Ross Hudgens: GEO Is Becoming an Off-Site Marketing Problem

AI search visibility may increasingly depend on affiliate sites, YouTube creators, communities and third-party publishers, according to Ross Hudgens, who says marketers need to pay much closer attention to who the models actually cite.

For years, search marketers have largely worked from a familiar premise: make your site better, build some authority around it and Google will eventually reward you.

Speaking at BrightonSEO San Diego, Ross Hudgens laid out a case for thinking about GEO beyond the confines of your own website. If this is all kinda “deja vu” 2001 when the reality of Google pagerank shocked SEO’s into realizing that marketing is happening in content off your site that you may/may not own.

Rosses’ argument was fairly straightforward: if ChatGPT, Gemini and other AI systems are leaning heavily on 3rd sources when forming answers, then those sources are becoming part of your search strategy whether you control them or not.

And in some industries, they may actually matter as much, if not more than your own site.

Follow the citations, not just the rankings

Hudgens started with citation data across several markets, including finance, SaaS and consumer ecommerce. The heady results mix varies considerably by industry.

Finance appears heavily influenced by a relatively concentrated group of affiliate publishers. SaaS has its own ecosystem, with sites such as G2 and user-generated content playing an outsized role. Consumer markets are more scattered, with YouTube, Reddit, established publishers and niche review sites all appearing. That difference matters.

There isn’t one universal list of websites marketers need to get mentioned on for AI search. Instead, the job is to determine which sources repeatedly appear for the questions that matter in your particular market. And to me, that sounds suspiciously like an old school SEO lesson wearing new clothes: that is simply stated as: figure out where the authority lives.

The twist is that authority is no longer measured only by backlinks or organic rankings. We now also have to ask whether a source is repeatedly influencing AI-generated answers.

Affiliate marketing suddenly looks a lot like GEO

One of the more interesting parts of Hudgens’ presentation was the connection between affiliate marketing and AI citations (raise eyebrows – yes, I know).
In markets such as consumer finance, many of the sites being cited by AI systems already operate on an affiliate model. That creates a rather obvious incentive.

If a handful of publishers repeatedly influence how AI systems describe and recommend products, brands may have considerably more reason to develop relationships with those publishers. And this goes beyond simply getting another link.

Hudgens suggested that brands should identify the publishers already receiving citation visibility in their market and then determine whether affiliate, partnership or other commercial relationships make sense. Hmmm, this is a decidedly less sexy or romantic version of GEO.

Sometimes the answer to “How do we get mentioned by AI?” may simply be “work with the sites AI already trusts.”

YouTube may be the biggest missed opportunity

Like the previous article we posted, Hudgens repeatedly returned to YouTube. Affiliate and performance marketing teams have historically had a difficult relationship with video because attribution can be lousy. Someone watches a product comparison on a television, sees another video on a phone and eventually buys the product three days later from a laptop. Good luck putting that journey neatly into an attribution dashboard.

AI search changes some of that calculation. The goal isn’t necessarily getting the click directly from YouTube anymore. The video itself can become a source that influences other discovery systems. That potentially makes creators far more valuable.

Tutorials, product comparisons, “best X” videos and detailed explanations of complicated products can all provide the kind of material AI systems need when answering commercial questions. For complicated SaaS products in particular, Hudgens noted that good creator content can have another benefit: it helps prospective customers understand what the product actually does. So the value of YouTube may stretch well beyond direct conversions. It can educate the customer, influence AI systems and help define how a product is positioned in the market. To me, that is a powerful combination.

Stop treating every AI engine equally

There was also a useful warning for anyone using AI vis tracking tools. Most platforms make it easy to check a box beside ChatGPT, Gemini, Perplexity and other engines, then roll the results into one visibility number. Hudgens argued that this can produce some very misleading conclusions.

Beware the AI citation farms

Wherever marketers find a metric, someone will eventually build a business around manipulating it. AI citations appear unlikely to be an exception. Hudgens pointed to a growing collection of sites seemingly designed primarily to generate content that appears attractive to AI systems. Some may rank. Some may even show up frequently inside AI citation-tracking platforms. But that doesn’t necessarily make them good targets. His suggested stress test is remarkably close to traditional SEO due diligence: look at whether the page and domain have real traffic, real visibility and some evidence of an actual audience.

If a publisher has no meaningful organic presence, little traffic and no reason for a human being to visit it, its long-term value as an AI citation source should probably be viewed with considerable skepticism. The temptation will be to chase whatever URLs appear inside the latest GEO tool. That could become the AI-search equivalent of buying links from sites built solely to sell links. We have seen that movie before.

Brands need clearer positioning

Third-party citations only help so much if nobody can agree on what your company actually does. Hudgens also emphasized the importance of consistent positioning. If your website describes the company one way, affiliates describe it another way and review sites use completely different terminology, AI systems have to reconcile all of those descriptions.
Brands can make that job easier.

Instead of trying to become “best overall,” Hudgens suggested establishing much more specific positioning. Being consistently described as the best option for a particular customer, use case or market segment gives publishers and AI systems something much clearer to work with. That positioning can then be pushed into brand guidelines, affiliate materials, creator briefs and first-party content. It is basic marketing, but it takes on new importance when machines are trying to synthesize hundreds of references into a two-paragraph answer.

Reddit isn’t dead just because citations dropped

Reddit citation visibility has fluctuated, but Hudgens cautioned against interpreting that as a reason to abandon community marketing. If your customers are still on Reddit, Reddit still matters. The smarter play may be building enough real customer adoption that those customers eventually participate in conversations organically.

Brands can also study which Reddit discussions rank in traditional search and which topics continually surface around their category. The objective isn’t flooding Reddit with company mentions. It is understanding where category conversations are happening and making sure the company has enough genuine users and advocates to participate in them. That is a much slower strategy than posting a few manufactured comments. It is also much harder for competitors to copy! lol iykyk

GEO starts looking a lot like marketing

Perhaps the biggest takeaway from Hudgens’ session was how quickly GEO stops looking like a technical SEO discipline once you follow the citations upstream.

The conversation moves into affiliate marketing.

• Then influencer marketing.
• Then YouTube.
• Then public relations.
• Then community building.
• Then positioning.
• Then brand.

Which raises an interesting question about whether GEO ultimately becomes a standalone discipline at all. The companies that perform well in AI search may not be the ones that discover a secret optimization technique for ChatGPT. They may simply be the companies that are talked about, reviewed, compared, recommended and accurately described across the places AI systems already trust. For search marketers, that means the job is expanding again. We still need to optimize the website. But increasingly, we also need to optimize the world around it.

What search marketers should do now

Start by looking at the actual citations appearing across the commercial prompts that matter to your business.Identify the domains and YouTube creators that repeatedly appear. Separate genuine publishers and communities from sites that seem to exist primarily to manufacture AI visibility.

Look at whether your affiliate and partnership teams are already working with those sources.

• Then examine how those third parties describe your company.
  
• Are they using the positioning you want?
  
• Are they describing your products accurately?
  
• Are competitors consistently associated with a use case that you should own?
  
• Finally, don’t allow a single blended “AI visibility score” to make those decisions for you.

Different AI platforms have different audiences, citation behavior and market penetration.

The next phase of GEO may require considerably less obsession with the score at the top of the dashboard and considerably more attention to the sources underneath it.