Source Attribution (AI)
Tracing each AI description of a brand back to the specific URLs, press articles, reviews, and structured sources that taught the model.
Definition
Source attribution is the practice of mapping each AI-generated description of a brand to the underlying sources that taught the model. Without it, fixing perception is guesswork.
Why it matters
Most teams try to fix AI answers by rewriting their own website. The website is rarely the source. Source attribution identifies the small set of pages, often 5–10 URLs, that actually shape the model's framing.
How it is done
For models with explicit citations (Perplexity, Google AI Overviews), source mapping is direct. For models without, retrieval-probing techniques and link analysis are combined to identify the most likely influencing URLs.
See how your brand performs across AI assistants
Strajist tracks your visibility, share of voice, and citations across ChatGPT, Gemini, Claude, Perplexity, and more.
Start free trialRelated terms
The discipline of measuring how AI models describe, categorize, and contextualize a brand, and which sources shape those descriptions. One layer beneath visibility tracking.
The week-over-week change in how AI models describe a brand, driven by shifts in the model's source mix.
Monitoring which of a brand's pages AI engines cite as sources, how often, and which prompts trigger each citation.
Whether AI-generated answers describe a brand in positive, neutral, or negative terms.
Tracking which competitors AI models recommend in your category and how their visibility evolves over time.
Designing and refining the set of prompts used to test AI visibility, so that the prompts reflect real buyer questions and produce stable, comparable results.