Diagnose

Source & citation analysis

Which domains AI assistants actually cite when they answer questions in your category — the real answer to what to get featured on.

  • Every citation parsed out of every answer and attributed to its domain.
  • Ranked by how often each domain supports an answer in your category.
  • Your own domain scored the same way, so you can see whether your pages are doing any work.

Map the sources that shape the answer

Recommendations are supported by evidence. Knowing which domains supply that evidence tells you where being present would actually change the outcome.

See it in the live demo

Cited domains

Which sites the AI platforms draw on when they answer.

#DomainShare of citations
1
g2.com
31.22%
2
vercel.comYou
17.56%
3
render.comCompetitor
13.17%
4
netlify.comCompetitor
12.2%
5
railway.appCompetitor
11.22%
6
aws.amazon.comCompetitor
7.8%
Sample data for a fictional brand. Not measured, and these brands are not customers.

How it works

01

Collect cited sources

Citations are parsed from each answer as it lands and stored against the run they came from, with the domain extracted.

02

Rank by influence

A nightly rollup counts how often each domain appears across your tracked questions, so repeated sources rise to the top.

03

Find the gaps

Domains that support competitors but never cite you become concrete targets rather than a vague content plan.

What to act on

A metric that does not change a decision is a metric worth deleting. These are the ones that do.

A review site dominates your category's citations.
Being absent or poorly rated there is costing you answers regardless of how good your own pages are.
Your own domain barely appears.
Your pages are not being treated as evidence. That is a content and structure problem you can act on directly.
A forum thread keeps getting cited.
Unowned sources shape recommendations too. Knowing which ones lets you decide whether to engage.

Keep the loop connected

View all features

Map the sources that shape the answer

Enter your domain and start tracking. One URL sets it up, and it runs every day on its own.