Is GEO Real or The Great SEO Hallucination? The SEO Community Speaks!

Generative Engine Optimization (GEO) has moved from theory to playing and testing ground. We were curious to see where this was all going to land with marketers. What we found is that marketers are experimenting across the entire spectrum from ChatGPT, Bing Copilot, Perplexity, Gemini, and Google's AI Overviews trying to figure out what sticks and what potentially drives results. Unlike tried-n-true SEO, where rankings and CTRs could be mostly tracked with relative ease and clarity, GEO is best described as seriously messy and frustrating with mentions, citations, and brand visibility at-this-juncture being feel good metrics that no one can benchmark (yet).

To get a clearer picture, we asked practitioners across the industry how they're approaching GEO today. Are these early experiments producing ROI, or are we still in the spray-and-pray and the your-guess-is-as-good-as-mine stage?

The responses below show the divide: some see measurable gains in citations, referrals, and brand lift, while others argue GEO is snake oil dressed up with a new acronym.

Lets start with a couple of folks with data-driven wins to show that something is working and producing tangible results:

Victoria Shepherd Technical Program Manager at Healthcare Success says:

"Mentions in AI Overviews increase when schema, structured FAQs, and physician expertise align, yet CTRs remain below 5%. The “lost clicks” we see in GA4 (40–50% traffic declines) are often resurfacing as brand inclusions and citations in generative platforms. Perplexity and Bing Copilot deliver the strongest referral lift, while Google Overviews suppress clicks, and ChatGPT acts more as an influence channel via shareable links. At the same time, AI bots have overtaken Googlebot as the top crawlers on some healthcare sites, underscoring the shift toward zero-click discovery. Schema and entity mapping are proving decisive - fragmented markup undermines inclusion, while structured MedicalOrganization, Physician, and FAQPage schema consistently improve citations."

Steve Wiideman, of Wiideman Consulting Group:

"We worked with social teams and customer service to help amplify the visibility of passages from the website in social networks, forums and discussions, and within contributed content.The results, while still slower than one might expect, show an uptick in referral traffic from ChatGPT of 71.6% over a 90-day window. We also traffic from Reddit increased by 120%. Both destinations were sending converting traffic, not just random visitors.

It's still early, but we plan to see these tests and experiments run through to the end of the year before we make any conclusive decisions."

Nate Henry, VP Channel Birdeye, says that rigorous structured approaches and testing for local results:

"Search AI is giving us clarity on how location-based businesses show up in generative engines. We're now analyzing prompts and benchmarking visibility across ChatGPT, Gemini, Perplexity, and Bing.It's showing us which sources the engines cite, auditing accuracy and sentiment, and publishing AI-ready content. GEO requires the same rigor as SEO with structured content, expert sources, and quality reviews, but success is measured by how often AI models cite the local business in question (not by rankings)."

---

While others think there isn't much new here and it is best to proceed with caution:

Jeff Ferguson, Owner at Amplitude Digital and author Killing SEO

"There's nothing new to optimize for. AI Overviews pull from the same ecosystem. They're not some separate layer you need to ‘game' with new tools or tactics. At best, GEO is a placeholder for discussions we might be having later about prompt-level optimization, structured data for LLMs, or answer engine visibility, but right now? It's snake oil with a shiny acronym."

Ben Fisher of Steady Demand

"We get results so far for our test and clients by doing good regular SEO and some rep management types of activities. As wel as providing aggressive feedback on AIO's. Which one does not matter at this time; all of the same techniques work across all answer engines. Good old fashioned good business and good Seo still works."

---

This is the crux of the core technical problem all SEOs are wrestling with on a daily basis:

Dixon Jones, CEO of Inlinks Optimization LTD:

"We are getting results from GEO, because people on Demos are saying they found us by interrogating LLMs. But we need to find new ways to track, because Google Analytics is JavaScript and LLMs do not (yet) render JavaScript.This makes for a super interesting conundrum. Using Log Files was what we did back in the 2000s, but if you are behind a CDN nowadays, that will not work either. I think (for us) there is a possible solution, looking at logs in our CDN provider."

