These Are the Earned Media Assets That Impact GEO Most

As AI search grows in importance as a channel for learning about business solutions, marketers are expected to ensure that AI describes our brands accurately, positively, and prominently in responses to relevant questions. That, in turn, means making sure that your brand is mentioned accurately, positively, and prominently in the places where LLMs prefer to look for information.

PR hits, also known as earned media brand mentions, play a key part in this mission. Research from Edelman found that as much as 90% of AI-powered search citations derive from earned media, with contextually relevant mentions in trusted media outlets and authoritative content pulling more weight than the traditional keywords and backlinks.

As we’ll see, generative engine optimization (GEO) requires landing specific types of content, formatted in particular ways, on certain platforms and publications.

Why AI Visibility Matters

The rise of conversational AI and AI-powered search has changed the face of digital marketing. AI engines like ChatGPT, Perplexity, Google AI Overviews, and Bing Copilot have replaced the traditional SERP list with cogent, concise answers.

What’s more, users increasingly trust these replies as knowledgeable recommendations. Attest’s 2025 Consumer Adoption of AI Report found that only 17% of users trust AI answers less than those from traditional search engines, and 40% trust them more. 

Source: Attest

Buyers rely on AI for brand recommendations, pre-purchase research, information about features and capabilities, and more. This is evident in the rise of zero-click searches.

When they’re learning about the various solutions offered by vendors, today’s prospects don’t necessarily encounter alternative views or check other sources. They’ll ask follow-up questions and eventually move from an AI-generated response straight to a brand website – could be yours or could be a competitor’s – giving AI answers a lot of impact on customer decision-making.

Traditional search occupies a single point in a buyer journey that leads people to check out many channels and platforms. AI search, in contrast, is increasingly located in a single place: the AI search platform or answer engine. This is where prospects keep asking more questions, iterating on their prompts, perusing AI-generated comparison tables, or running deep research dives based on their pain points.

Note that AI searches tend to be ongoing conversations. Visitors are more engaged, with sessions lasting an average of 8 to 13 minutes, far longer than quick hit-and-run Google searches.

A buyer can move from the top to the bottom of the sales funnel in just one session in AI Mode, ChatGPT, or Perplexity, without the brand having had any direct voice or any opportunity to capture or nurture a lead. As a result, someone who ends up on your website after a conversation with an AI engine is likely already primed to journey further, because the AI has pre-qualified you as credible and relevant.

This is good news in many ways, but only if you, your marketing team, and your PR reps know how to optimize earned media activity for how AI search works. Some PR mentions are more impactful for GEO than others, and it’s all evolving extremely quickly.

Maximizing the impact of your PR campaign strategy requires different tactics from optimizing a PR campaign for human audiences, and it’s even more different from SEO link-building campaigns. Instead of trying to directly reach and convince your ideal customers that you’re an authority in your space, you’re trying to send signals to AI engines that inform how they answer questions related to your space.

This requires working with contributors, platforms, and publication-side editorial teams to produce the types of content assets that AI likes. If your agency knows how to land these assets in the right locations, while leveraging the right voice and formatting, then you’ll be primed to surface in the AI answers that matter most to your business.

Content That’s Published on High-Authority Platforms

While owned media certainly shows up quite a bit in the sources cited in AI search answers, multiple studies have found that AI prefers authority publications, such as Forbes, Reuters, or The Guardian, over your brand’s product pages or blog posts. Depending on the company you’re promoting, these publications might be out of reach, but let’s remember that the clickstream data that informs these studies encompasses all types of AI prompts – not just those relating to business solutions.

Source: SE Ranking

In this sense, niche trade publications and local business journals can wield disproportionate influence, such as MarTech for marketing, Dark Reading for cybersecurity, or InfoWorld for data analytics. For example, if your expert commentary appears in SC Media, AI learns that your brand is authoritative when it comes to cybersecurity. It associates your brand with the topic, making the AI more likely to suggest your brand when someone asks for recommendations for a cybersecurity vendor.

