GEO – AI SEO
OpenAI’s ChatGPT gained 100 million users in two months making it the fastest growing consumer product ever.
Google has gone all in on AI with AI Overviews and AI Mode in search, and with Gemini, their standalone chat/LLM interface. And, Google is not new to AI as you’ll learn if you read on.
In the not too distant future, search as we know it today will become the interface for intelligent agents who don’t just answer questions, but get jobs done for us.
Polls suggest that users love the quick answers and the ability for AI Search to help them get jobs done and answer questions. So they’re not going away. AI will become pervasive in all our software products and in the way we approach information and commerce.
In fact, in a recent study we sponsored with UPCEA, a majority of respondents indicated they were more likely to trust a business they could find in Google’s AI Overviews and other generative search engines.
What is AI SEO?
AI SEO is the work we have to do to assure you are recognized by the AI engines for what you do, where you do it, and for whom.
AI SEO is not that much different than traditional SEO if you’ve been doing SEO well for the last decade. However, the way AI Chat and generative search work makes technical SEO much more important than ever before.
Another great way to think of AI SEO is as “Semantic SEO.” The launch of Google’s knowledge graph in 2012 got the SEO community to start thinking of Things vs. Strings – a fundamental shift that most SEO professionals are still adapting to.
Strings, or the search queries that humans type into search engines require lexical analysis. You have to understand the language and nuance.
Things are defined, unambiguous entities. An Apple computer is not the same as a Honeycrisp apple even though they are lexically similar.
What is AI-Powered SEO?
AI-Powered SEO encompasses the AI tools and automation we are using to help us do SEO better.
Using proprietary and commercially available SEO software, we are revolutionizing the way we approach SEO work. The work of SEO, including creating content, optimizing pages, analyzing results, and telling the story of our clients and their services and products is all benefitting from artificial intelligence.
This includes understanding the relationship between our clients, their products and services, and the world (also known as The Knowledge Graph).
What about GEO (Generative Engine Optimization)?
At Search Influence, we believe that if you have been doing good SEO for the last decade, you are 70% of the way to generative engine optimization (or any of the other new acronyms the community wants to use).
Why only 70%? Many of the approaches to content creation and technical SEO we’ve been using are completely applicable. This is because we have always been a little geeky, and we’ve recognized that most good SEO doesn’t happen on your website.
Content distribution and promotion has always been a part of our approach to SEO. In fact, our founder Will Scott, coined the term Barnacle SEO. The idea of Barnacle SEO is to attach yourself to more authoritative sites in a non-harmful way to get in front of prospective customers, even when your site isn’t the answer.
The Three Biggest Changes to SEO for GEO
The SEO community is generally unprepared for the transition from classic SEO practices to the emerging demands of AI-driven search technologies. Here are the three critical areas that require immediate attention:
- Structure – thinking about your content and website like a machine
- Chunking – making your content scannable and digestible for humans and robots
- Distribution – getting your relevant business entities everywhere
Our CEO often says that the secret sauce for generative engine optimization is “structure it, chunk it, distribute it!”
We say “yeah right, Will, but what does that really mean?”
What Does Structure Mean in Semantic SEO?
Structure in Semantic SEO refers to alignment of your business entities to the knowledge graph.
This is done both on the page in the way you craft content, and in structured data like Schema.org definitions. Aligning your business entities to the knowledge graph connects you unambiguously to the knowledge that powers the web.
Another way to structure your content – both on your site and in media you distribute – is by making obvious what you do, for whom. Imagine describing your business as subject, predicate, object. This is called a semantic triple and makes a concrete connection for knowledge graphs and language models.
Semantic triple examples:
- Search Influence provides expert AI SEO
- Steve Jobs cofounded Apple Computers
- New Orleans is popular with tourists
The clearer that connection, the easier it is for search engines and LLMs (Large Language Models – the core of many AI systems) to know you are the right answer to their user’s query. This semantic relevance is what separates successful AI SEO from traditional approaches.
What is Chunking in Semantic SEO (AI SEO)?
Chunking is the way AI breaks up content to make it easier to understand.
Language models like ChatGPT and Google Gemini don’t read full pages the way people do. They read short passages called chunks. These chunks are usually a few sentences long, often between 150 and 300 words or tokens. Each chunk is converted into a mathematical format called an embedding, which lets the AI understand and compare meanings.
A lot of valuable content devices are naturally great for chunking:
- Short intros and callouts
- Lists: bulleted, numbered
- Captions and descriptions
- FAQ and How To content
If you’ve spent any time looking for information on the internet, you’ve probably seen a bunch of content that could have been better chunked. The typical wall of words was definitely not written with chunking for information retrieval in mind.
Why Does Content Distribution Matter for AI SEO?
