What is generative engine optimisation (GEO)?

Generative engine optimisation is the practice of making a company’s content, facts and reputation easy for AI assistants and AI search features to find, understand, trust and cite, so the company appears accurately in AI-generated answers to buyer questions.

Do software buyers really use AI assistants to shortlist vendors?

Yes, according to recent surveys. Forrester reported in January 2026 that 94% of business buyers use AI in their buying process, and G2’s March 2026 survey found that 54% of B2B decision makers named generative AI chatbots as the top source influencing their vendor shortlist.

How can a scale-up measure its visibility in AI answers?

Build a fixed set of 30 to 50 buyer questions, run them regularly across the assistants target buyers use, and record how often the company is mentioned, in what position, next to which competitors, which sources are cited and whether the facts are correct. Add AI referral tracking and an AI option to how-did-you-hear questions.

Does Google require special markup to appear in AI Overviews?

No. Google’s guidance states that there are no additional requirements or special optimisations needed to appear in AI Overviews or AI Mode. Pages need to be crawlable, indexed and eligible to show a snippet, and structured data should match the visible text.

Who should own GEO in a scale-up?

Marketing usually runs the programme, but it works best as part of the wider revenue engine, with input from product marketing on positioning, from sales and customer success on buyer questions, and from leadership on which categories and markets to prioritise.

Insights / AI search and GEO

By Niraj GemawatPublished 3 Nov 2026

How do buyers find software in the age of AI search?

More and more software buyers start their research by asking an AI assistant, not by clicking through ten blue links. ChatGPT, Gemini, Perplexity, Copilot and Google’s AI Overviews and AI Mode now summarise a category, name a handful of vendors and explain the trade-offs before a buyer visits a single website. For a scale-up, that means being part of the AI answer matters as much as ranking in search. Generative engine optimisation is the discipline of earning that place.

Title graphic reading How buyers find software in the age of AI search, with a network of connected nodes on a navy background

What has changed in how buyers find software?

The shortlist is increasingly formed inside an AI answer. Buyers still visit websites, read reviews and talk to peers, but the first framing of the category, and often the first list of names, now comes from a generative model.

The evidence points the same way across several sources. In a January 2026 Forrester post, principal analyst John Buten reported that 94% of business buyers use AI in their buying process, and that buyers in Forrester’s Buyers’ Journey Survey, 2025 named generative AI or conversational search as a meaningful source of information at twice the rate of any other source. G2’s 2026 AI search insight report, based on a March 2026 survey of 1,076 B2B decision makers, found that 54% named generative AI chatbots as the top source influencing which vendors make their shortlist, ahead of software review sites at 43%. G2 sells marketing products to software vendors, so its findings are best read alongside independent data.

Search itself has changed shape too. Google said in its Q2 2025 earnings remarks that AI Overviews had over 2 billion monthly users across more than 200 countries and territories, and that AI Mode already had over 100 million monthly active users in the US and India. Fewer of those searches end in a click. Pew Research Center found in 2025 that users who saw an AI summary clicked a traditional result in 8% of visits, against 15% for those who did not, and clicked a link inside the summary in just 1% of visits. An Ahrefs study published in February 2026 of 300,000 keywords found that the presence of an AI Overview correlated with a 58% lower average click-through rate for the top-ranking page.

The commercial consequence is simple. If the vendor is not named in the answer, the buyer may never reach the website where the marketing team has invested most of its effort.

What is generative engine optimisation?

Generative engine optimisation (GEO) is the practice of making a company’s content, facts and reputation easy for AI systems to find, understand, trust and cite, so that the company appears accurately in AI-generated answers. The term was introduced in a 2023 research paper by Pranjal Aggarwal and colleagues, accepted to KDD 2024, which reported that content changes could raise visibility in generative engine responses by up to 40%, with results varying by domain.

GEO is sometimes called answer engine optimisation (AEO) or AI search optimisation. The labels differ; the job is the same.

How is GEO different from SEO?

GEO builds on SEO rather than replacing it, but it optimises for a different outcome. SEO aims to rank a page and win a click. GEO aims to have the company named, described correctly and cited inside an answer, whether or not a click follows.

Unit of success. SEO measures rankings and organic sessions. GEO measures mentions, citations and the accuracy of what the model says about the company.
Query shape. Buyers ask assistants long, specific questions ("Which contract management tools work for a 200-person legal team in Germany?"). Google describes AI Overviews and AI Mode as using query fan-out, running multiple related searches across subtopics to build a response.
Sources beyond the website. Models draw on review sites, analyst coverage, forums, comparison articles, documentation and press. What others say about a company carries real weight.
Foundations still apply. Google’s own guidance on AI features states that there are no additional requirements or special optimisations needed to appear in AI Overviews or AI Mode: pages must be crawlable, indexed and eligible for a snippet. Weak technical SEO therefore undermines GEO from the start.

How do AI answers choose which vendors to mention?

No provider publishes a full ranking formula, but the observable pattern is consistent: AI answers favour sources that are clear, specific, corroborated and easy to extract.

