What Sources tracks
When an AI model answers a prompt, it often cites external URLs as supporting evidence. Sources extracts every cited domain across all your tracked prompt runs and organizes them into a ranked table. This tells you:- Which domains AI models trust and cite most often
- How your domain compares to competitors’ domains in citation frequency
- What categories of content (institutional, editorial, UGC, etc.) dominate AI responses in your market
Tabs
Domains
A table of every root domain cited in AI responses, sorted by total citation count for the selected period. Each row shows:| Column | Description |
|---|---|
| Domain | Root domain (e.g. example.com) with favicon |
| Source type | Category assigned to this domain |
| Citations | Total number of times this domain was cited |
| Share | Percentage of all citations going to this domain |
| Trend | Direction vs. previous period |
| Models | Which AI models cited this domain |
| Type | Examples |
|---|---|
| Institutional | Government sites, universities, regulatory bodies |
| Corporate | Brand and company sites |
| Editorial | News publishers, magazines, industry blogs |
| Own brand | Your registered domains |
| Reference | Wikipedia, encyclopedias, knowledge bases |
| UGC | Reddit, forums, review platforms |
| Competitor | Domains linked to tracked competitor brands |
| Other | Uncategorized or mixed content |
Usage over time
A line chart showing how the citation share of a selected domain has changed day by day. Use this to:- Confirm whether a domain’s citations are growing or declining
- Detect model-specific trends (e.g. Perplexity citing a domain more than ChatGPT)