AI monitoring
Ongoing brand monitoring in ChatGPT, Claude and Gemini
Regular measurement runs with your questions, history across all runs, competitors and sources in view. Plus the optimizer, which tells you after every run what to do next.

- Three models includedChatGPT, Claude and Gemini in every plan, with no extra charge per model.
- Two variants per questionUnprompted and when asked for sources. That shows whether your website is really cited.
- Actions, not curvesThe optimizer derives concrete steps from every run and re-checks completed actions.
- Made in GermanyGerman-speaking, billed in euros, a personal contact.
History
See what has changed since your last action
Every run goes into the history: AI Visibility Index, mentions, position and your own website as a source. Completed actions appear as markers in the chart.
- Index from mention, position and source, using only questions without your brand name
- Comparison per model and per region
- Effect of completed actions on the questions affected

AI evaluation
Ask your evaluation
Why is your brand missing from a question? Is a particular listing worth it? The AI evaluation answers from the data of your run and shows what the answer is based on.
- Answers only from your measurement data, with evidence
- Suggestions for the obvious follow-up questions
- 20 to 250 questions a month, depending on the plan

Competitors and sources
Who gets mentioned instead of you, run after run
Which providers the models recommend instead of you, how they justify it and which websites serve as evidence.
- Competitors with share and position
- Cited domains and your own citations
- Where a placement is realistically within reach

Plans
Four plans
Minimum term three months, after that you can cancel monthly.
monthly
Basic
€49 per month, net
15 questions, 90 answers per run · monthly run
- ChatGPT, Claude, Gemini included
- Overview with optimizer
- AI evaluation: 20 questions a month
- History, competitors, sources
- Domain and page check
- PDF report per run
monthly
Starter
€99 per month, net
50 questions, 300 answers per run · monthly run
- ChatGPT, Claude, Gemini included
- Overview with optimizer
- AI evaluation: 50 questions a month
- History, competitors, sources
- Domain and page check
- PDF report per run
recommended
Professional
€199 per month, net
100 questions, 600 answers per run · run every 14 days
- ChatGPT, Claude, Gemini included
- Overview with optimizer
- AI evaluation: 100 questions a month
- History, competitors, sources
- Domain and page check
- PDF report per run
monthly
Growth
€449 per month, net
250 questions, 1,500 answers per run · run every 14 days
- ChatGPT, Claude, Gemini included
- Overview with optimizer
- AI evaluation: 250 questions a month
- History, competitors, sources
- Domain and page check
- PDF report per run
With implementation support from €239 a month. A call after every run, agreeing actions, reviewing copy. +€190 to +€490 depending on the plan.
How the support works →For agencies
Several brands, one login
Set up projects for your clients and pass on reports as PDFs. White-label reports and client logins on request.
Frequently asked questions
What customers want to know first
The answers are plain text in the page source and marked up as FAQPage, so AI models can quote them.
What does AI monitoring cost?
From €49 a month net for 15 questions. Starter with 50 questions costs €99; both measure once a month. Professional with 100 questions costs €199, Growth with 250 questions €449; both measure every 14 days.
How often does VISIBILIS measure?
Basic and Starter once a month, Professional and Growth every 14 days. New pages and directory listings take weeks before the models pick them up. One run a month shows whether anything has really moved.
Why don't you measure daily?
AI models don't give the same answer to the same question every time. Measuring daily mostly shows that movement inside the models rather than how your own brand is developing. With only a few questions, one mention more or less shifts the curve even though nothing has changed. Instead, we evaluate every answer word for word: who is mentioned, for what reason and with which source.
Why do Professional and Growth measure more often?
From 100 questions upwards, individual outliers carry less weight. There, a run every 14 days shows real movement, for example in markets with many competitors.
Which models are included?
ChatGPT, Claude and Gemini in every plan, each with web search. There is no extra charge per model.
What counts as a question at VISIBILIS?
One customer question in two variants, put to three models. 50 questions give 300 answers per run.
How long does monitoring run?
At least three months, after that you can cancel monthly. It takes three runs to see whether an action is working.
Measure first, then decide.
A free sample with two questions, or go straight to the fixed-price visibility analysis.