AI search is changing how people discover information.
Instead of browsing a list of results, people are increasingly finding what they need directly in tools such as ChatGPT and Google’s AI Overviews. In many cases, the search journey ends with the AI generated answer, or, if there are citation links, it can drive a new meaningful source of traffic to your content - on ustwo.com, around 7% of our visitors now arrive this way.
A new approach
Google has said that you don’t need more than standard SEO to rank well in their AI Overviews. However, ChatGPT and other LLMs work differently. They don’t just crawl and rank content, they also retrieve and summarise it based on their training and ability to understand, trust, and recognise authority.
This has led to the emergence of Generative Engine Optimisation (GEO), sometimes called Answer Engine Optimisation (AEO) or Generative Search Optimisation. These practices build on strong SEO fundamentals, but add specific considerations to help ensure your content is part of AI generated answers.
How to implement GEO
AI search is influenced not just by a single page or website, but by a consistent coverage of a topic across multiple channels, and whether that content is structured in a way that works naturally as answers.
Ensure AI can understand you
AI search tools favour content that is easy to interpret and can be easily lifted. This is just as much about publishing structure as it is about copywriting.
Practical ways to do this:
- Write answers to user prompts, use question-style headings, FAQs, how-to guides and comparisons
- Use plain, jargon-free language: clear headings, short paragraphs, lists and tables
- Write self-contained sections that make sense when quoted out of context
- Don’t use vague, self-promoting language
- From the tech side, use simple, predictable CMS data structure and semantic markup
Ensure AI can trust you
AI search tools prefer content that feels credible and backed up with evidence. If your content demonstrates real experience, it becomes safer to use and easier to cite.
Practical ways to do this:
- Support key statements with evidence: data, references, case studies, and quotes
- Focus on the end-to-end experience: what are the outcomes, impact, examples, what you learnt, why you are qualified
- Ensure content on key pages is current and up to date
Ensure AI recognises your authority
Multiple channels of trusted content makes it easier to recognise your authority and have confidence in your expertise. Your website still matters, but it’s only one part of the picture.
Practical ways to do this:
- Distribute content in multiple places: website, talks, articles, profiles, press, videos, podcasts and communities
- Spread the message across highly cited channels, e.g. reddit, wikipedia, linkedIn, and medium
- Cover not just popular, key questions, but also edge cases and gaps in industry topics
- Use consistent naming for key concepts, services, people, and products, reinforcing the same core ideas across your online presence
Bringing GEO into your marketing practice
AI search and LLMs can make this feel like a new game with new rules, but the reality is a lot simpler and a lot more human. The same long-standing SEO fundamentals still apply: create genuinely useful, high-quality content in a voice people (and now machines) can understand and trust. GEO just frames it through a new lens of “answers” and casts the net across a wider network.
In other words, do what good marketing has always done: no hacks, just clarity, evidence, and consistency. Do the basics well, show up where your audience already is, and AI systems will be able to confidently include you in the answers that matter.
Where do we go from here
Based on the principles above, we have started making some meaningful strategy and content changes to strengthen our online presence. But to do this well, we first need to test and measure how visible we are in AI generated answers, that way we can identify where changes will have the biggest impact. No doubt we will share what we learn as we go.