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 honored with awards including the German Design Award 2026.
Classic SEO aims for a position in a list. The user sees ten results and makes a choice. Answer engines, on the other hand, provide a summarized answer and cite a few sources for it.
This results in three differences.
There are no ranks. Either content is used or it isn't. Position seven does not exist.
The unit is the paragraph, not the page. A system adopts a self-contained thought, not an entire page. Therefore, the structure within the text is what decides.
A large part ends without a click. The mention itself is the value – it creates awareness and often leads to a direct search for the name later on.
What remains the same: content must be accessible, readable, and categorizable. Those who are technically invisible will not be cited here either. AEO does not replace SEO; it builds upon it.
Classic SEO aims for a position in a list. The user sees ten results and makes a choice. Answer engines, on the other hand, provide a summarized answer and cite a few sources for it.
This results in three differences.
There are no ranks. Either content is used or it isn't. Position seven does not exist.
The unit is the paragraph, not the page. A system adopts a self-contained thought, not an entire page. Therefore, the structure within the text is what decides.
A large part ends without a click. The mention itself is the value – it creates awareness and often leads to a direct search for the name later on.
What remains the same: content must be accessible, readable, and categorizable. Those who are technically invisible will not be cited here either. AEO does not replace SEO; it builds upon it.
1. Comparisons. A versus B, with clear criteria and a reasoned recommendation. This is by far the most frequently cited format, because that is exactly how users ask – what is better in my case – and because a comparison has a structure that can be easily adopted.
2. Concrete figures. Prices, ranges, durations, quantities. Content that answers what does X cost with a number is cited. One that says "it depends" is not.
3. Step-by-step sequences. Numbered processes with sequence and condition. Systems pass them on almost verbatim.
4. Definitions. A term, explained in two to three sentences, at the beginning of the section. This is the building block from which explanations are put together.
What is rarely cited: experiential reports without figures, opinion pieces, company announcements, and anything that only gets to the point after several paragraphs.
A paragraph that can be adopted is understandable on its own. It works without the sentence before it.
Bad: As described above, this depends on several factors. Usually, it is in the medium range.
Good: A B2B website with twelve to fifteen pages costs between 12,000 and 25,000 euros at an agency. The difference mainly depends on whether texts and images are supplied.
Three rules:
No back-references. As mentioned, see above, as described – each of these makes the paragraph unusable as a quote.
Name the subject. Not it costs, but a B2B website costs.
The statement first. Figure or core sentence in the first sentence, justification afterwards.
This is not a contortion for machines. It is the same structure that also works for skimming readers.
Systems prefer sources they can classify. Four details make the difference.
An author with a name and role. Not the editorial team, but a person with a function. In structured markup as author of type Person with jobTitle.
A date. Publication and last modification, visible on the page and as datePublished and dateModified in the markup. For topics that change, timeliness is a strong selection criterion.
An organization behind it. Organization with name, logo, address, and profiles. This anchors the content with a real sender.
Verifiable details. Figures with source, laws with section, statements with condition. A text whose assertions can be verified is preferred.
These four points take only a few minutes per post and distinguish a citable source from any random one.
This is the part that is most underestimated: Generative systems rely heavily on what others write about you.
A brand that appears in specialist articles, comparison lists, directories, interviews, and forums is visible to a model. One that is written about nowhere remains invisible – even with a perfect website of its own.
This results in concrete tasks that have nothing to do with your own site:
Appearing in comparison lists and provider overviews that already exist in your field.
Specialist articles in industry media – one guest article a year has a longer-lasting effect than ten of your own.
Directories and chambers with consistent details.
Interviews and podcasts, because they produce transcripts that can be analyzed.
This is classic public relations – just with a new reason to do it.
Visibility in answers does not appear in any standard report. It must be actively measured.
The simple method: Formulate twenty questions as customers would ask them. Enter them monthly into three systems, without logging in. For each answer, note: Does the brand appear? Is the website cited as a source? Which competitors show up?
This results in three key metrics: mention rate, citation rate, and share of voice among all mentioned providers.
Two indirect signals in existing tools:
Increasing searches for your brand name in Search Console. Users who are mentioned in an answer often search directly for the name afterwards.
Referral traffic from AI services in website statistics. The numbers are small, but the trend is telling.
The first measurement point is important. Those who measure today can prove in six months whether something has happened.
