Get recommended by AI: How to make the list and reach the top
Key takeaways
- AI assistants such as ChatGPT, Gemini and Perplexity almost always recommend companies in lists. In our GeoUp analysis, 92% of all company mentions appeared in list or ranking answers.
- When a company is mentioned, it is often right at the top. More than one in three mentions (36%) went to first place, and in our ongoing measurement series, five of six visible client websites have an average rank between 1.3 and 2.2.
- First place depends less on size than on the source. In our data, the more a company’s mentions rely on its own website, the higher it ranks.
- Five levers get you onto the list and to the top: clear specialization, provable strengths, comparison and guide content, consistent facts, and mentions in the sources AI systems read.
As of September 2026
The owner of a physiotherapy practice is looking for a tax advisory firm that knows the healthcare professions and is located nearby. She asks ChatGPT and gets three firms suggested, each with a short explanation. That same day, she calls the firm at the top of the list.
For many people, searching like this has become second nature. According to the ARD/ZDF Media Study 2026 by Germany’s public broadcasters, 59% of people aged 14 and over in Germany use AI tools at least occasionally. In a survey by the German digital association Bitkom published in November 2025, half of internet users said they use AI chats instead of traditional search engines at least occasionally. Many prefer a compact answer to clicking through search results, according to Bitkom CEO Bernhard Rohleder.
Few spots, a lot of weight: How AI assistants recommend
A traditional search delivers a results page full of links. Searchers decide for themselves which provider fits. An AI assistant takes this preselection off their hands. It summarizes what it finds and names a handful of companies.
Our GeoUp analysis of 12,769 AI answers from January to July 2026 shows how consistently AI systems rely on lists. It covers 27 companies from the DACH region (Germany, Austria and Switzerland) and 100 complete measurement runs across ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews. Of 966 company mentions, 885 appeared in answers of the “The best providers for …” type, which is 92%. And 343 of the 966 mentions, or 36%, took first place.
This gives every spot on such a list real weight. What counts is not only whether you are mentioned, but also where.
First place is within reach, even for specialists and regional providers
Since July 2026, we have been measuring 19 client websites with GeoUp using a uniform method: 30 questions, five AI systems, 150 answers per run, every two to four weeks. By September 25, 2026, this added up to 119 measurement runs with 17,790 answers, and for every mention we recorded its position in the list. Six websites were mentioned at least ten times in their most recent run. Five of them reached an average rank between 1.3 and 2.2:
- a manufacturer of floor care equipment: average rank 1.3
- a regional automotive service provider: average rank 1.4
- a manufacturer of kitchen appliances: average rank 1.6
- a B2B software provider: average rank 1.8
- an exhibition stand builder: average rank 2.2
The average rank is the mean list position across all mentions in a measurement run. A value of 1.4 means mostly first place, occasionally second or third. Across the six websites, the median share of mentions in the top three was 63%.
One connection stands out: the more a company’s mentions rely on its own website as a cited source, the higher it ranks. For the automotive service provider, 81% of mentions since July are based on the AI system retrieving its website and linking to it as evidence. For the B2B software provider, the figure is 96%. Both rank between first and second place on average, while the sixth website, a well-known automotive supplier, is mainly mentioned via dealer and comparison portals in long product lists and ranks 3.5 on average.
The explanation lies in how these systems work. An AI answer is built from the sources the system retrieves for the question. If your website provides the right answer, it forms the core of the recommendation, and you end up at the top.
Five levers: How to make the list and reach the top
Which companies an AI recommends depends on what it finds for the question. Five levers make sure it finds you.
Clear positioning and specialization
In its announcement of AI Mode for Germany in October 2025, Google reported that early AI Mode users asked questions two to three times as long as typical Google searches. Anyone who asks like the practice owner in our example describes the service, target group and region all at once. This is exactly where specialized and regional providers have their opportunity.
Phrase your specialization as specifically as your customers ask, for example “tax advice for medical and therapy practices in Leipzig” instead of just “tax advice”. Give every core service its own page that says who it is for and where you operate. For local search, Google describes this principle as relevance: the more complete and detailed the information in a Business Profile, the easier it is to match the listing to the right searches.
Provable strengths
A recommendation needs a reason. The more concretely your website proves your strengths, the more material the AI finds for it, ideally in numbers and facts: year founded, completed projects, reference clients, certifications and awards with issuer and year. Evidence of your specialization is especially valuable, such as the number of practices you serve. You can also add awards to your schema markup via the award property, for your company as well as for individual services.
Comparison and guide content
Recommendation lists come up for questions like “Which provider is best for …?” or “Which suits me better, A or B?”. These are exactly the questions your website should answer. In its guidance on writing high-quality reviews, Google sets out principles that transfer well: explain what sets a solution apart from the alternatives, which option works best for which use, and give reasons for every recommendation.
For the tax advisory firm in our example, this could be a guide on “What practices should look for when choosing a tax advisor” or a comparison such as “Doing your own bookkeeping or outsourcing it”. A GeoUp measurement shows which questions these are in your industry.
Consistent facts
Your name, address, phone number, services and specialization should read the same everywhere: on your website, in your Google Business Profile, in business directories and on LinkedIn. For local search, Google writes that businesses with complete and accurate information are more likely to be shown. Consistent details also make it easier for an AI to attribute every mention clearly to you.
