Generative Engine Optimisation: How To Rank in AI Search

Generative Engine Optimisation (GEO) is the process of optimising a website to rank in AI search results. GEO includes optimising content to gain features in AI Overviews, the current AI-generated feature on Google Search, and getting your business referenced and recommended in AI-generated chatbots like ChatGPT, Claude, Perplexity, and Google Gemini.

This guide will break down the jargon surrounding GEO, explain where it came from, and what you need to do to rank effectively in AI-driven search. We’ll cover practical steps you can take to optimise your website and search strategy to ensure maximum visibility even when users bypass traditional search engines.

What Is Generative Engine Optimisation (GEO)?

Generative Engine Optimisation, or GEO, is a technique to optimise website content so it’s referenced and prioritised by AI-driven search engines. Unlike traditional SEO, which focuses on ranking in search engines, GEO helps generative models understand and highlight your content when crafting their outputs.

AI is generating answers, articles, visuals, and more. Generative models don’t rank content in a list like current Google search results; instead, they generate a rich response by incorporating multiple sources. If you want your website to be among the top sources AI presents, GEO is where to invest your marketing budget.

GEO focuses on increasing visibility within these generative responses instead of aiming for a higher rank on a search results page. It prioritises factors like succinctness, relevance, factual accuracy and authority to rank and reference products, services and businesses.

The Origins of the Term ‘Generative Engine Optimisation’

The term “Generative Engine Optimization” (or “optimisation” if you’re on the British side of the pond) evolved naturally from SEO. With the rise of tools like OpenAI’s ChatGPT, Google’s Gemini, Claude and Perplexity, the need for a new type of optimisation emerged. GEO takes the basics of SEO and adapts them to the requirements of AI-driven search.

The term “Generative Engine Optimization (GEO)” was first used in an academic paper published on arXiv and has started to be used among marketers, content creators, and AI enthusiasts to describe this new style of SEO. It’s a bit jargony, in our opinion, and is not to be confused with “geo” as in “geographical” locations, but it gets the job done.

How Generative Engines Work

Generative models use vast datasets to create responses to user prompts. They don’t rank content in the traditional sense but prioritise the most relevant, credible, and fluent data.

For example, if I ask ChatGPT for a recommendation on the best coffee makers, it will prioritise content that includes direct product comparisons:

Instead of using keyword density or backlinks, generative models focus on contextual relevance, user intent and the ability to fulfil the user’s query. The way AI evaluates content is different from traditional search engines like Google. It’s all about usefulness and context.

If I give ChatGPT more information about how I like to brew and drink coffee, it uses this context to provide more tailored recommendations:

More detail is needed in website content now to maximise the likelihood of these generative models understanding the specificities, nuances and relevance of your product or service to different types of users.

Struggling to know how to use AI with your marketing?

Download a free copy of AI Marketing: Everything You Need to Know and get immediately up-to-speed

Download my free copy

Front cover of Thinkplus's "AI Marketing: Everything You Need to Know" guide.

GEO vs SEO: Similarities and Differences

SEO, AI Overviews (previously SGE) and GEO are the three organic search terms that we are hearing a lot about, but they serve different purposes and use different strategies.

SEO (Search Engine Optimisation): SEO is the traditional approach to ensuring your content ranks well on search engines like Google. The goal is to rank on search engine results pages (SERPs) when users search for related topics.
AI Overviews: AI Overviews are the AI-driven summaries directly at the top of search results (previously called SGE or Search Generative Experience). It provides users with quick overviews of information sourced from various web pages.
GEO (Generative Engine Optimisation): GEO takes optimisation further into the AI realm, focusing on referencing content within generative models like ChatGPT, Perplexity, and Claude. GEO is not just about search rankings; it’s about becoming a part of the AI’s knowledge base so your content is favoured during answer generation.

How Is GEO Similar to SEO?

Visibility Goals: Both GEO and SEO aim to increase the visibility of your content, whether it’s in traditional search engines or AI-generated responses.
Content Quality: High-quality, authoritative content is crucial for both GEO and SEO. Search engines and generative models favour content that is reliable, informative and well-structured.
Keyword Strategy: Just like in SEO, GEO involves identifying relevant keywords and phrases to make content easily discoverable by search engines or AI models.
Technical Optimisation: Both GEO and SEO benefit from a technically optimised website, including proper use of schema markup, metadata and fast page loading times.
Importance of Backlinks: Backlinks remain important for establishing authority and credibility in both GEO and SEO.

How Is GEO Different from SEO?