This is why links don’t really matter when it comes to GEO. The web of mentions creates context that affects how AI interprets signals about your brand, whether or not those mentions are hyperlinked using keywords in the anchor texts. Just make sure that the publications your PR team targets are open to AI crawlers, that the content isn’t gated behind a lead capture form (although paywalled top-tier media is fine, apparently), and there are no disallow designations in the robots.txt. 

AI also takes user reviews, opinions, and mentions on social media and knowledge platforms very seriously. One study that explored citations by the leading AI answer engines found that knowledge platforms reign supreme. Among the top ten most cited sources, Reddit comes up more frequently than Forbes or Gartner. Google’s AI Overviews, for example, cites Reddit, YouTube, Quora, and LinkedIn more often than Forbes and Business Insider.

These platforms have authority for GEO purposes, even though many of them aren’t traditionally considered “earned media.” According to Gini Dietrich’s PESO model of the marketing media mix, they are either paid (as in the case of sponsored influencer videos on YouTube, for example) or shared (as in the case of brand-authored posts on Medium or customer community discussions on Reddit).

But there’s no arguing with AI answers, and GEO-ready digital PR firms know how to secure steady flows of contextually relevant mentions on all of these platforms as part of a holistic strategy.

Listicles, Product-Focused Features, and Expert Commentary

If you want AI to recommend your brand based on your niche expertise or as a leader in its field, to accurately explain your product differentiation, and/or to showcase the love your customers have for it, then you have to bring those narratives to AI’s attention. And that requires packaging these narratives in content that AI can easily make sense of.

Setting aside thought leadership for a moment, when it comes to recommending what makes your product special, product-focused content rules the GEO roost. This encompasses product comparison articles, product-focused feature articles, and “top product” category lists wherever they are hosted, including authentic peer product review platforms.

Listicles, which have been derided for years, are tops when it comes to influencing AI answers. Research by xFunnel into over 14 million AI citations found that 250 pages make up more than 40% of total citations. Of those top 250 pages, 38% are listicles. 

Source: XFunnel

Original research and data-led content is also valuable. Journalists particularly appreciate data-led content, which means that those types of assets are more likely to get published in the authoritative third-party publications that AI favors as sources. Diverse publishers and writers will also cite your original research, which contributes to the web of mentions that convinces AI of your authority.

Bylined thought leadership op-eds and expert commentary interviews still play a significant role in driving prominence in AI search results. These types of pieces showcase your knowledge and depth of understanding, helping AI view you as experts in your niche.

Content That Is Authoritative and Easy to Summarize

Content written with definitive statements helps to show LLMs that your take is trustworthy, and this type of authoritative voice also helps the AI engines to summarise and repackage your words as factual answers. It’s generally best to format content in ways that make it easy for AI search to quote it. That means ordered lists with bullets, guides, glossaries and definitions, and FAQs, with section labels and schema markdown.

AI engines seek to answer the user’s question as closely as possible, so they favor content that closely matches people’s actual conversationally-phrased prompts. AI prompts are longer than yesterday’s longtail keywords, and they’re more likely to be part of an extended conversation. Kevin Indig’s research has found that the top 10% of content cited by LLMs is long-form, with high sentence and word counts.

As he explains it, that’s not because AI prefers longer content per se, but because the longer a piece is, the higher the chances that it meaningfully answers a specifically-worded prompt.

Source: Profound

AI engines also prefer content that is readable and easy to understand. Indig found that the content cited most by AI often has a high Flesch score, which indicates simple comprehensibility.

PR Strategy Is Changing Before Your Eyes

These early days of GEO have been described as a land-grab moment. The old search order is being shaken up, as AI search is calling the shots. Now is the time for brand marketers to establish authority under the new rules, before your competitors beat you to it. With a made-for-GEO PR strategy, you can establish a powerful footprint in AI answers.

About the Author

Ben Jacobson Chief Content Officer at InboundJunction

A seasoned marketing strategist, Ben Jacobson serves as the chief content officer at InboundJunction, a PR, SEO and GEO agency for growth-stage tech companies. His writing has appeared in SEMrush, HubSpot and The Next Web.
 

See full profile