How it works Bewertungsmethodik Example Report Blog

Transparency

How rank8 evaluates

The AI Visibility Score isn't a gut feeling — it's based on real queries to real AI systems, weighted evaluation and a clear formula.

1. Realistic queries — no tricks

rank8 generates search queries that match real user intent. The questions deliberately don't mention your brand — they ask e.g. "Which car manufacturer is known for sports cars?" instead of "Is Porsche recommended?".

This measures whether AI assistants organically mention your brand — exactly as a real user would experience it.

2. Weighting of AI systems

Not all AI systems have the same market relevance. rank8 weights results based on the current distribution and usage of individual systems.

AI System Model Weight
ChatGPT GPT-4o 25%
Included
Claude Sonnet 4 20%
Included
Gemini 2.5 Flash 20%
Included
Perplexity Sonar Pro 15%
+ 0,99 €
DeepSeek DeepSeek Chat 10%
+ 0,99 €
Mistral Mistral 10%
+ 0,99 €
Llama 3.3 70B via Groq 5%
+ 0,99 €

3. Score calculation

For each query rank8 checks whether your brand is mentioned and at what position. This gives a score per query which is then weighted and combined.

Mentioned + early position → high score Mentioned + late position → medium score Not mentioned → Score 0 Visibility Score = Σ (AI Score × Weight) / Total weight

The final Visibility Score is between 0 and 100 and indicates how present your brand is in AI responses.

0 – 25
Barely visible
26 – 50
Low visibility
51 – 75
Moderate visibility
76 – 100
High visibility

4. Why this matters

More and more people use AI assistants as their first point of contact instead of classic search engines. Those who don't appear in ChatGPT, Claude or Gemini lose potential customers — without noticing.

The rank8 AI Visibility Score makes this gap visible and shows concretely where action is needed.

How visible is your brand?

Find out now — from €2.99 for 3 AI systems.

Start rank8 →