Target Audience: SEO focuses on improving content visibility in search engine results pages (SERPs), whereas GEO aims to have content referenced in AI-generated responses, such as by ChatGPT, Google Gemini and other generative engines.
Content Presentation: SEO focuses on ranking on search pages where users choose which link to click. GEO aims to have your content directly included in an AI-generated answer, eliminating the need for users to click through multiple links.
Authority Signals: While SEO relies heavily on keywords, backlinks, and on-page signals, GEO also depends on brand mentions and having information available in trusted aggregators or directories that AI models may prioritise.
User Intent Handling: SEO focuses on optimising for a wide range of search intents, from informational to transactional. GEO, on the other hand, focuses on optimising for AI-generated answers, which tend to be more informational and less about driving direct traffic.

“SEO and GEO share a lot of similarities and they pretty much both use all the same building blocks but they put them together in a slightly different way, improving one almost always improves the other as well. You do need to start with a solid foundation of SEO though. SEO is about getting your website found by people, and bots. If the AI bots can’t find and, accurately, index your website, then they can’t evaluate it properly to use in its answers, and that’s where the SEO comes in.”

Comparing SEO, AI Overviews and GEO

Let’s get some more details on SEO, AI Overviews and GEO. Here’s how they stack up:

Feature	SEO (Search Engine Optimisation)	AI Overviews	GEO (Generative Engine Optimisation)
Purpose	Rank content well on search engines like Google.	Provide AI-driven summaries directly in search results.	Optimise content for AI models like ChatGPT to be included in generated responses as well as AI-driven summaries on search results.
Key Focus	Keyword optimisation, backlinks, page speed, and other traditional SEO elements.	Formatting content for summarisation, ensuring visibility in search result summaries.	Optimising content to provide context, relevance and authority, ensuring recommendations in generated outputs.
User Experience	Improves traditional SERP rankings to help users find content through search engine listings.	Provides users with concise, AI-driven overviews of web content to save users time.	Gets content directly referenced by AI in conversational responses or answer generations.
Main Outcome	Higher visibility in Google SERPs, more clicks and traffic.	Inclusion in AI summaries that appear at the top of Google searches, resulting in more clicks and traffic from the AI Overviews positions.	Inclusion in AI-generated responses, leading to direct exposure without relying on traditional searches and search engine rankings.
Content Requirement	High-quality, keyword-rich content that also delivers value and uses SEO best practices.	Structured content with concise answers and clear formatting for easy summarising.	Content should be authoritative, relevant, and mentioned frequently online.
When to Use	To attract organic search traffic by ranking highly in search engine results.	When aiming for featured visibility in evolving AI-driven search experiences.	To ensure your content is recommended by AI-driven tools, search features and included in AI-generated conversations.

TL:DR:

SEO is about ranking well on Google and other search engines.
AI Overviews (previously known as SGE) is about being visible in AI-driven search result summaries.
GEO is about being recommended by AI tools and generative engines that provide answers or generate content directly. This includes AI chatbots and covers AI Overviews.
To succeed, you need a mix of all three. You need traditional SEO to build a strong foundation, AI Overview tactics to gain top placement on certain Google searches, and GEO techniques to rank highly within AI-generated outputs.

How to Rank in Generative Engine Optimisation

Want your business recommended more often in generative AI models? We’ve spent a whole lot of time hypothesising, testing and experimenting across industries to understand how to optimise for generative engines. Here are the current theories of how to make it happen:

1. Optimise for Web Search: Many AI tools perform a web search in the background to generate their responses. GEO is built on traditional SEO (yep, spoiler: SEO isn’t dead yet). This means you need to rank in traditional search results to be then picked up by generative models. Essentially, you need to rank well in the old SEO game to be able to play the new GEO game.

Get to the top of Google

Learn how to get your website to the very top of Google (and turn that traffic into revenue).

Download my free copy

2. Optimise for AI Overviews: Google’s AI Overviews summarise content from web searches to deliver users an answer more quickly. These answers dominate the top of the search results page, though the latest figures show that 7% of searches feature AI Overviews. You want your business to be featured in relevant searches where an AI Overview is present. Read our detailed guide here on how to optimise content for AI Overviews

3. Optimise Your Prompts: Be clear and direct in your content. Craft prompts into your content that make it easy for AI to understand what you’re aiming for. Different types of prompts work well:

Question-Based Prompts: Asking questions like “What is the best strategy for X?” tends to get specific, in-depth answers.
Instruction-Based Prompts: Content explaining “How to do X step by step” is highly valuable for AI-generated summaries.

4. Use Relevant Keywords: Forget keyword stuffing (hopefully, you forgot this in 2011). Instead, naturally include relevant terms that help AI see the value and purpose of your content. Use latent semantic indexing (LSI) keywords* to improve content accuracy and help the model form associations.

