Citation gaps need limits before they need a free tool
A gap report is useful when the panel, capture conditions and source availability are explicit. Absence records a finding, not a guaranteed repair.
Founder of TomHow, based in Stevenage. SEO and AI search consultancy.
What can a citation-gap check actually prove?
A completed check can show that an agreed domain was not cited in a defined set of source-bearing answers. It cannot show universal absence or identify the cause from the answer alone. Brand mentions, owned-domain source citations, failed requests and source-unavailable answers need separate fields. A useful report preserves the exact questions and captures, then distinguishes independently confirmed defects from proposed experiments.
A measurement workflow, not a causal diagnosis: define the panel, record capture status, count separate outcomes and verify proposed repairs.
Define the panel before reading the score
The panel is the set of questions, platforms and modes actually checked. It should reflect buyer decisions and keep branded questions separate from questions that do not name the business.
A result from five questions is a five-question observation. Do not describe it as the whole market or compare it with a larger panel without explaining the difference.
Work a small example correctly
Illustrative example only: 20 scheduled captures produce 16 completed answers. Twelve expose inspectable source lists. Four completed answers name the brand and two source-bearing answers cite its agreed domain.
Mention rate is 4/16, or 25%. Owned-domain source-citation rate is 2/12, or about 16.7%. Report four failed captures and four completed source-unavailable answers alongside the rates. They are not zero-citation observations.
Review what appeared instead
Preserve cited URLs and open the pages. Check whether they support the answer, which topic they cover and whether the intended business page answers the same question.
A competitor mention is distinct from a competitor-domain citation. A directory citation is distinct from a link to your own site. These differences change which source deserves review.
Choose the next action without inventing a cause
Independently check access, indexing where relevant, page relevance, current business facts and normal internal links. Fix a confirmed defect and rerun its technical check.
Treat unconfirmed content or profile explanations as experiments. Recheck comparable answers after the change, keep the original captures and state other changes that could affect the comparison. A citation-gap report does not guarantee rankings, leads or citations.
Check the platform guidance.
Continue with the useful next step.
Thomas Howcroft
Founder of TomHow and senior operator for SEO, GEO, AEO, and AI search evidence.
View LinkedIn profile ->