Prominence + The Machine: How to Create a GEO Strategy in an AI-dominated Media Landscape

In 1997 the ‘Deep Blue’ supercomputer beat reigning world champion, Garry Kasparov, at chess for the first time. This was a watershed moment in artificial intelligence, marking a symbolic shift where machines began to outperform humans in complex strategic tasks. Fast forward to 2026 and AI has changed the state of play in the comms industry.

So, how on earth are you meant to create a PR strategy in an AI media landscape?

Businesses now need to shift from prioritising SEO to becoming the cited answer for AI tools like ChatGPT, perplexity or Gemini. The goal is no longer just traffic; it’s citation and inclusion. But what does this actually look like in terms of building a PR strategy?

Here’s our advice for building your PR strategy in an AI world.

Identify Content Opportunities

To find where you can provide insight, you need to understand where AI currently lacks information. To do this, you can prompt various LLMs with industry specific “how to” or “best of” queries. Where the AI gives vague, outdated, or incorrect answers, there is a content vacuum for you and your PR team to fill.

You can also try source mapping, where you use specialised GEO tools to see which domains are being cited in AI overviews for your target keywords. If a competitor is cited, try to analyse why. It could be because of a specific whitepaper, content in a trade publication, or a Reddit thread. Once you figure it out, adapt and improve on their content.

Finally, you can use social listening to identify emerging terms and trends that haven’t hit mainstream media yet, and coin them in your content. If you are the first to define a new industry term (e.g. AI slop), AI models will link that definition to your brand as the originating authority.

Tailor Your PR Content for LLMs

Once you’ve identified content opportunities, you need to draft your content differently from traditional PR because your primary audience isn’t just human, it’s the training data and retrieval systems of AI models. LLMs prioritise content that is easy to summarise and provides new information, so you’ve got to add unique and digestible data or insights in your content.

It’s also important to consider how you structure your content for LLMs. When writing, place the direct answer to a potential prompt in the first sentence of a section, followed by supporting evidence and expert quotes. This format lends itself to the Q&A nature of LLMs.

You should also plan content to write using topical clusters. Instead of simply writing about the subject you want to be the authority on, write a variety of content around the central pillar, creating an interconnected network which establishes you and your business as the topical authority. For example, instead of writing an article on ‘How to become a PR thought leader’, you should write articles on ‘An introduction to PR and GEO’, ‘GEO Isn’t Broken; But Your Website Might Be’ and ‘How to create a GEO strategy in an AI-dominated media landscape’.

Semantic & Fact Density (The Accuracy)

LLMs look for fact-dense passages to ground their answer to so it’s important to cite high-authority sources. An easy way to do this is to prioritise newsjacking your competitors’ new reports or reputable data announcements (such as from the ONS) with your own additional insights to build your credibility.

As with traditional PR, it’s important to use industry-specific terminology and statistics. Research shows that data-heavy content with specific numbers has a higher inclusion rate in AI summaries. AI models also love structured data and certified statistics, so if you release an annual “state of the industry” report with unique data points, it will more than likely be scraped and referenced by AI models.

Brand Footprint & Earned Media (The Reputation)

Similarly to traditional PR strategies, AI models don’t just look at your website; they look at the sentiment of the entire web regarding your brand. So, it’s important to build a strategy that focuses on earned media. You can do this through classic digital PR such as thought leadership articles, press releases, and case studies.

Where AI differs from traditional PR strategies is its focus on community engagement. LLMs are increasingly trained on Reddit and niche forums because authentic brand mentions in these communities serve as “social proof” for AI. You could build this into your strategy by encouraging loyal customers to review your brand on Google and tasking your comms team to review and respond to relevant Reddit threads like they do for customer queries on social media platforms.

Measuring Success in a “Zero-Click” World

To know if your strategy is working, you’re also going to have to create a strategy for measuring success. Traditional SEO metrics like “Clicks” and “Impressions” are becoming less relevant because most users no longer need to click on websites or articles to get the information they need. So, for GEO, you should track the Share of Model (SoM) instead. This tracks how often your brand is mentioned when a user asks an engine for a recommendation in your category. When doing this, it’s also important to measure citation quality to make sure you’re being linked as a primary source as opposed to a passing reference.

And finally, you need to check that there’s sentiment alignment and the AI describing your product accurately and in line with your brand’s core values.
If you can do all of this, then you’ll be working with Deep Blue, instead of fighting against it.

If you liked this article, look out for our future PR and AI blogs at Rooster.co.uk.