* What’s an example of LSI keywords? If you’re a personal injury law firm, your primary keyword may be “Personal Injury Lawyer”. Some LSI keywords would be: “Accident Attorney,” “Legal representation for injury,” “Car accident claim lawyer”, “Compensation lawyer for injury” and “Slip and fall lawsuit.”

5. Get Mentioned Across the Web: The more your business or content is mentioned online, the more likely it is that generative models will pick it up.

Brand Mentions: Digital PR and Content Marketing are key here. Getting brand mentions on other websites ensures AI (these LLMs) has plenty of data about your business during its training phase.
Recommendations from Influencers: Consider setting up affiliate schemes or having influencers test your product/service and share their thoughts on their websites and/or social media.
Reviews: It’s not just brand mentions that matter, but reviews, too. Getting first-hand customer reviews on your website and third-party platforms like Google Reviews, Trustpilot, Feefo, Reviews.io and other feeds AI answers. You want both quality and quantity here. If reviews are specific, even better.

6. Show Authority and Authenticity: AI values well-sourced, factually correct content. Back up your information with reputable sources to establish authority and ensure it’s written (or at least reviewed) by an authority on the topic. Essentially, follow the principles of E-E-A-T. If the information you are communicating is based on data you have created yourself, provide the data itself (not just the conclusions from it) in the content. This way the AI can read the data directly and combine it with other datasets it knows, extending your GEO reach.

7. Make Short, Helpful Videos: People want answers, not jargon. Quick, informative videos and shorts addressing popular questions like “How do I apply for a mortgage?” can improve the quality of your content (embed them in your blogs) and also have the potential to rank on SERPs and video platforms in their own right.

Best Practices for Implementing GEO

It’s not just what you do but how you do it. Increasingly, we’re seeing GEO prioritising content that follows these best practices:

- Good Quality: Nobody wants to read crap content. Be detailed, factual and cover topics comprehensively. Include statistics and data points to reinforce credibility. Write a good author bio and include it in the blog post.
- Consistent Brand Voice: A consistent brand voice helps AI recognise your authority and credibility. We’ve found the best way to do this when generating content through AI tools or to even out tone consistency is to feed it a full Brand Accelerator report that includes tone of voice, copy examples and customer personas.
- Simplifies Complex Concepts: Make your content easy to understand. Generative models value fluency and readability, so aim to simplify complex topics while keeping key information intact.
- Skimmable Content: Make your content easy to skim by using headers, bullet points, and visuals. Generative models prefer skimmable content that can be easily summarised.

Integrating GEO with SEO

GEO and SEO work hand-in-hand to increase your content’s visibility across both traditional search engines and AI models. By combining these strategies, you ensure your content ranks well in search engines and is referenced by AI models.

Strategies for Effectively Combining GEO and SEO

To get the best of both worlds, we recommend combining GEO and SEO. Here are the strategies we recommend to effectively integrate GEO and SEO into your marketing strategy:

Optimise for Both Web Search and AI Context: Focus on ranking well in traditional search engines while ensuring your content is detailed enough for AI models to understand its value and relevance.
Content for Humans and AI Readability: Create well-structured content that is easy to skim by using clear headings, bullet points and concise paragraphs. This helps with traditional SEO while making your content easier for generative models to summarise and reference.
Backlinks and Brand Mentions: Go bigger and broader with Digital PR, focusing both on building backlinks and gaining brand mentions across many different websites.
Monitor Visibility in Both Channels: Track your visibility in and traffic from both traditional SERPs and AI-generated responses.

Does Implementing GEO Actually Work?

Research on the effectiveness of implementing GEO is limited, as it is still a new and emerging field. From our experience at Thinkplus, we’ve found that it is effective. We’ve been pioneers in this new type of optimisation, helping clients rank in AI Overviews and teaching other agencies to do the same.

Through using the GEO techniques we discussed above, our Ninja team has secured a whole heap of AI Overview spots for clients across industries. Here’s one example of an AI Overview we’ve secured (and held) for a client in the cleaning sector. This page is consistently the number 1 traffic driver to this client’s website, overtaking the homepage by a lot.

However, we also know that these AI search features and AI chatbots are evolving, and fluctuations happen. Just as with getting the top of Google’s search results, securing AI Overviews or ChatGPT mentions doesn’t mean you’ll maintain them. What marketers can and should do is experiment: figure out how to get them and try different ways to maintain those positions.