Your brand is more likely to be discovered and deemed prominent the more present you are online.
There have been a number of studies published of what websites are most prominent in generative engine results. In short, there’s only one winner across all of them.
The winner is Wikipedia.
For the others, it is a range of prominent and not so prominent sites with little overlap.
One of the reasons certain sites are more prominent than others in large language models has to do with how they were trained. Training is how the LLMs built their models and informs how they “see” the world.
Wikipedia is the common thread among them. In fact, Google’s original Knowledge Graph relied heavily on Wikipedia data. Other common sources include news sites, forums, and other user generated content sites like Reddit and YouTube.
As you can see, content distribution is a cheat-code to increase your likelihood of inclusion and prominence in LLMs and AI search.
Does SEO Even Matter in an AI World?
Good SEO definitely matters for AI SEO and Generative Engine Optimization (GEO).
The AI that we use every day is not just trained data mathematically regurgitating content in response to our prompts. It is actually a combination of approaches called Retrieval Augmented Generation that brings together search and LLM capabilities.
What this means for AI SEO is “it depends.” SEO joke.
What it means is that top ranking in traditional search will have benefits to your prominence in AI search, too. However, the future of SEO will focus more on building trust than just achieving the number one ranking.
Another reason to consider top ranking is that third party search engines like Duck Duck Go may use modified versions of Google or Bing search results.
A final point about ranking: savvy users will still scroll past the AI results and may have more trust for organic search results. This means that if a journalist or creator is looking for a source they may be more likely to reach out to those with the top organic results.
What is Retrieval Augmented Generation (RAG)?
Retrieval Augmented Generation combines search with language models to generate informed responses.
RAG enhances large language models by injecting real-time knowledge into the generation process. Instead of relying solely on what the model was trained on, RAG systems retrieve relevant content from external sources like websites or databases and feed that context into the model before it generates a response. This hybrid approach improves factual accuracy, recency, and specificity.
And in the case of Google’s AI Overviews and AI mode, the database the language model is looking at is the Google index.
So SEO is still critical in a GEO/AI SEO world because you must be both discoverable in search, and part of the model to be most successful.
Can SEO Be Done by AI?
Yes, AI can assist with many SEO tasks, but it cannot completely replace human SEO expertise.
AI tools can help with keyword research, content optimization, technical SEO audits, and even content creation. However, SEO specialists still need to provide strategic direction, understand search intent, and make decisions about SEO best practices. The most effective approach combines AI-powered tools with human expertise.
Which AI is Better for SEO?
Different AI tools excel at different aspects of SEO work.
For content creation, tools like ChatGPT and Claude can help generate blog posts and optimize content. For technical SEO, specialized SEO software like, Screaming Frog, SurferSEO and others provide AI-powered insights. The best choice depends on your specific needs – whether you’re focusing on keyword research, content strategy, or technical optimization.
Is SEO Going Away with AI?
No, SEO is not going away with AI – it’s evolving.
AI Overviews and generative search engines still rely on web content and search results to provide answers. SEO professionals need to adapt their strategies to focus on structured data, semantic relevance, and content that AI can easily understand and cite. The fundamentals of good SEO – understanding user intent, creating valuable content, and optimizing for search engines – remain crucial.
Can ChatGPT Do SEO?
ChatGPT can assist with many SEO tasks but cannot replace comprehensive SEO expertise.
ChatGPT can help with keyword research, writing meta descriptions, creating SEO content, and analyzing search trends. However, it cannot access Google Search Console, track keyword rankings, or perform technical SEO audits. It’s best used as an SEO content assistant alongside other marketing tools and human expertise.
Why Should You Work with Search Influence for AI SEO?
Search Influence combines decades of SEO and digital marketing expertise with the technical know-how to win in AI Search.
Our CEO Will Scott is an AI SEO expert who writes and speaks frequently about SEO for AI and generative search. He is also the instructor for the SMX Master Class on Generative Engine Optimization.
Our approach to SEO has been semantically focused since Google rolled out the knowledge graph and we are deep schema nerds. We understand how to:
- Optimize content for AI engines and traditional search
- Implement structured data and meta tags effectively
- Build internal linking strategies that work for both humans and AI
- Create content that satisfies search intent across multiple pages
- Develop local SEO strategies for client sites
- Analyze search trends and track keyword rankings
- Execute link building campaigns that improve organic traffic
- Work with WordPress sites and online stores
- Support SEO agencies and marketing agencies with client projects
And we’re fun to work with. If you need support executing AI SEO, get in touch.
Whether you’re an SEO professional looking to understand generative engine optimization, a site owner wanting to improve your website’s visibility, or a business needing comprehensive SEO work, we have the coding skills and expertise to help you succeed in the age of AI search.
