from what others have written about that brand, not from what the brand says about itself. <\/p>\n\n\n\n

In fact, in a recent audit for a client, we counted how many of the sources ChatGPT cited when talking about their category came from their own website. The figure was 4%. The remaining 96% were pieces written by third parties: media, professional associations, user opinions\u2026 The operational consequence is straightforward: the client can devote all their resources to optimising their website, but the conversation about their brand is being built somewhere else. <\/p>\n\n\n\n

This makes the communications professional an actor with direct consequences for model responses<\/strong>, whether they know it or not. Whether they care or not. Every piece of coverage secured, every press release published, every spokesperson interview or industry analysis that features the brand is, involuntarily, feeding the image that AI builds and reproduces of that company. <\/p>\n\n\n\n

What AI does with your reputation<\/h2>\n\n\n\n

When someone asks a language model about a brand or a product category, the model synthesises what multiple independent external sources have said about that brand, without giving particular credibility to what the company says about itself. If several sources agree on an attribute, the model treats it as a verified fact<\/a>.<\/strong> If that attribute appears only on the company\u2019s own website, the model interprets it as self-interested information and tends not to repeat it.<\/p>\n\n\n\n

This has a practical consequence that any communications professional can easily picture. Think of an established brand in the mattress sector, known for the quality of its materials and a 10-year warranty. A competitor launches models that look similar at half the price. The media, especially during peak-demand periods like Black Friday or the Christmas campaign, start publishing comparisons in which that competitor consistently appears as \u201cthe affordable alternative\u201d to the leading brand. No one is lying, but the narrative that justifies the premium price is missing: material quality, durability, after-sales service\u2026 If the brand doesn\u2019t actively work on that narrative in those same outlets, the competitor fills the gap with the story they want to tell. <\/p>\n\n\n\n\"\"\n\n\n\n

What changes in a communications professional\u2019s day-to-day<\/h2>\n\n\n\n

All of this translates into something concrete: adding a new layer of awareness to decisions that are already made routinely,<\/strong> without needing to become an SEO expert or use keyword<\/a> analysis tools.<\/p>\n\n\n\n

Choosing which outlets to publish in becomes a key ranking decision: some outlets carry far more weight than others among the sources language models prioritise for certain categories. Crafting the message of a press release now has a double challenge: you need a headline that interests a journalist, and you also need to think about which attributes you want associated with the brand in the text that journalist will publish, because that text is what AI will read later. And if we coordinate key messages with the SEO team, we\u2019ll ensure the attributes we want to project appear consistently across all external sources, not just on the corporate website. <\/p>\n\n\n\n

It\u2019s about moving from thinking of the journalist as the only end recipient of the message to understanding that, once published, that message becomes part of the information ecosystem that AI feeds on.<\/p>\n\n\n\n

In practice, this changes everyday decisions. For example, when a client in banking and insurance asked us to prioritise between two outlets for a piece with proprietary data, we chose the one that was cited more often by Gemini and Perplexity when discussing financial products. Interestingly, the outlet with the smaller circulation carried more weight as an AI source, so the piece ultimately ran there. <\/p>\n\n\n\n

The metrics that are starting to matter<\/h2>\n\n\n\n

Communications professionals are used to measuring in terms of impressions, share of voice<\/em> or advertising value equivalency. But none of those metrics reflect whether AI is recommending your brand or not, or which attributes it\u2019s using. These are the new metrics starting to gain relevance: <\/p>\n\n\n\n