We’re not the only ones seeing this. A recent research paper set the SEO world ablaze with its data, finding that implementing GEO techniques can increase visibility in GE responses by up to 40%. The table below from the same paper shows the percentage increases in visibility after making changes to website copy. Greens are new additions, and reds are deletions. Take this relative improvement data with a grain of salt as improvement measures are limited.

Table from GEO: Generative Engine Optimization by Aggarwal et al.

How Can You Measure Results from GEO?

Tracking AI Overviews

SEMrush’s Position Tracking tool tracks AI Overview features for keywords tracked for your campaign. Within the Position Tracking tool, navigate to SERP Features and this will display keywords from your campaign that triggered an AI Overview and whether your website was highlighted.

Tracking Visibility

The primary measure currently being used to track generative engine results is overall visibility. This includes general online visibility metrics like backlinks and brand mentions. You can check your backlink profile using SEMrush’s Backlink Analytics tool. These measures are based on the assumption that the more you’re talked about online, the more likely it is that generative models will include your brand in their responses.

AI Sentiment Analysis

HubSpot has introduced an AI Search Grader tool that measures brand presence by analysing results from GPT-4o. This tool provides a helpful overview of brand sentiment and share of voice.

In practice, the tool runs a series of queries based on the business information you provide, meaning its analysis is limited to the terms you input. However, it can be used repeatedly to evaluate different products or services, giving a more comprehensive insight over time.

Measuring Traffic Referral

The most useful current measure of GEO results are traffic, sessions and conversions on your website that have come through generative engines. At Thinkplus, we set up custom reports using your GA4 data to show GEO results. These reports include AI platform traffic over time, AI platform sessions, AI platform conversions, top GEO landing pages and AI platform traffic and conversion splits.

Tracking Mention Frequency

There is no clear way to track mentions and recommendation frequency within AI chatbots. Currently, the only other way to measure mentions is through manual searches on AI engines. This process is time-consuming and — with the rapid evolution of AI search tools — the gathered data can quickly become outdated.

Challenges of GEO

GEO has its own challenges. One major issue is AI’s inherent bias — generative engines are only as good as the datasets they’re trained on. Bias in those datasets can affect which content is generated. So, content should draw on a diverse range of reputable sources and include information representing different perspectives to mitigate bias.

There are also ethical concerns around misinformation and over-optimisation and ensuring factual accuracy in your content is key to avoiding these pitfalls.

AI is evolving rapidly, and staying up-to-date on algorithm changes and new capabilities is crucial to keeping your content visible. Additionally, the winner-takes-all nature of generative models means that only a handful of top responses get the bulk of visibility, making it essential to ensure your content is as authoritative and high-quality as possible.

That’s where Thinkplus comes in: we’re pioneers of AI search, having led the way in uncovering ranking factors for and getting our clients featured in AI Overviews and on AI chatbots. We keep on top of all the AI search updates so you can focus on other important stuff in your business. Get your free website and marketing review and find out how we can do the same for you.

“Traffic from these generative engines is increasing rapidly, so although they are very new and currently small fry against traditional search, smart marketers are putting in the groundwork now to build visibility in anticipation of the continued consumer shift towards AI.”

The Future of Generative Engine Optimisation

While only a fraction of searches are carried out on AI chatbots and Google still has the vast majority of search market share, some marketers are viewing this moment as a land grab. Integrating GEO into your marketing strategy now puts you ahead of competitors. Many have not yet understood the shift that’s happening.

We Ninjas don’t have a crystal ball here. Still, it’s not absurd to think that generated AI answers could become the primary method for content discovery, especially with its integration into search engines and virtual assistants. If you’re serious about dominating your marketplace, then GEO should be in your game plan.

Move beyond SEO as we know it and get to grips with GEO techniques to future-proof your business

Get GEO insights from our Ninjas by requesting a free website and marketing review. Contact us directly to discuss our Generative Engine Optimisation services.

Listen to Our Podcast on GEO

Join Charlie, Tim and Jess for our podcast on GEO and the future of search. They talk about the rise of GEO and why it matters to marketers.

Written by Charlie Marchant and first published on 15th October 2024. Updated 4th November 2024. Expert review by Andy Tuxford on 11th November 2024.

Charlie Marchant is the CEO of Thinkplus. With over a decade of experience in organic search and content creation, Charlie has driven the growth of the agency’s Content Marketing, Digital PR and Branding services. She is co-author of the book The Ultimate Guide To Content Marketing & Digital PR.

Andy Tuxford is the Head of SEO at Thinkplus. Andy has over 8 years of direct SEO experience and over 20 years in various marketing and management roles. He has driven significant organic growth for businesses of all sizes, from local plumbers to multi-national eCommerce websites. He has played a leading role in AI search experimentation and rankings at Thinkplus.