GEO / SAIO: Everything About SEO for AI Search Engines

AI is dramatically changing the way we search the internet. Website operators must ask themselves: How does SEO work for ChatGPT, Perplexity, and others?

How SEO Used to Work

Anyone who is no longer a complete novice to the internet knows how a search used to be conducted: You would enter a search term (in technical jargon: keyword) or a combination of terms into a search engine like Google, Bing, Ecosia, or DuckDuckGo and receive a long list of more or less relevant search results. SEO is the art of optimizing your content for these search queries to rank as high as possible on the list.

There has always been a small group of users who, instead of entering a few keywords like “Hotel Munich,” would input a whole question, such as “Which hotel in Munich is close to the Olympic Center?” However, such queries are relatively rare in traditional searches. We SEO professionals refer to them as “long-tail” search queries, and if you happen to run a hotel near the Olympic Center, considering such search queries could be worthwhile for you.

The Revolution of AI Search Engines and the Importance of GEO / SAIO

With the growing popularity of dedicated AI search engines like Perplexity or the new Google AI, as well as generative AIs like ChatGPT that also search the internet, search behavior is changing.

What is particularly noteworthy is the speed at which this change is occurring. In some analyses and through targeted inquiries with our client websites, we are seeing more than 30-40% of referrals coming from AI search engines—this trend is strongly increasing. This is why optimizing for queries in AI search engines is so relevant, whether you call it GEO (“Generative Engine Optimization”) or SAIO (“Search Artificial Intelligence Optimization”).

The Difference Between Traditional Search and AI Search Engines

There are numerous differences between traditional searches and queries made through AI search engines. Here are some key points:

• Longer Search Queries: Search queries are significantly longer, ranging from 3-5 words in traditional searches to 20+ words in AI queries (without follow-up questions!).
• Optimization Goals: The goal of optimization is no longer to land in the top 3 search results but to gain as many source citations as possible, linking from AI results to your own site. Therefore, it is no longer about ranking in search results or how often you are listed or clicked (measured by the so-called click-per-impression rate), but rather about visibility, positive mentions of your brand, and the clicks that result from them.
• Changing Evaluation and Ranking Factors: AI search engines perceive websites differently. While backlinks, keywords, and traffic were once important factors, today it’s about relevance, source authority, readability, word count, and content freshness.
• Visitor Intent: When visitors click on a link to a website, their intent today is different than before. Previously, they wanted to learn more; now they have already read the important information and want to take targeted action. This also means that the conversion rate for visitors from AI search engines is significantly higher.

Concrete Tips for Successful GEO / SAIO

What follows from these differences for successfully positioning your brand in AI queries? How can you effectively implement good GEO / SAIO?

The Key: What Does the Visitor Need?

We have been preaching this for ages, but regarding AI search engines, the topic gains new relevance: What does the customer search for and need? AI search engines have a much more precise “sense” of this than Google & Co. ever did. It is therefore crucial that your website meets these needs and answers the questions. If it does so in a relevant and high-quality manner, you will be provided with fewer but more targeted and highly motivated visitors.

Relevant and Authoritative Content

To appear as an authority and source for a specific search query in AI search results, it is no longer sufficient to optimize a subpage for one specific keyword. Instead, traditional content marketing approaches come into play: Write truly (!) relevant content! This means:

• Comprehensive Content Covering All Aspects: AIs perceive context. By thoroughly addressing all possible questions and points on a topic, the likelihood of being cited and linked in search queries increases. Dedicated FAQ blocks are also a valuable tool here.
• Smart Website Structure with Pillar and Cluster Content: A structure that works with a “Hub-and-Spoke Model” is much more effective for AI queries, where a comprehensive topic page links to various relevant subpages.
• No Traditional SEO Tricks: Avoid artificially inflating keyword usage to create relevance (“keyword stuffing”) or artificial optimization that offers no value to the visitor.
• Clear, Readable, and Understandable Language.
• Clear Structure on Pages: Organize content with headings, paragraphs, lists, formatting, etc.
• Media and More: Ensure all content is easy to read and process, including media, tables, etc., using alt texts, titles, and captions.
• No AI for Content Creation: This point is still debated, but it seems that AI search engines prefer original, authentically human content over AI-generated content.

Developing a Clear Brand

AIs do not view a website through individual URLs; they also assess the coherence and clarity of statements across all pages. Therefore, it is particularly important for AI search queries to sharpen your brand and communicate consistently and coherently across all channels.

Adapting Your Website for High-Intent Users

As mentioned above, users coming from traditional result lists differ significantly from those directed to your website by an AI search result. The former have little context and visit your site to learn more. The latter have already received the entire context from the AI. If they click through, it indicates that their intent is now focused on taking action.

This means that for conversions on your website, you should place greater emphasis on clear calls to action (CTA) and a smooth transaction process, rather than on basic persuasion.

No Technical Barriers

When the content is appropriate, you must ensure that AI search engines can find and evaluate it. These topics are also found in traditional SEO, but they remain important:

• No exclusion of AI in the robots.txt file
• Clear schema.org specifications on the website
• No client-side JS rendering
• Setting self-referential canonical tags
• Smart internal links with meaningful anchor texts

Looking Beyond Your Own Website

AIs assign authority to a website, but they do so not only based on the content on the website itself but also across the entire internet. This is similar to traditional backlink building but has a different focus. Sources with high authority that AIs prefer to “tap into” include Wikipedia, YouTube, Quora, Reddit, LinkedIn, and Forbes. Many other sources that were traditionally significant for backlinks (e.g., news portals) often exclude AIs from their pages and are therefore less attractive in GEO.

Conclusion and Remaining Questions About GEO / SAIO

The technology of AI search engines is still new, and emerging companies are trying to compete against giants like Google or Microsoft. The entire landscape remains volatile, and one must expect significant changes and disruptions. In particular, it remains to be seen whether and how one will have to pay for mentions in AI search results in the future.

If you have previously oriented yourself to the standard of presenting genuinely visitor-focused, relevant content and an authentic brand, you will likely be less caught off guard by the AI wave. However, if you have relied more on traditional SEO tricks without optimizing content for visitors, it will certainly be more challenging to bridge the gap between impressions and clicks that AI search engines create.

Helpful Resources

For further exploration of the topic, I recommend the excellent checklist by Aleyda Solis [Google Docs].

Peter Schnoor has been working as a web designer and web developer for over 20 years. As the founder of Netjutant, he helps clients communicate more authentically and successfully online. To balance his screen work, he enjoys nature in his home region of Upper Bavaria.