What is a citation gap?
A citation gap is the set of target questions where the brand is absent and some competitor is cited. It is not the same as general absence: it is absence with proof that the question is citable, because someone is already being cited on it. It is the most actionable work list an AI Visibility operation produces.
Why the gap is worth more than the average
A 24% citation rate says there is work to do, but not which work. The citation gap says: these eighteen questions are being answered with another firm’s source, and these are the sources chosen. Every line on that list is a topic with proven demand and an identified competitor.
The gap also reveals the kind of content the engine is favouring on that topic. If every cited source is an FAQ page, format matters more than depth; if they are signed analyses, authority is the tiebreaker. That changes next month’s editorial decision.
How the gap becomes production
Each uncited question is classified by commercial value and by effort. High-value, low-effort ones, usually those the firm’s archive already answers in PDF, go first, through the data-transformation workstream. The ones that require new content go through the authority-content workstream.
In the next cycle the same list is re-tested. A gap that closed becomes evidence of result; one that did not becomes diagnosis: either the page is not being retrieved, or it is being retrieved and losing the selection, and those two causes call for different fixes.
Related terms
The citation gap comes out of the citation rate measured against the baseline, and it closes with content built on latent intent. See how AI Visibility is measured.
Frequently asked questions
What is the difference between a citation gap and simply not being cited?
A gap is comparative absence: the question produces a citation for someone, just not for you. Questions where no brand is cited are a different problem, usually a sign the engine does not treat the topic as commercial or found no trustworthy source.
How long does a gap take to close?
It depends on crawler revisit cadence and on how contested the topic is. We do not promise a citation deadline. What we guarantee is the measurement: the same list is re-tested every month on the same ruler.
Is the gap useful for picking keywords?
It is useful for picking topics, which is different. The unit of work is the natural-language question and the sub-questions it generates, rather than the keyword.