For months, the search industry spoke as though the break had already happened. SEO was the old internet. GEO was the AI era. Traditional rankings were supposedly fading out, AI citations were moving in, and any business slow to pivot would quietly vanish from the next generation of search. The confidence behind those claims? It travelled a lot faster than the actual evidence did.
Then Neil Patel asked a question on X that pulled the conversation somewhere less comfortable: “Does more SEO traffic mean more GEO traffic?” The replies did not read like a clean consensus forming. They read more like marketers slowly realising the systems they had been separating into different budgets and different teams might still be pulling from the same foundations underneath.
Key Points
- Neil Patel shared findings from research covering more than 384 websites, looking at whether stronger SEO traffic also correlates with stronger GEO traffic and AI citation visibility.
- Marketers joining the discussion argued that AI systems increasingly reward trusted, decision-support content, authority signals, and expertise depth, rather than informational content built purely for rankings.
- Google’s own ranking systems documentation and the AI Overviews rollout have reinforced a growing belief that AI discovery still leans heavily on relevance, reliability, helpfulness, and original content infrastructure.
Neil Patel’s Post Landed While the Industry Was Already Reshaping Itself Around AI
That context explains why the reaction moved so fast. GEO conversations have been accelerating quicker than most businesses can actually keep up with. Agencies introduced AI search services almost overnight. Publishers started rebuilding content strategies around AI discoverability. “LLMO” entered marketing vocabulary, riding alongside predictions that traditional SEO would eventually lose its relevance as AI-generated answers absorbed more and more search behaviour.
Patel’s post cut past most of that noise.
“Although there are differences between SEO and GEO, there are also many similarities. And technically more similarities than differences.”
From there, the conversation moved straight into data:
“So we looked at 300+ sites to see whether more SEO traffic means more GEO traffic.”
That single line shifted the mood immediately, because it went after one of the quieter assumptions driving the whole AI-search conversation, the belief that AI discovery systems are breaking away from the authority models traditional search was built on.
A translated response referencing NP Digital research covering 384 websites pressed the same point even harder:
“In NP Digital’s survey of 384 sites, a correlation emerged: the more SEO traffic a site has, the higher its AI citation rate. The content assets you’ve built up with SEO aren’t wasted.”
That finding spread quickly, and not just because it was interesting. It spread because many businesses had already started making strategic decisions as though the opposite were true.
Some pulled back on traditional SEO infrastructure. Others pushed resources hard toward AI optimisation without fully understanding what AI systems actually and consistently reward. Visibility strategies started splitting into separate lanes, even as the systems underneath them kept pointing toward the same foundations explored in the Princeton GEO research paper.
The Conversation Moved Fast Toward Trust and Recommendation Behaviour
What stood out beneath Patel’s post was the speed at which marketers stopped talking purely about rankings and traffic. The discussion shifted toward recommendation systems.
Shel Welker put the difference plainly:
“SEO visibility still is a key component to getting GEO citations. If your content only answers questions, AI will reference you. If it supports decisions, it will recommend you.”
That difference carries real weight. Recommendation systems operate under different pressures than traditional indexing systems ever did. Search engines historically rewarded discoverability. AI systems appear increasingly tied to confidence. A generated answer does not simply pull up a page; it compresses research, filters out uncertainty, and publicly connects information to sources the system seems willing to stake its credibility on.
That changes what kind of content actually matters.
Informational content can still generate visibility. But decision-support content, comparisons, implementation guidance, contextual expertise, reliable explanations, all of it appears to matter more now because AI systems need stronger confidence before surfacing recommendations directly inside generated responses.
One reply beneath the discussion compressed the entire argument into a single line:
“The overlap between SEO and GEO is bigger than most think.”
The conversation moved on without dwelling on it, but that line kept echoing.
Search Intent and Authority Signals Were Never Really Gone
Part of the confusion around GEO comes from how aggressively some marketers framed AI visibility as an entirely new discipline, something fully disconnected from traditional search behaviour.
Neil Patel’s own on-page SEO guide tells a different story. His writing on on-page SEO describes search intent as the foundation of optimisation and argues that relevance stays critical because search engines are trying to surface “the results that offer the most value.” The same material stresses that trust matters to both users and search engines, and that content needs to be “authentic, accurate, and relevant.”
Those principles now sit directly inside the GEO conversation whether marketers choose to frame them that way or not.
Patel also describes how quality content helps build “authority and expertise,” keeps readers engaged longer, and increases the chance of earning backlinks. That reasoning sounds remarkably close to how AI recommendation systems appear to judge source confidence.
