GEO: Just a passing trend or major challenge for marketing teams?

For several months now, a buzzword has been creeping into almost every marketing meeting: GEO, for Generative Engine Optimization. We are promised a new territory to conquer within ChatGPT, Perplexity, and Google’s AI Overviews, complete with its own techniques and dedicated tools.

However, the topic is sparking debate within the SEO community. What triggered it? A viral exchange on the podcast The Edward Show, where consultant David Quaid challenged a staunch GEO advocate by asking him to provide concrete proof for his claims. So, is it a true revolution or just a marketing stunt?

The topic has ignited Reddit and divided professionals. This debate is worth examining because it touches on a fundamental question that our clients are also asking: is GEO a truly new discipline, or an evolution of the search engine optimization we already practice?

At the heart of the debate: what are the AI’s sources?

The basic principle is very simple: where does AI get its answers? It doesn’t invent anything. To answer you, it simply pulls information from traditional search engines like Google or Bing.

Result: for ChatGPT or Google AI to talk about your company, your site must already rank very well on traditional Google. Being cited by AI is simply proof that your usual SEO is working. This is why many experts doubt “GEO”: ultimately, there is nothing truly new about it.

The tools measuring your presence in AI are still unreliable

The second point of disagreement concerns the new tools claiming to measure your visibility in AI. Several experienced professionals are calling for caution, without completely writing off the entire sector. Their concern is easily understood. These tools ask a series of questions to AIs, then give you a visibility score. The problem is that this score is highly unstable, and the measurement KPIs it relies on can vary from one tool to another.

You only have to change a word in the question or switch to a different AI, and the result shifts completely. We therefore end up with a nice number presented as a serious indicator, even though we cannot really trust it. The right attitude is simple: some tools work diligently, but others are just guessing; in any case, these figures must be read with a great deal of perspective.

SEO and GEO are not enemies, they complement each other

Should we therefore bury GEO? No, that would be an exaggeration. The two words look alike and rely on the same technical foundations, but they do not serve quite the same purpose. SEO is primarily used to attract visitors to your site. GEO, on the other hand, mainly builds your brand awareness, which can subsequently drive visitors. We therefore see these two approaches as complementary, not competing.

The only truly useful thing to monitor for now is whether your brand appears or not in AI responses. And the good news is that some of this information is becoming more accessible: Google announced on June 3, 2026, the arrival of dedicated reports for generative AI performance in Search Console. However, this data should still be interpreted with caution, as rollout is gradual depending on markets and not all accounts have access yet. Bing Webmaster Tools can also provide useful signals, but Bing represents a much smaller share of the global search market, around 4 to 5%. This data must therefore serve as complementary indicators, not absolute truth.

Our agency perspective: focus on what really works

More than an SEO versus GEO duel, this topic questions our practices. Above all, it pits substantive work against grand promises. The reality—somewhat uncomfortable for those selling miracle cures—is that there are ultimately very few genuinely new techniques outside of the fundamentals of good search engine optimization.

What mainly needs adjusting is the strategic approach: prioritizing topics differently, tailoring content to be better understood, better cited, and better repurposed, and placing greater importance on certain well-known SEO optimizations. True new developments remain limited, sometimes even in beta or alpha testing, such as certain initiatives around llms.txt.

Here is what makes a difference today:

  • Relevant content: clear texts that answer users’ real questions.
  • Flawless technical execution: a well-built site that is easily crawled by Google.
  • A solid reputation: awareness and credibility built over time.

Rand Fishkin, founder of Moz and a prominent name in search engine optimization, puts it very simply:

“The best way to sell something is not to sell anything at all. You have to earn the awareness, respect, and trust of those who might buy.”

Put differently, no trendy tool or magic score will replace this foundational work. This is precisely where we come in: our role as an agency is to invest the client’s budget in concrete actions, not in useless theoretical reports.

Bad advice is costly, regardless of the label you give it:

  • Case #1: A company wanted to pay for a “GEO audit” when its site didn’t even appear on Google. Problem: if search engines can’t find you, AIs won’t find you either. Good SEO remains the essential prerequisite.

  • Case #2: A brand was advised to copy dozens of identical pages just by changing the city name. An outdated strategy that brings zero value.

GEO or SEO, the rule doesn’t change: bad advice remains bad advice. And it’s always the poorly informed companies that pay the price.

What we recommend today

Our advice is simple. As long as the vast majority of searches still take place on Google, it seems more strategic to first consolidate SEO foundations while gradually integrating levers related to mentions on ChatGPT, Perplexity, and other AI engines. Keep an eye on your presence in AI responses, because it is a useful and easy signal to track, but without obsessing over it.

At the end of the day, the real question is what investor profile you want to adopt: a defensive approach, based on proven practices and long-term benefits, or a more offensive approach, with higher risk, to test newer GEO signals and capture potential quick wins.

The decision also depends on the data you want to rely on: historical, solid, and qualitative SEO data, or more dynamic signals related to citability and brand presence in generative responses. Are we looking to generate clicks or impressions? To capture traffic or strengthen brand presence?

In any case, a website optimization mandate remains the safest foundation, because content quality, structure, and clarity serve both SEO and GEO. Focus your efforts where they truly pay off: clear content, genuine expertise published on your site, and a good online reputation. AIs will continue to evolve, and measurement tools will improve. In the meantime, the best way to be visible in AI remains simply excellent search engine optimization.

When Wikipedia arrived in 2001, we weren't stopped from using it. Above all, we were taught to use it with discernment: verify sources, cross-reference information, understand the tool's limitations, and maintain a critical eye. This is a mindset that still seems very relevant today, particularly with GEO and the evolution of search driven by AI.

Our role is to anticipate these innovations that are redefining visibility, but experience teaches us not to confuse novelty with strategic priority. Being innovative doesn't mean chasing every signal, but knowing which ones to test, measure, and explain accurately to our clients. GEO is an exciting evolution of SEO: let's approach it with curiosity and lucidity, without overexcitement, but without standing still.

Roland Merelle
Author of this article and Director of SEO Marketing at Dialekta

Want to learn more?

Join us for our breakfast conference specific to this topic on September 16, 2026