AI Conversation Citation Rate Analysis

Citation Rate Analysis shows how often AI platforms support appearances of Your Brands with cited sources.

It helps you distinguish an unsupported brand mention from an answer in which the monitored-brand provider record has one or more citations attached.

What the chart measures

For each AI platform and monitoring date, Wheremore calculates:

latest monitored responses containing Your Brands with an attached citation ÷ latest monitored responses containing Your Brands × 100

The result is a response-level citation rate from 0% to 100%.

A response enters the calculation only when:

one of the names saved under Your Brands appears as a structured provider;
the provider satisfies the selected Provider Position filter; and
the response is the newest processed attempt for that configuration, question, platform and monitoring date.

Each qualifying response counts once, even if it contains several Your Brands aliases or several citations.

Example

Suppose ChatGPT has 20 qualifying responses on one monitoring date:

Your Brands appears in all 20 responses;
15 of those responses have at least one citation attached to the monitored-brand provider appearance; and
those 15 responses contain 31 distinct attached citations.

The chart displays 75%.

The tooltip displays:

ChatGPT: 75.0% · 15 of 20 eligible responses cited sources

31 distinct citations in total

The 31 citations do not form the denominator. The metric measures cited responses, not the average number of sources per response.

Citation rate and citations to your domain are different

Citation Rate Analysis does not mean that the AI platform linked to your own website.

It measures whether the structured appearance of Your Brands has at least one source attached. That source might be:

your own website;
an independent publication;
a review or comparison site;
a partner;
a competitor; or
another publicly available source.

To check whether an answer links specifically to your website:

Open AI Platform Conversation Responses.
Select the relevant project, configuration, question and platform.
Use Cites your domain.
Inspect the displayed citations and source URLs.

Use Citation Rate Analysis for sourcing frequency. Use Cites your domain for owned-domain citation.

Before data appears

The selected project needs:

An AI Visibility monitoring configuration.
Monitoring questions assigned to journey phases.
One or more selected AI platforms.
Completed and processed monitoring responses.
At least one company or product name under Your Brands in AI Visibility Settings.
Structured provider records containing a saved brand name.
Structured citation records attached to those provider appearances for the numerator to rise above 0%.

If Your Brands has not been configured, the chart asks you to add brand names.

Quick start

Open the AI Visibility Dashboard.
Select a project.
Choose a date range.
Leave Journey Phase, Configuration and Platform set to all values for the broadest view.
Set Provider Position to Mentioned Anywhere.
Compare citation rates across the platform lines.
Hover over a measured point to see cited responses, eligible responses and attached citations.
Narrow the filters to investigate a phase, audience configuration or platform.
Compare Mentioned Anywhere with Top 3 and #1 Recommendation.
Open AI Platform Conversation Responses to inspect the answers and sources behind an important movement.

Apply the dashboard filters

Project

Project selects the monitoring configurations, processed responses and Your Brands settings used by the chart.

Only projects available in the active workspace can be selected.

Date Range

Choose a preset period or a custom start and end date.

Dates are interpreted as UTC calendar dates. Both custom boundaries are included. If custom boundaries are entered in reverse order, Wheremore uses the earlier date as the start.

Last 30 Days means today and the preceding 29 UTC dates.

Journey Phase

Choose all phases or restrict the chart to:

Awareness;
Consideration; or
Decision.

This helps reveal whether sourcing changes as a buyer discovers the problem, evaluates alternatives or approaches a final choice.

Configuration

Choose all configurations or one monitoring configuration.

Use this to isolate a specific audience, market context, product or question set.

Disabling a configuration stops future scheduling but does not remove its completed historical responses from this chart.

Platform

Choose all platforms or one platform.

The chart supports:

ChatGPT;
ChatGPT Search;
Claude;
Claude Search;
Gemini;
Gemini Search;
Perplexity; and
Perplexity Search; and
Google AI Mode.

When All Platforms is selected, each platform is shown as a separate line.

Provider Position

Provider Position restricts which Your Brands appearances qualify:

Mentioned Anywhere — accepts Your Brands at any valid provider rank.
Top 3 Recommendations — accepts ranks 1, 2 and 3.
#1 Recommendation — accepts only rank 1.

This changes the eligible response population. It does not simply zoom the same data.

For example, #1 Recommendation answers: “When Your Brands is the first structured provider, how often is that appearance supported by citations?”

Read the chart

Each platform row uses this format:

ChatGPT: 42.9% · 3 of 7 eligible responses cited sources

12 distinct citations in total

The hard numbers explain the result:

3 is the number of eligible responses containing at least one citation.
7 is the total number of eligible Your Brands responses for that platform and date.
42.9% is 3 divided by 7, expressed as a percentage.
12 is the number of distinct citation records across the eligible responses. It provides context but is not used to calculate the rate.

The citations can point to any source. They do not necessarily cite or endorse Your Brands or link to your own domain.

A rising line — a larger share of qualifying Your Brands responses has attached citations.
A falling line — more qualifying brand appearances are being returned without attached citations.
A high rate with few eligible responses — encouraging, but supported by a small evidence base.
A lower rate with many eligible responses — a broader and potentially more reliable view of sourcing behaviour.
Different results by platform — platforms may differ in retrieval, grounding and citation behaviour.
Different results by journey phase — informational questions may attract more sourcing than recommendation or decision questions.
A higher rate at #1 than Mentioned Anywhere — leading recommendations are sourced more often than general appearances.
A lower rate at #1 than Mentioned Anywhere — the strongest recommendations are less consistently supported than broader mentions.

