Blog

AEO - 3 tips for optimizing your content for AI-powered chatbots

Katharina Rösner
AEO Inhalte für KI optimieren Titelbild
öffnet größere Ansicht, auf dem im FrachtPilot der Lagerbestand mit allen wichtigen Informationen geöffnet ist

AI-powered chatbots are already registering 800 million users per week. Since users now prefer using ChatGPT or other AI bots to search for information, brands and companies must adapt to this trend. Otherwise, they remain invisible to algorithms or users if their content and data is not AI-optimized. A central concept is EEAT — Experience, Expertise, Authoritativeness and Trustworthiness. We explain how you can achieve this here.


The most important thing in brief:

  • Information search with Answer Engines is becoming more and more popular
  • SEO must be supplemented with AEO
  • Expertise, Experience, Authority and Trustworthiness (EEAT) are the main criteria for an AI to cite a website
  • Organic traffic is becoming less important
  • In addition to structured content you also need structured data

What is AEO?

AEO stands for Answer Engine Optimization and describes the optimization of content for AI-powered search engines, which provide direct answers to user questions. A distinction is made between Google Answer Engines and AI-powered chatbots. The former are for example AI Overview and featured snippets. They integrate AI to generate summaries. Users ask AI-powered chatbots questions using prompts in order to obtain information.

What is changing as a result of AEO?

  • Prompts instead of keywords: users formulate conversational questions instead of searches from 1 to 3 keywords

There are detailed questions asked respectively prompted by users, instead of entering three keywords in a search bar, the length of stay on the Answer Engines rises from about 1 minute to 12 minutes. This is because the AI provides a detailed answer that is solution-oriented and anticipates the problem. This removes the need to click on page links. The result: it is not the clicks on the website that count, but that the AI cites the website or cites it as a source.

  • user needs are in the foreground: Content must provide precise and concise answers and not just provide general, keyword-optimized content.

Therefore, we must adapt the content to new user behavior and prepare it in question-answer sequences. This helps both algorithms and users.

  • Instead of generating clicks on the website, it counts citation through AI-powered chatbots.

ChatGPT cites on average two sources, Perplexity For example, an average of six.

  • Answer Engines Cite different sources: The Google rankings do not match the pages that ChatGPT quotes. However, with AI overview they are almost identical and with Perplexity there is some overlap.

It is problematic for companies, but also for users, that ChatGPT, for example, only sometimes gives the sources. In fact, this makes the citation process opaque. In addition, users cannot check how valid information is.

Is SEO still important?

SEO continues to ensure high-quality content AEO in addition for visibility. So, search engine optimization still forms the basis for high-quality content. Mentioning the website in AI bots or AI summaries should also ensure visibility.

How do I get quoted by AI platforms?

  • For a placement in AI-powered summariesn above the search results, the page must be in the Top 10 on Google ranks
  • Using faster load times the AI can crawl and find websites better. This increases the chances of a citation.
  • AI can read structured data easier: Algorithms recognize FAQPage markup better and distinct them better than questions in the H-headings: If markups for HowTo instructions and FAQ markups are included on blog pages, they ensure visibility and citations.
  • Alt-text in pictures and videos makes content readable and understandable for algorithms.
  • Transcriptions and subtitles in videos, AI makes crawling easier and increases user understanding.
  • With the EEAT principle the AI is shown that you have expertise, experience, authoritativeness, and Trustworthiness.

What is EEAT?

Algorithms evaluate quality and are based on phrases that are based on Experience, Expertise, Authoritativeness, and Trustworthiness (EEAT). In fact, AI needs clues and proof of your trustworthiness and expertise because it doesn't “understand” the content like a human does.

  • A main criterion is that entrepreneurships or the brand presents itself as a trustworthy source with expert status.
  • The reference to own studies and stats points to experience, authority and expertise of a company.
  • The naming and linking of the sources used promises reliable and trustworthy content.
  • Trustworthiness and authority Is also displayed by personalized and updated about pages and author profiles: This is how the AI recognizes that they are real people and classifies the content as trustworthy.
  • In addition to external links internal links show thematic depth and expertise.
  • experiences and quotes can prop up validity of statements.

