SEO vs. GEO: Which to Prioritize?

I know of no definitive research showing whether generative AI platforms are replacing Google Search or how they affect buying decisions.

ChatGPT sends very little traffic to websites: 0.43% overall, per Ahrefs. That percentage aligns with my own data, which found the highest share, at 0.70%, going to technology sites.

Given genAI’s minuscule traffic impact, should businesses invest in optimizing it?

Search optimizer Marie Haynes says no in her article “GEO is a distraction.” Search Engine Roundtable summarized her study and those of two others.

Recommended

The Race to Own AI Shopping

September 6, 2026

AI Visibility?

A common defense of GEO claims AI visibility is not measured by clicks alone. I agree. Traffic is not a valid key performance indicator for a channel designed to provide answers, not clicks.

Yet we have no reliable metrics to gauge the impact of visibility alone. Yes, consumers presumably research purchase decisions on genAI platforms, but don’t ask me to prove it.

I can cite a combination of metrics, but none provide a clear picture.

• AI visibility scores from tools such as Profound and PeecAI rely heavily on prompts. For example, a prompt that includes a brand’s name will almost certainly include that company in the response. Ditto for niche queries. My name is rarely cited in prompts for “top seo experts.” But “top female SEO experts” typically includes it.

• AI traffic. Comparing the performance of a 30.0% traffic channel to one at 0.4% is fruitless since the numbers are too low for meaningful conclusions.

• Branded queries. I’m guilty of asserting that genAI can increase brand-name searches on traditional search engines. An entrepreneur may search “top female SEO experts,” see my name in the response cited but not linked, and then search “Ann Smarty” on Google. Yet even that is difficult to measure owing to the variety (high) and volume (low) of traditional brand-name searches.

Hence many businesses now ask customers in post-purchase surveys, “Where did you hear about us?” But that, too, is imperfect, as buyers with extended journeys may not recall a genAI mention.

Given the fuzzy attribution options, how can businesses justify GEO spend?

Here’s my take.

Prioritize SEO

Traditional search engine optimization is the priority, for two reasons.

Google is the only intent-based, measurable organic search channel. GEO metrics remain a black box.

And organic search visibility is the most predictable route to AI visibility. The correlation between top-ranking URLs in Google and top-cited URLs in genAI platforms varies. ChatGPT tends to cite large publications (many are OpenAI partners) and user-generated content (Reddit), while Gemini and AI Mode rely on Google’s index. Yet studies conflict as to what that correlation is, and the impact on genAI citations of high Google rankings versus low.

But know this: I’ve seen no highly ranked site on Google with no visibility on ChatGPT. For that reason alone, organic search should remain the priority until we have proof that AI visibility materially drives purchases.

GEO Next

I’ve come to believe GEO does differ somewhat from SEO.

• GenAI crawlers are less advanced than Google’s. Thus clean and clear websites readable without JavaScript are more important for GEO.

• GenAI platforms tend to seek diverse sources when answering prompts. Visibility across channels matters more for GEO, especially when the brand messaging is consistent and user feedback is positive.

I’m not pursuing GEO-only tactics, however. I can’t measure performance even if successful. Plus the visibility process of genAI platforms varies — there’s no straightforward way to be visible everywhere.

Everything I do for GEO results in additional benefits. For example, ChatGPT frequently cites Reddit threads, but more importantly, brand awareness on Reddit drives traffic and conversions — with or without ChatGPT.

Likewise, genAI platforms tend to cite list-type articles. But inclusion in high-authority listicles also spurs traffic and conversions.