In your website’s code, you record the same facts in machine-readable form with schema markup. Two properties are particularly relevant for recommendation questions: areaServed for your service area and serviceType in the Service type for the kind of service you provide. You can read more about company markup in What does AI say about your company? Complete, accurate markup that matches the visible information is standard for our clients as part of our AI visibility service.
Mentions in sources AI reads
Besides your own website, AI systems draw on other sources. In our analysis of 12,769 AI answers, industry and supplier directories, trade media and press portals were among the most frequently linked sources. Every mention there confirms that you exist and what you stand for. For local search, Google also names a business’s prominence as a factor, based in part on the number of websites that link to the business and the number of reviews.
Listings and articles have the greatest impact on the sites that AI systems in your industry actually draw on. Authority Building in GeoUp identifies these portals and also checks whether your name, address and phone number match everywhere. The two complement each other: mentions elsewhere make you known and confirm your details, while your own website delivers the core of the answer that moves you to the top.
Measure where you rank
You can get a first impression yourself. Ask ChatGPT, Gemini and Perplexity for the best provider of your service in your region, and note where you appear. In a simple mention rate, first place and seventh place count the same. In their impact, they do not.
We developed GeoUp to give you a reliable trend over time. With every measurement, our tool asks ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews 30 typical customer questions from your industry and records whether your company is mentioned, in what position and which sources the AI cites for it. Measurements run regularly, up to weekly, because rankings fluctuate from run to run and only the trend shows the direction. You can add your own questions.
If you want to know why a competitor is recommended more often, use the Market Battle Mode. It places your visibility in all five AI systems directly next to that of your competitors and shows which sources the AI draws on when it recommends others. This shows you where the gap is largest and where a targeted measure will have the most impact.
Frequently asked questions
In what form do AI assistants recommend companies?
Almost always as a list. In our analysis of 12,769 AI answers, 92% of all company mentions appeared in list or ranking answers, and more than one in three mentions went to first place.
Does a small or regional company have a chance at first place?
Yes. In our measurement series, a regional automotive service provider ranks 1.4 on average. What matters is not the size of the company, but whether its own website answers the questions that trigger recommendation lists.
Which content helps most to get recommended by AI?
Pages that directly answer the recommendation and comparison questions in your industry: clear service pages with your specialization and region, proven strengths, and guides and comparisons. Schema markup and consistent details in directories round this off.
How do I find out where AI ranks me?
With regular measurement across several AI systems. GeoUp asks five AI systems 30 typical customer questions and records the position and the cited sources for every mention. The free GeoUp quick check gives you a first impression.
Can I hand off the work on my AI recommendations?
Yes. Our SEO and GEO expert team regularly writes new articles on exactly the questions your customers ask AI, strengthens your presence on your website and wherever AI systems look things up, and uses GeoUp to measure the results. A monthly report shows what has become more visible. Learn more on our AI visibility page, or write to us directly at our email address.
Conclusion
AI assistants make the preselection for their users and recommend providers in lists. Fewer spots mean more weight for each one, and first place is within reach for specialized and regional providers too. The way there is clear specialization, proven strengths, content on the questions that trigger lists, consistent facts and mentions in the right sources. If you measure regularly, you can see how your position develops.
Does AI recommend your company?
With the free GeoUp quick check, you can see in moments, based on sample questions, whether ChatGPT, Perplexity, Claude, Gemini and Google AI Overviews mention your company. GeoUp is a product by njudev.
Would you rather hand off the implementation? Our team will take care of it.
You can book our SEO and GEO expert team at any time in addition to GeoUp. Simply write to us at our email address or learn more about our AI visibility service.
Sources
- GeoUp: “Wer genannt wird, steht oben: Warum Platz 1 in KI-Antworten erreichbar ist”, September 3, 2026 (measurement series data as of September 25, 2026). https://geoup.io/news/ki-empfehlungen-listen-platz-1 (in German)
- GeoUp: “12.769 KI-Antworten ausgewertet”, July 24, 2026. https://geoup.io/news/ki-sichtbarkeits-auswertung-12769-antworten (in German)
- ZDF press portal: “ARD/ZDF-Medienstudie 2026: Nutzung von KI-Tools nimmt Fahrt auf”, press release, September 22, 2026. https://presseportal.zdf.de/pressemitteilung/ard-zdf-medienstudie-2026-nutzung-von-ki-tools-nimmt-fahrt-auf (in German)
- Bitkom: “Internet-Suche im Wandel: Die Hälfte nutzt bereits KI-Chats”, press release, November 20, 2025. https://www.bitkom.org/Presse/Presseinformation/Internet-Suche-Wandel-Haelfte-nutzt-KI-Chats (in German)
- Google: “Der KI-Modus startet in Deutschland, Österreich und der Schweiz”, Google Blog, October 8, 2025. https://blog.google/intl/de-de/produkte/suchen-entdecken/ki-modus-startet-in-dach/ (in German)
- Google Search Central: “Write high quality reviews”, last updated December 10, 2025. https://developers.google.com/search/docs/specialty/ecommerce/write-high-quality-reviews
- Google Business Profile Help: “Tips to improve your local ranking on Google”. https://support.google.com/business/answer/7091?hl=en
- schema.org: “areaServed”. https://schema.org/areaServed
- schema.org: “Service”. https://schema.org/Service
- schema.org: “award”. https://schema.org/award