Nick Stamoulis, President of Brick Marketing

"...we are seeing branded searches and direct traffic increase across many of our b2b clients. This could be due to users not clicking on source links in the LLMs as they would through a traditional search. For tools, they are very basic and often revolve around potential content holes that tie into a digital brand presence, but they are not too helpful yet."

Dave Davies Head of SEO/GEO at Weights & Biases by Coreweave

"...at this stage I'm having difficulty tying causation to the visibility metrics I pull out of the tools I use (mainly Scrunch AI for GEO monitoring). My scenario is made more difficult than average by the fact that we have developers surfacing us through models via APIs in tools like Cursor AI, where we can be added to their systems before they've ever visited our site or signed up.While I'm definitely not going the "spray-and-pray" route, there is still a lot of testing and refining being done, and I don't expect that to ever end."

 

---

While the experts are obviously divided on the long-term impact of GEO, their experiments reveal several common threads. For SEOs looking to get ahead, here are a few key strategies emerging from the front lines:

• Lets Double Down on Structured Data: As Victoria Shepherd noted, comprehensive schema is crucial. Prioritize MedicalOrganization, Physician, FAQPage, LocalBusiness, and Product schema to provide clear, machine/bot-readable information.
  Even though we've had people saying LLMs don't uses schema directly, I look at it this way:

  "If the data is telling us that only 23% of the web is using Schema, and it also shows that 40% of the results in AIO's are using Schema, it is a safe bet that there is a filter, a ranking bonus, or cull being applied somewhere in the algo stack for structured data. So why wouldn't we use it?" - Brett Tabke"

• Embrace Entity Optimization: Dixon Jones is a proponent of this (disclosure, we have a side interest with Inlinks. which does not impact inclusion here).
  He suggests that we need to ensure your brand, people, and products use well-defined entities with clear connections across the web (e.g., via knowledge panels, Wikipedia, industry sites).
• Amplify Your Reach Beyond Google: Steve Wiideman's success with Reddit and social forums highlights a key shift.
  While panda and penguin updates destroyed most UGC plays, Google is now smitten with it again as generative engines source information from a very wide range of discussions and communities. You need to actively participate where your audience is at (with your brand if possible).
• Master Your Digital Brand Presence: Nick Stamoulis and Ben Fisher both allude to this. The rise in direct and branded searches means your overall brand reputation, reviews (as per Nate Henry), and off-site mentions are more important than ever.
• Provide Aggressive Feedback: A tactical tip from Ben Fisher. Use the feedback mechanisms within AI Overviews, Gemini, and other platforms to correct inaccuracies and reinforce your desired information. Those little thumbs up/down buttons are wired to something on the back end - lets literally poke those buttons and see if anything happens.

Perhaps the most significant challenge presented by GEO is the breakdown of our traditional analytic process. As Dixon Jones points out, server-side tracking is becoming necessary as LLMs  don't render the JavaScript required for tools like GA4. This creates a "black hole" where visibility doesn't translate to trackable sessions. Consequently, as SEOs we must adapt our KPIs. Success in GEO is not measured by rankings and CTRs, but by:

• Mentions & Citations: How often is your brand, product, or expert cited as a source? (ugh, yes I know it looks worthless on the outside looking - it really is - but its something.)
• Increased Branded Search & Direct Traffic: As users get answers and discover brands in AI Overviews, they are more likely to search for the brand directly or type the URL, as Nick Stamoulis observed.
• Referral Traffic from AI Favored Platforms: While difficult to track, platforms like Perplexity and ChatGPT can show up as referrers. Steve Wiideman's 71.6% increase from ChatGPT is a key early indicator. Indirect targets are once again in vouge. eg: backlinks still rule the web. See if you can land some today!

 

Whether GEO is a revolutionary new discipline or simply good SEO under a new name remains a point of contention. However, the evidence is clear: the path from discovery to conversion has been fundamentally rerouted. The decline in traditional CTRs, the rise of zero-click slop, and the new challenges in measurement are not theoretical. They are impacting traffic and strategy today. The SEOs who succeed will be those who embrace this messy new reality, shifting their focus from chasing clicks, backlinks, to building an unimpeachable, authoritative brand presence across the entire digital ecosystem. The game has changed and theplaying field is bigger than ever before. Step up and see where you land in the search engine world.