Mechanics change. The pressure toward trust, relevance, expertise, and contextual usefulness keeps surfacing underneath both conversations anyway, especially as Google’s emphasis on E-E-A-T principles becomes more central to how trustworthy information is surfaced.
Google’s Own Documentation Still Comes Back to Reliability
Google’s broader ranking infrastructure points in the same direction. The company says its systems evaluate “many factors and signals” across “hundreds of billions of web pages” to surface useful and relevant results.
Those systems include:
- PageRank
- RankBrain
- BERT
- Neural matching
- Passage ranking
- Freshness systems
- Original content systems
- Spam detection systems
- Reliable information systems
- Review systems
Google also explains in its How Search Works documentation that its systems work to surface “authoritative pages,” lift “quality journalism,” and give original reporting more visible placement in results.
Separating that from the GEO conversation becomes difficult, because many AI-search discussions still treat AI discovery as though it operates completely outside traditional authority infrastructure.
The systems themselves keep changing. But the emphasis on relevance, helpfulness, originality, and reliability keeps showing up almost everywhere inside Google’s own explanations, including its Search Quality Rater Guidelines overview and documentation around the Helpful Content Update.
AI Overviews Are Already Changing How People Discover Content
Google’s rollout of AI Overviews made the shift easier to observe. The company describes them in its official AI Overviews documentation as AI-generated snapshots designed to help users “quickly understand information from a range of sources.”
Traditional web links still appear beneath those summaries, but the interface compresses discovery immediately. Search journeys get shorter. Fewer links absorb attention. Recommendation visibility narrows into smaller and smaller spaces.
Google also warns users that “AI responses may include mistakes.”
That quiet disclaimer actually explains something important; authority signals may become more critical inside AI-driven search systems, not less. Generated answers still lean heavily on reliable underlying sources. Recommendation systems cannot keep surfacing weak or inaccurate information without eventually damaging confidence in the interface itself.
David Perlov described the tension this way:
“Overlap makes sense if you treat both engines as the same recommendation problem with different surface areas. The variable that matters isn’t traffic correlation, it’s intent depth. GEO surfaces a smaller cohort but they’re already past the ‘is this real’ question…”
That smaller cohort may end up mattering more commercially than raw traffic volume, especially if AI systems absorb more of the trust-building process before a user ever clicks through to anything.
The broader mechanics behind that shift are increasingly being discussed across the industry through concepts like Generative Engine Optimization (GEO). The core question is no longer only where a page ranks, but whether a brand or source is trusted enough to appear inside AI-generated answers.
The Industry May Have Built a Divide That Was Never Really There
The budget discussion beneath Patel’s post exposed a different kind of problem. Many businesses had already split GEO and SEO into competing internal priorities.
Pawan Singh addressed it directly:
“If the data shows SEO traffic and GEO traffic moving together then every budget conversation treating them as competing priorities is solving a problem that does not exist while creating one that does.”
That line landed because the behaviour it describes is already widespread.
Some organisations rushed toward AI publishing systems. Others treated GEO as a standalone workflow, detached from broader SEO infrastructure. In certain companies, visibility teams now operate with entirely different assumptions about what AI systems reward, even as they chase the exact same outcome.
Another translated response pushed back against one of the louder narratives circulating around AI search:
“The claim that ‘SEO is dead, from now on it’s LLMO’ is half wrong.”
Half wrong probably describes the industry’s current state better than most of the predictions going around.
Nobody serious believes AI search changes nothing. At the same time, evidence keeps building that AI systems may still inherit many of the same authority signals traditional search engines were rewarding long before GEO even entered the marketing vocabulary.
What this means for SEO and GEO strategy
The key takeaway is not that SEO and GEO are the same. It is that AI visibility appears to build on many of the same foundations that already support strong SEO: trust, authority, relevance, expertise and useful content.
That matters for strategy. Businesses with strong topical authority, trusted content and mature SEO infrastructure may already have an advantage in AI search. Their existing signals, from backlinks to entity recognition and original content, may help AI systems decide which sources to cite or recommend.
The risk is higher for publishers built on large volumes of shallow content. AI-generated answers compress visibility into fewer sources, which means thin pages created only for keyword coverage may struggle.
This connects to a broader debate we covered in our analysis of Google’s claim that AI search optimization is still SEO. AI search may change the surface, but the underlying need for trusted, useful and well-structured content has not disappeared.
For search teams, the practical move is not to split SEO and GEO into separate silos. It is to build content and authority systems that can work across both.
The next phase of search may reward the sites with the strongest trust infrastructure, not simply the most content.
