Challenge
Companies planning a staff party or business dinner increasingly ask ChatGPT, Gemini or Perplexity which caterer to book. At the baseline measurement on January 27, 2026, Catering met Passie ranked #267 in the AI industry ranking, with a visibility score of 0.3% and 0% share of voice. In three months, AI models had sent just 3 visitors to the site and crawled it once.
Solution
Starting from that baseline, Catering met Passie worked on its visibility in AI answers: technical optimization, including a properly configured robots.txt and llms.txt, and structured data markup across the site. The Trackbase.ai platform served as a continuous measuring point: for ten realistic buyer questions, across 11 AI models, it showed where Catering met Passie ranked, with what sentiment, and against which competitors.
Catering met Passie went from barely visible to the top of its competitor ranking in AI answers, with numbers instead of assumptions.
Optimizing without measurement is guessing. Catering met Passie's journey shows what changes when a regional caterer can see exactly what AI models say about it, and which competitors they name instead.
Barely visible when event planners asked AI
More and more companies no longer start their search for a caterer on Google, but in ChatGPT, Gemini or Perplexity: who can run a festival-style staff party, which caterer does live cooking in Gelderland, or who handles tents, styling and furniture as well as the food. At the baseline measurement on January 27, 2026, Catering met Passie ranked #267 in the AI industry ranking. Its visibility score was 0.3% and its share of voice 0%.
Of the ten tracked buyer questions, Catering met Passie appeared in just one: a regional question about Barneveld and Amersfoort, at position #4. AI models had sent only 3 visitors to the site in three months, all from ChatGPT, and AI crawlers had visited the site exactly once. For most models, the caterer simply did not exist yet.
Optimizing and measuring
With the baseline set, Catering met Passie worked on the signals AI models use to decide which caterer to recommend: technical optimization, including a properly configured robots.txt and llms.txt, and structured data markup across the site.
The Trackbase.ai platform made every step measurable. Per prompt and per AI model, from ChatGPT to Google AI Overviews, it showed where Catering met Passie was mentioned, at which position, with what sentiment, and how often AI crawlers came back to read the site.
Results after eight months
The numbers after eight months are unambiguous. The visibility score rose from 0.3% to 3.2%, more than ten times higher, and share of voice grew from 0% to 0.8%. In the competitor ranking, Catering met Passie now holds the #1 spot with an average position of 1.6 across 11 AI models, ahead of regional players like mercuurs.nl, dekortcatering.nl and hutten.eu. Sentiment rose from 75% to 90%: when AI models mention Catering met Passie, they do so overwhelmingly positively.
AI models also began to recommend the caterer on more buyer questions. On the Barneveld/Amersfoort question it climbed from #4 to #2, and it now also ranks #2 for corporate parties with styling and furniture in central Netherlands and #5 for large-scale corporate events with chefs cooking on site.
Behind the scenes, AI models discovered the site in earnest. Where the site had been crawled once at the baseline, it received 1,631 crawler visits in the last three months: 515 from ChatGPT, 505 from Google/Gemini, 364 from Perplexity and 247 from Claude. AI-referred visits tripled from 3 to 9, the average session grew from 0:53 to 1:32, and visitors now land on ten different pages instead of three, from custom catering and corporate events to the contact page.