The Open Data Standard for AI Context Ranking
GenRank is a community-driven, transparent platform that measures how AI models recommend brands, products, and entities.
What is GenRank?
GenRank is an open data standard that tracks how leading AI models like ChatGPT, Claude, Gemini, and others recommend entities across various categories. Unlike proprietary analytics tools, GenRank believes that AI visibility data should be transparent, accessible, and community-driven.
Our mission is to create a standardized, open benchmark for understanding AI recommendations—empowering brands, researchers, and individuals with actionable insights into the AI-influenced discovery landscape.
Community-Driven Data
GenRank is built by and for the community. Anyone can participate in shaping the questions we ask AI models and contribute to our growing dataset.
Submit Questions
Propose questions you want AI models to answer. Questions are validated for quality and relevance before being added to our polling system.
Vote & Discuss
Upvote questions you find valuable and engage in discussions about AI recommendations, methodology, and data quality.
Access Data
All ranking data is publicly available. Use GenRank scores for research, competitive analysis, or building your own applications.
How It Works
Question Curation
Questions are submitted by the community or curated by our team. Each question undergoes AI-powered validation to ensure quality, relevance, and that it will elicit meaningful ranked recommendations from AI models.
AI Polling
Approved questions are sent to multiple leading AI models on a regular schedule. We capture raw responses and extract structured ranking data, including entity names and their positions in each recommendation list.
Entity Resolution
Our system identifies and normalizes entities mentioned in AI responses. Variations like "ChatGPT", "OpenAI ChatGPT", and "GPT-4" are mapped to canonical entities, ensuring accurate aggregation across responses.
Score Calculation
GenRank scores are computed using our transparent formula that weighs ranking positions and AI model market share. Higher positions earn more points, and scores from more influential AI models carry greater weight.
The GenRank Formula
Our scoring methodology is fully transparent. Here's exactly how we calculate GenRank scores:
We use logarithmic decay to create a balanced scoring system. Top positions are rewarded significantly, but lower positions still contribute value. This prevents winner-take-all dynamics while recognizing prominence.
Position Value Examples
Example Calculation
Consider "Notion" ranked #1 by ChatGPT (market weight 0.45) for the question "What are the best productivity apps?":
If Notion also ranks #2 by Claude (weight 0.25) and #3 by Gemini (weight 0.30), the scores accumulate across all models and questions to form the final GenRank score.
AI Models We Track
GenRank monitors recommendations from 18 AI models across 7 providers. Models are weighted by estimated market share and can be activated or deactivated based on availability and relevance. Active model weights sum to 100%.
OpenAI
5 modelsAnthropic
3 modelsPerplexity
2 modelsDeepSeek
2 modelsxAI
1 modelMeta
1 modelCurrently Active (9 models = 100%)
Model weights are updated quarterly based on market research, usage data, and API traffic estimates. Inactive models are preserved for historical data and can be reactivated as needed.
Our Open Data Commitment
GenRank is built on the principle that AI visibility data should be a public good. We commit to:
- ✓Transparency — Our methodology, formulas, and scoring systems are fully documented and publicly available.
- ✓Accessibility — All ranking data is freely accessible to the public without paywalls or restrictions.
- ✓Community Governance — The community shapes what questions we ask and helps maintain data quality.
- ✓Neutrality — We don't accept payment to influence rankings or methodology.