Citation tracking

AI citation tracking: find the pages briefing the answer

When an assistant cites its sources, it is showing you what shaped the answer. We record every cited URL for every question, rank the domains that keep appearing, and flag the answers that cite your page while naming somebody else.

Start 7-day free trialAll seven features

Card required, nothing charged for 7 days

Google · “ai search visibility tool”

AI Overview

AI search visibility tools measure how often assistants such as ChatGPT and Gemini mention a brand. Options include AI Mention Tracker, which reports each engine separately with sample sizes…

aimentiontracker.aisearchengineland.comg2.com

Collected from the live result page, so it moves when Google moves.

Illustrative figures from a demo workspace, not a customer's data.

Cited and named are two different questions

Most tools collapse these. The four combinations need four different responses.

  1. 1

    Cited and named

    Your page is a source and your brand is in the prose. The outcome you are working toward.

  2. 2

    Cited, not named

    You briefed the answer and a competitor got the credit. Retrieval without framing.

  3. 3

    Named, not cited

    The assistant knows you without reading you today. Durable, and harder to influence.

  4. 4

    Neither

    You are outside the conversation on this question. The domains cited instead tell you where to start.

The awkward finding

The pages describing you are mostly not yours

Run this for a week and the pattern is usually the same: the domains citing hardest in your category are review sites, comparison posts and forum threads, not vendor sites. Which means the description an assistant gives of your product is assembled largely from pages you do not control. Seeing the ranked list is uncomfortable and it is also the most actionable thing on this site — it is where a recurring criticism usually turns out to live.

Every URL, every answer.
The full citation list per answer, not a sample, with rank and domain.
Domains ranked.
Who briefs your category most often, across all six engines.
The answer kept too.
Check whether the cited page actually supports the claim made about you.
Answer · Perplexity · “best project tracker for agencies”

For agency work the strongest options are Northstar, Planwise and Tandem. Northstar handles multi-client billing best, while Planwise has the clearest approval flow for client sign-off…

Sources cited

  1. northstar.com/agenciesnamed
  2. yourbrand.com/blog/agency-billingcited, not named
  3. planwise.io/approvalsnamed
  4. g2.com/categories/project-trackingthird party
Illustrative figures from a demo workspace, not a customer's data.

Citations are not rankings, and the difference matters

A cited page is not a ranked page, and optimising for one does not reliably get you the other.

A search result is a ranked list of documents, ordered, positional and reasonably stable. A citation panel is a set of pages the model consulted while writing prose, unordered in any meaningful sense and regenerated each time. A page can be cited constantly and rank nowhere, or rank first and never be cited. Treating a citation panel as a leaderboard produces confident, wrong conclusions.

Google's AI Overviews make this especially visible, which is why AI Overview tracking reports the citation panel separately from the prose, and why visibility is reported as a rate rather than a position. Our methodology sets out what we will and will not infer from a citation.

  • No citation rank invented. We record the order the engine returned and do not dress it up as a position.
  • Absent citations recorded as absent. An engine that cited nothing is not an engine that cited you.
  • Everything stored. Answers and citations kept from your first day, so you can re-read them later.

Common questions about AI citation tracking

What people ask before they build a plan around citations.

What is AI citation tracking?
Recording which web pages an assistant links to when it answers a question. Assistants that browse show their sources, and those sources are the raw material of the answer: the pages being cited are the pages shaping what gets said about your category.
Can my site be cited without my brand being named?
Yes, and it is more common than people expect. The citation panel and the answer text are separate. An assistant can read your page, use what it learned, and then name a competitor in the sentence the buyer actually reads. You have the retrieval and not the framing, and we report it as its own state because the fix is different.
Which domains get cited most in my category?
You will usually find it is not the vendors. Review sites, comparison posts, forum threads and documentation tend to dominate citation panels, which means the pages deciding how you are described are frequently not pages you own. We rank the domains so you can see who is actually briefing the model.
Do all engines show their sources?
Not equally. Engines that browse the live web cite heavily; ones answering from training data may cite nothing at all. We record what each engine actually returns rather than inferring sources it did not give, and the platform pages say what to expect from each.
How do I use this to get cited more?
Look at which domains are cited for the questions where you are absent, and work on being represented on those pages. If three comparison sites brief every answer in your category, being accurately described on those three matters more than another post on your own blog.
Is the full answer kept alongside the citations?
Yes. Every answer is stored in full with its citation list, and both are searchable. That is what lets you check whether a citation actually supported the claim made about you, or whether the assistant cited a page and then said something the page does not say.

The rest of what we measure

Six more, all on every plan.

See who is briefing your category

Add your questions and read the first citation lists within minutes. Card required, nothing charged for 7 days.

Start 7-day free trialSee how it works