Tools · Citability
LLM citability scores for federal profiles
Citability measures how likely a large language model is to cite a profile when a voter asks about that office holder. Each score is a deterministic, explainable blend of content depth, structured data, source freshness, and measured citation rate. The score is free and crawlable — open any profile to see the full per-component breakdown and what would raise it.
- Profiles scored
- 582
- Average score
- 76.8
- Strong (≥ 80)
- 46
- Solid (60–79)
- 536
How the score works
The composite (0–100) is a weighted blend of four components. The three deterministic heuristics are computed for free from the published profile; the empirical component is measured on a cadence against a controlled question set and cached — it is never computed on a page request.
Content depth
Weight 30%
Narrative length, sources, and structured sections vs the depth floors.
Structured data
Weight 25%
How many JSON-LD-enriching fields are populated for machine extraction.
Source freshness
Weight 15%
Recency of the profile’s source-verification timestamps.
Measured citation rate
Weight 30%
Cadenced measurement of how often a model panel cites the profile URL.
Dataset version 1.20260604 · scored
Presidents46
U.S. Senators100
U.S. Representatives436
Related on The Candidate
- Our methodologyHow we source facts and structure every profile for machine readability.Open
- Public read APICitability scores are exposed as a documented JSON endpoint.Open
- Federal officesThe hub-of-hubs for every federal office we cover.Open
- About The CandidateOur mission, sourcing posture, and neutrality contract.Open