Citations
Every URL cited in your category, what each column means, and what to do with a row.
On this page
What are citations?
When an engine answers, it cites sources. Mentionova stores every one of them across all six channels — the LLMs, Google and Reddit — so the Citations tab is the whole picture of what your category is built on, not only the answers that named you.
URL view and Domain view
A two-button segment switches between URL and Domain. URL is one row per cited page; Domain aggregates by site and expands to show the pages underneath.
The columns
- Brand Referenced — whether your brand actually appears in the cited page's own text. It's read from the scraped page, not inferred from the answer, so it under-reports on sites that render in JavaScript.
- Brand Sentiment — the tone of the AI answer about your brand across the runs that produced this citation. It says nothing about the cited page's own tone.
- Influence — a 0–100 composite: 55% domain authority, 25% log-scaled citation frequency, 10% channel breadth, 10% sentiment lift. It's blank while domain authority is unknown.
- Domain Type — social, communities, media, products, reviews, docs, marketplaces, government, owned, competitors, other.
- Page Type — listicle, review, comparison, forum thread, video, doc, homepage, listing, other.
- Tags — competitor, reddit, news, blog, forum, government, docs, other.
Enrichment is asynchronous: rows show a dash until the classifier and the authority lookup land, and a badge tells you how many are still filling in.
Filtering and export
Search by URL, domain or title; filter by channel, tag, prompt or period; flip on Competitor only. Export CSV re-queries the raw rows with your current filters rather than dumping the aggregated view.
Acting on a row
Open the detail sheet for a citation and you can open the URL, or pull the people behind the domain with Get contacts — available on every non-Reddit row, not just competitors. Reddit rows hand off to Reddit Engagement instead.