Edip Kilinc is the founder of Webwavers. Together with his team, he develops websites, brands, and digital presences for ambitious companies – with over 250 completed projects and accolades including the German Design Award 2026.
Classic search provides a list of links. The user decides which one to follow. Generative systems – ChatGPT, Perplexity, Google with AI answers – provide a formulated answer instead and cite some sources.
This results in a different competition. It is no longer just about being in first place, but about being included in the answer. And different rules apply to this: it is not the page that is the most comprehensive that wins, but the one from which a precise sentence can be extracted.
Generative Engine Optimization – GEO for short – describes the work of preparing content in such a way that it appears in such answers.
Classic search provides a list of links. The user decides which one to follow. Generative systems – ChatGPT, Perplexity, Google with AI answers – provide a formulated answer instead and cite some sources.
This results in a different competition. It is no longer just about being in first place, but about being included in the answer. And different rules apply to this: it is not the page that is the most comprehensive that wins, but the one from which a precise sentence can be extracted.
Generative Engine Optimization – GEO for short – describes the work of preparing content in such a way that it appears in such answers.
Before discussing wording, a fundamental point must be clarified: Most AI crawlers do not execute JavaScript. GPTBot, ClaudeBot, PerplexityBot, and the others see the delivered HTML – and nothing else.
This has consequences that many companies underestimate. Content in expandable sections that is only loaded upon a click is invisible. Glossaries whose definitions are retrieved via a query are invisible. Everything that is only created in the browser does not exist for these systems.
The test takes two minutes: Open the page, display the source code, and search for a sentence from the content. If it is not in there, it does not exist for AI systems – no matter how good it is.
Generative systems extract text blocks. The better a paragraph functions on its own, the more likely it is to be adopted.
Three habits follow from this. First: the answer is at the beginning, not at the end. Anyone who first leads up to a point and then resolves it does not deliver an extractable paragraph. Second: one thought per paragraph, three to five sentences. Third: resolve references. A paragraph that begins with "This leads to" loses all meaning outside of its context.
Definitions are particularly effective. A sentence following the pattern "X is Y, which does Z" is cited with above-average frequency because it can be used without adaptation.
Headings are guideposts for generative systems. An H2 formulated as a question directly signals which question the following section answers – and that is exactly what is being searched for.
Lists and tables are adopted disproportionately often because they already have the format in which the answer should appear. Anyone who hides a list within body text is throwing this opportunity away.
Added to this is structured markup: Article, FAQ, and definition markup give machines a second, unambiguous copy of the information. This does not replace good content, but it removes ambiguity.
It is striking how strongly generative systems prefer content that is verifiable. A statement with a number, year, or named source is adopted significantly more often than the same statement without.
This also applies to one's own experience, as long as it is concretely named. "In over 250 completed projects" carries more weight than "many years of experience" – not because it sounds more impressive, but because it is formulated in a verifiable way.
The same goes for people: A text with a named author and function is weighted differently than an anonymous one. This is one of the reasons why author details on specialist articles are not just an accessory.
The honest answer: only to a limited extent. There is no Search Console for AI answers.
What does exist: the direct test. Ask the questions you want to be mentioned for in ChatGPT and Perplexity – exactly as a customer would ask them. Note down who is mentioned. Repeat this monthly. It's unscientific, but it is the only measurement that points directly to your goal.
Additionally, you can check the server logs to see if GPTBot and others are crawling your pages at all. If they aren't coming, any content work is in vain.
