CitenzoCitenzo AI Recs
Citenzo's own intelligence layer that rates each placement site for one goal — getting you cited by AI — from Top pick down to Avoid. It's on your side, so it flags weak and risky sites, not just good ones.
How the score works
Every site in the catalog gets a number from 0 to 100 and one of six badges. The number is a weighted sum of seven signals; then, if a placement is a mass-syndicated press release, we discount it — more on why below.
Score = 25·Cited + 20·Relevant + 20·Editorial type + 15·Mention + 8·Authority + 8·Fresh + 4·Dofollow
| Grade | Score | What it means |
|---|---|---|
| Top pick | AI-cited · or ≥ 82 | AI already cites it in your space, or it scores high on everything that earns citations. |
| Great | 66–81 | Relevant, editorial, and the kind of source AI pulls from. |
| Good | 50–65 | Solid, safe placement — a dependable choice. |
| Fair | 36–49 | Middling — fine, but nothing special for citability. |
| Weak | 22–35 | Low relevance or a low-citation format — usually skip. |
| Avoid | < 22 · or flagged | Spammy, thin, or mass-syndicated; AI rarely cites it. |
Why not just Domain Rating?
Because it barely predicts whether an AI cites you. A meta-analysis of 54 ranking-factor experiments put Domain Authority near the bottom of the pile, well below relevance and retrieval signals (Zyppy). Correlation studies land in the same place — DA explains only a few percent of the variance in AI citations. It still helps a little, but only indirectly, because authoritative pages tend to rank and get retrieved. As a number to chase, it's mostly noise dressed up as signal — which is exactly why link sellers love to quote it.
The evidence, factor by factor
Whether AI already cites the source. Citations are wildly concentrated: Reddit, Wikipedia, YouTube, LinkedIn and a short list of major outlets and review sites account for most of them (Search Engine Land). A placement on a domain the engine already pulls into answers walks straight into the citation pool; a placement on a site it never quotes can't be quoted, however high its DR. We check this against your own tracking data — the sources that cite your rivals in your niche — which is the closest thing to a direct lever we have.
Relevance beats authority. Retrieval is semantic: the engine first pulls documents that match the question, then decides who to name. Topical fit correlates far more strongly with being cited than raw authority does. A mention dead-on your topic on a mid-size site out-earns one on a bigger, off-topic outlet.
Editorial, not syndicated. This is the finding that surprises people. Earned, original editorial makes up roughly 84% of AI citations, while paid and advertorial content is a rounding error (Muck Rack). Worse, a press release blasted across dozens of outlets is cited almost never — the duplication reads as spam, not consensus (Search Engine Journal). One real guest post on a relevant site beats the same story syndicated two hundred times.
A mention outweighs a link. Ahrefs' study of 75,000 brands found that unlinked brand mentions track AI visibility about three times more strongly than backlinks do (analysis of the data). Engines resolve you as an entity from the text; the hyperlink is almost beside the point. So we reward a clean, contextual mention over a bare anchor.
Dofollow versus nofollow: a coin flip. For classic SEO it's an article of faith; for AI citation it's irrelevant. Semrush found the two link types correlate almost identically with AI visibility (Semrush). We keep a token four points for it, purely as a hedge for old-school ranking — not because it moves citations.
And what's on the page itself. This part you control after the placement is booked. Princeton's GEO study (around 10,000 queries) found that adding statistics, quotations and named sources to a page lifts its AI visibility by roughly 20–40%, with the biggest gains going to pages that weren't already ranking at the top (Princeton GEO paper). A well-sourced, quotable article earns far more than a thin one on the same site.
What Google itself says
Google's own line is that there's no special markup and no separate index for AI Overviews — the same people-first content and structured data that ranks in ordinary search is what surfaces in the AI answer (Google Search Central). You can't sneak in with a schema trick; you earn it the way you earn a good ranking, and then relevance and citations decide who gets named.
The honest caveat
These are correlations from vendor and academic studies, not physical laws, and the exact numbers move around depending on who ran the study. We lean on the direction of the evidence — which is remarkably consistent — rather than the decimals, which aren't. This formula is our best current read; as sharper data lands, the weights will change with it.
In practice: trust the badge to steer your budget toward relevant, editorial placements on the sources AI already reads — and away from the cheap, high-DR links that look impressive on a spreadsheet but change nothing about whether AI names you.