Entity clarity. The model needs to understand exactly what the company is, what it sells, for whom and where. Inconsistent names, vague positioning and different descriptions across the website, LinkedIn and directories make that harder.
Citable, structured content. Pages that answer a question directly, define terms, use clear headings and include specific facts are easier to quote than pages built around slogans.
Third-party corroboration. Mentions in trade press, analyst notes, partner pages, community discussions and comparison articles tell a model that others recognise the company in its category.
Reviews and customer evidence. Review platforms and published case studies supply the proof points that answers often repeat when they explain why a vendor fits a use case.
Freshness and accuracy. Outdated pricing pages, retired product names or old funding news can be repeated back to buyers long after they stop being true.

How do you measure visibility in AI answers?

Measure AI visibility the way a buyer experiences it: by asking the questions buyers ask and recording what comes back. Traffic alone understates the effect, because many AI-influenced journeys end without a click.

A fixed prompt set. Write 30 to 50 buyer questions by stage: category ("What tools do X?"), comparison ("A vs B"), use case, objection and pricing. Run them across the assistants the target buyers use.
Share of answer. Track how often the company is mentioned, in what position, alongside which competitors and with which sources cited.
Accuracy. Record every factual error: wrong product scope, outdated pricing, a confused company name. Each is a fix list item.
Referral and assisted signals. Segment analytics for referrals from AI assistants, and add "How did you first hear about us?" with an AI option to forms and discovery calls.
Cadence. Answers vary from run to run and change as models update, so trend the results monthly rather than reacting to a single response.

What should a scale-up’s marketing team actually do?

Treat GEO as a programme with an owner, a baseline and a quarterly review, not as a one-off content push. A practical sequence looks like this.

Fix the entity. Write one plain description of the company, its category, its customers and its markets, and make it identical on the website, LinkedIn, review profiles, partner listings and press boilerplate. Add organisation and product structured data that matches the visible text.
Check the technical basics. Confirm that important pages are crawlable, indexed and rendered as text, and decide deliberately which AI crawlers the robots.txt file allows.
Build answer pages. Map the buyer questions from the prompt set to pages that answer each one directly in the first paragraph, then go deeper: use-case pages, honest comparison pages, integration pages, pricing logic and implementation guides.
Publish citable evidence. Original data, named customer stories and specific numbers give models something concrete to repeat. Opinion without evidence is rarely cited.
Earn third-party mentions. Prioritise the review platforms, analyst briefings, trade publications, partner ecosystems and communities that the assistants already cite for the category.
Correct what is wrong. Update outdated pages, retire old product names with redirects and refresh directory listings so that stale facts stop being repeated.
Close the loop with sales. Share the prompt results with sales and customer success, who hear first-hand what buyers were told by AI before the first call.

Where does GEO sit inside the revenue engine?

GEO is not a channel to bolt on to marketing; it is part of how a company earns its place on the buyer’s shortlist. As argued in What turns SaaS innovation into recurring revenue?, winning vendors are usually present before sales is involved. AI answers are now one of the main places where that early presence is won or lost.

That makes GEO a revenue question as much as a marketing one. Positioning, category definition, customer evidence, partnerships and sales enablement all feed the same answers. A company with unclear positioning will find the AI repeating that confusion back to buyers; a company with a sharp ideal customer profile and strong proof points gives the models something specific to say. The questions founders should ask are practical ones: which buyer questions matter most, who owns the prompt set, and how AI visibility is reported alongside pipeline.

How does TGC Capital Partners approach this?

TGC Capital Partners, your Growth Engineering Partner and the strategic investment arm of Gateway Group of Companies, works with tech founders in Europe, North America and India on the commercial system behind growth. 1X Revenue Engineering covers positioning, demand generation and the revenue engine, and AI search visibility is now part of that work. It sits alongside the other 6X Factors, applied where they fit the company. Revenue is the outcome TGC measures, not a promise.

Buyers now meet vendors inside an AI answer first, so the revenue engine has to be built for that answer as well as for the website.

Frequently asked questions

What is generative engine optimisation (GEO)?: Generative engine optimisation is the practice of making a company’s content, facts and reputation easy for AI assistants and AI search features to find, understand, trust and cite, so the company appears accurately in AI-generated answers to buyer questions.
How is GEO different from SEO?: SEO aims to rank pages and win clicks from search results. GEO aims to have the company named, described correctly and cited inside AI answers, whether or not a click follows. GEO relies on the same technical foundations as SEO but also depends heavily on third-party mentions, reviews and clearly structured content.
Do software buyers really use AI assistants to shortlist vendors?: Yes, according to recent surveys. Forrester reported in January 2026 that 94% of business buyers use AI in their buying process, and G2’s March 2026 survey found that 54% of B2B decision makers named generative AI chatbots as the top source influencing their vendor shortlist.
How can a scale-up measure its visibility in AI answers?: Build a fixed set of 30 to 50 buyer questions, run them regularly across the assistants target buyers use, and record how often the company is mentioned, in what position, next to which competitors, which sources are cited and whether the facts are correct. Add AI referral tracking and an AI option to how-did-you-hear questions.
Does Google require special markup to appear in AI Overviews?: No. Google’s guidance states that there are no additional requirements or special optimisations needed to appear in AI Overviews or AI Mode. Pages need to be crawlable, indexed and eligible to show a snippet, and structured data should match the visible text.
Who should own GEO in a scale-up?: Marketing usually runs the programme, but it works best as part of the wider revenue engine, with input from product marketing on positioning, from sales and customer success on buyer questions, and from leadership on which categories and markets to prioritise.