Definition and tiers

Across the prompt set, record every URL each engine cites. Classify each citation into tiers:

TierSource typeWeightExamples
T0 — OwnedYour domain, cited directly1.0yourbrand.com/guide, your docs
T1 — Authoritative earnedReference and institutional sources0.8Wikipedia, government, standards bodies, major trade press
T2 — Commercial earnedDirectories, review platforms, industry media0.5G2, Capterra, Yelp, vertical directories
T3 — CommunityForums and social discussion0.3Reddit, Quora, niche forums
TX — Competitor/unrelatedCitations that route value to others0Competitor pages, content farms

Citation Rate = Σ(tier weights of your citations) / Σ(tier weights of all citations in the answer set) × 100. A brand can have high Visibility with poor Citation Rate — named by the engine while the evidence links route to a review site or a competitor's comparison page.

Why citations compound

Citations are not just credit — they are training signal. Pages that get cited get crawled more, quoted more, and reinforced in future answers. Engine-specific citation behavior differs: Perplexity shows inline numbered sources on virtually every answer (the most transparent citation market); Google shows link cards beside the Overview; ChatGPT shows a sources panel on search-enabled answers; Claude cites sparingly and prefers primary documents.

How to raise Citation Rate

  1. Inventory first. Run the prompt set, harvest every cited URL, tier them. Most brands discover their category's answers are built on 5–10 domains they have no presence on.
  2. Fix owned sources to be citable. Direct-answer blocks, visible dates, named authors, clean URLs — engines prefer pages that are safe to send users to.
  3. Claim the T2 layer systematically. Directory and review-platform profiles are the fastest tier to correct: complete, consistent, current.
  4. Earn T1 slowly and honestly. Wikidata entries, standards participation, trade-press coverage with real data. No shortcuts survive editor review.
  5. Feed T3 authentically. Answer real questions on Reddit and niche forums under real names. Astroturfing is detectable and corrosive.

Frequently asked questions

Which engine is easiest to earn citations from?
Perplexity, because it retrieves live and cites inline on every answer — a well-structured new page can be cited within days. Google AI Overviews is hardest because citation leans on established index authority. ChatGPT sits between: its sources panel rewards Bing-indexed, well-structured pages.
Can I see which of my pages AI engines cite?
Yes. Perplexity shows sources inline; ChatGPT shows a sources panel on search answers; Google shows link cards. Systematically recording these across a prompt set is the core of the Citation Rate measurement — our Checker automates this harvest.
Do citations from my own site count as much as third-party ones?
They count for traffic routing (T0 weight) but engines corroborate claims across independent sources — a fact stated only on your own domain is treated as a claim, while the same fact on your site plus two independent sources is treated as established. Both matter; they do different jobs.