3 tips for your AEO strategy

  1. Create relevant content for your target group - different target groups require different strategies
    • Consider whether it even makes sense or is feasible for your company to publish figures and statistics or your own studies.
    • Based on the interaction with Answer Engines, you anticipate what questions your customers will ask or which problems they want to solve. This integrates questions and answers into your previous content or you create new pages. A blog format, for example, is not suitable for technical-related issues.
    • Be on the platforms that ChatGPT or other AI bots quote. These include Reddit and Wikipedia as well as YouTube. But here you also have to consider whether you can or want to implement this at all.
    • Incorporate your customers more into your content strategy when this is possible in terms of data protection. This can be implemented for social media into contributions or videos, in interviews for your blog or in the form of testimonials on your website. Customer references demonstrate your experience to AI.
    • Create comprehensive content, which you illuminate from different perspectives. AI rates high-quality content positively, classifies it as trusty one and she recognizes authority.
  2. Technically implement AEO
    • Where it makes sense, include structured data on your website. Schema.org markup such as FAQs or step-by-step instructions work very well for technical questions. Particularly suitable for AI are JSON LD code snippets.
    • Allow AI crawling on the website. Algorithms crawl, i.e. search, your website for specific information.
    • Use Structured data tools such as Google Search Console or Google rich results test. 
    • Include CTAs in your content strategy.
  3. Track your AEO strategy
    • Ask ChatGPT relevant questions and see if your page is cited.
    • Review various AI-powered bots to find out where you're being mentioned
    • With SEO tools, you can check the relevance of the questions.
    • Create custom channel groups in tracking tools like Google Analytics 4 for Answer Engine Traffic
    • Referral traffic, i.e. traffic generated by citation and mentioning by AI, is identified with Regex.
    • Use special AEO tools like AI Monitor, Profound or BrandRank.ai 
    • Track your AEO with AI Grader.
    • You can keep checking the meta data and meta descriptions with SEO plugins or SEO tools.

How do I optimize my content for AEO?

There are a few things you can consider to make content easier to understand. This is optimal for AI because it does not have a human understanding of language and must approach the meaning of models. Machine processing, for example, requires machine-readable content. Algorithms can also read and recognize structured content better. Clear headlines and precise wording are helpful for this.

What are precise formulations?

Precise formulations are defined as “subject-verb-object” structures, so-called “semantic triples.”

What is structured content?

  • The application of headline structures establishes a clear structure. It is also important for SEO. If headlines are formulated as questions, these can be answered directly in order to provide clear question and answer sequences. This combination is best citable because it was trained with dialogue-oriented data and therefore recognizes it better.
  • With the help of concise sentences and short paragraphs people and algorithms can better capture and process content.
  • Key points, numbered lists, summary tables and info boxes contribute to a concise representation of the content.

What is the best way to structure texts?

  • Build your content according to the pyramid scheme: Specific examples should only be given after the question-answer sequence. Experiences and quotes should be mentioned afterwards.

What does AI rate positively?

  • Algorithms rate extensive and comprehensive content as trustworthy from multiple perspectives. Trustworthiness demonstrates authority and expertise to AI.
  • The reference to external links and the mention of cited sources is evidence of expertise and trustworthiness - backlinks are still important.
  • If the date of the last update or edit is added, the AI recognizes the authors' authority and rates the content as trustworthy.
  • Meta descriptions are no longer just a teaser, but should already answer relevant questions.
  • The H-structure for headlines still works, but better is using FAQPage markups and HowTo markups.

Beispiel für ein FAQPage Markup in JSON LD von ChatGPT Wie kann ich frachtPilot testen? Sie können FrachtPilot 14 Tage kostenlos testen
Example of a FAQ markup in JSON LD (Source: ChatGPT)

What content is processed for AEO?

  • AI integrates complex product descriptions and technical documentation in its answers. With AI-powered summaries they are the first to appear in search results.
  • The AI is turning on holistic image of a company or a brand by including all channels in the search. Therefore, should played consistently on all channels become.
  • The more channels the AI finds with similar content, the greater the likelihood of being quoted

How can AI “understand” content?

AI-powered chatbots use structured data so that they can recognize keywords in context, or co-occurrences to be able to analyze them semantically and to classify them. This is done in knowledge graphs models. In addition, with it, they can do sentiment analyses and represent positive or negative meanings in matrices. In this way they get closer to the meaning because they cannot understand the content.

Beispiel für ein knowledge graph
Knowledge graph example

conclusion

FrachtPilot has as an expert status as an industry-specific ERP software in the field of regional direct marketing. If you are still looking for a suitable ERP system, have a look at our website, book a free webinar or try FrachtPilot straightforward for free. We are looking forward to you 😉