Always use the tooltip counts alongside the percentage.

What 0% means

A genuine 0% point means:

at least one eligible response contained Your Brands; and
none of those eligible responses had a citation attached to the monitored-brand provider appearance.

It does not mean the platform produced no answers.

It also does not mean the full answer contained no citations elsewhere. Citations attached only to competitors or other providers do not make the monitored-brand response cited for this metric.

Monitoring days and missing dates

Monitoring may run only on selected weekdays.

A platform and date with no eligible Your Brands responses is not plotted as 0%. It is treated as no observation.

The chart joins neighbouring measured dates to keep the trend readable. The connecting line is visual continuity only; it does not create a measurement for an unscheduled or unobserved date.

There is an important difference:

0% citation rate — eligible responses existed, but none had attached citations for Your Brands.
No observation — no eligible responses existed under the selected filters.

Latest response and retry handling

Provider calls may be retried.

For the same configuration, question, platform and monitoring date, the chart uses only the newest processed run. An older failed, incomplete or superseded attempt does not create an additional observation.

This prevents retries from inflating the eligible or cited response counts.

How Your Brands settings affect history

The chart applies the current Your Brands aliases to stored provider records.

Adding, removing or correcting an alias can therefore change which historical responses qualify. Current aliases do not rewrite the underlying historical response.

Use exact company and product names that match the structured provider names returned by monitored platforms. Avoid broad terms that could describe unrelated providers.

Add annotations

Select a date inside the chart to add a short annotation.

Useful annotations include:

changes to Your Brands aliases;
edited monitoring questions or audience context;
adding or removing an AI platform;
a launch or positioning change;
a major content or website update;
digital PR or campaign activity;
publication of new supporting evidence; or
a change to processing days.

Select an annotation label to delete it.

Annotations provide context only. They do not change responses, providers, citations or calculations.

What the chart does not measure

Citation Rate Analysis is not:

the percentage of all monitored answers containing any citation;
the percentage of answers linking to your own domain;
the number of citations per answer;
the authority or quality of a source;
proof that a citation supports a particular claim correctly;
Share of Voice;
Competitive Win Rate;
sentiment;
organic search visibility, rankings or traffic;
referral traffic from AI platforms;
revenue or conversion contribution; or
proof that a content or campaign change caused a movement.

It measures the response-level presence of attached citation records for qualifying monitored-brand provider appearances.

What should I do next?

Open AI Platform Conversation Responses to inspect the answer and its sources.
Use Cites your domain to isolate answers linking to your own website.
Compare Share of Voice to see whether improved sourcing accompanies greater visibility.
Compare Competitive Win Rate to see whether cited appearances also rank above competitors.
Review Conversation Sentiment Analysis to understand how the brand is presented.
Use Key Item Tracker to identify concepts consistently associated with sourced answers.
Review AI Visibility Settings if brand aliases are missing or too broad.
Review Manage AI Visibility Tracking if questions, phases, platforms or processing days need changing.
Strengthen first-party evidence, statistics, research, product documentation and clearly attributable claims where important questions remain unsupported.

Data timing

The chart is not real time.

Monitoring runs on each configuration’s selected processing days. Results appear after provider calls finish and raw responses have been converted into structured runs, provider appearances and citations.

The newest scheduled date can be incomplete while jobs remain pending, running, saved as raw responses or waiting to be processed.

Google AI Mode sources normally belong to the answer rather than a named provider. Compatible citation reporting can use this answer-level evidence, while provider-linked brand citation metrics still require a genuine provider relationship. Own-domain reporting uses the resolved source destination.

Troubleshooting

Select a project appears

Select a project in the global dashboard controls.
Confirm the active workspace is correct.
Confirm you have access to the project.
Allow the project selector to finish loading.

Add at least one of your brand names appears

Select Configure on the AI Visibility Dashboard.
Add the required company and product names under Your Brands.
Save the settings.
Reload the chart if it does not refresh automatically.

No citation data appears

Increase the date range.
Return Journey Phase, Configuration and Platform to their all-values options.
Set Provider Position to Mentioned Anywhere.
Confirm the selected configuration has completed monitoring runs.
Confirm saved aliases match the structured provider names.
Open AI Platform Conversation Responses to verify Your Brands appears.

The citation rate changed sharply

Hover over both dates and compare their eligible-response totals.
One additional cited response can produce a large movement when the denominator is small.
Confirm the project, filters, Provider Position and Your Brands aliases are unchanged.

The chart shows 0%

This is a measured result. Eligible responses contained Your Brands, but none had a citation attached to the monitored-brand provider appearance.

Inspect the responses to see whether:

the answer contains no sources;
citations are attached only to other providers;
the provider structured data is incomplete; or
the platform returned citations in an unsupported format.

The chart is high but Cites your domain finds nothing

The two reports answer different questions.

Citation Rate Analysis accepts a citation to any source attached to the monitored-brand appearance. Cites your domain requires the citation URL to match the configured website.

This is expected. One cited response can contain several distinct citations.

The chart connects across days without monitoring

This is expected. The connection joins measured observations for readability and does not create intermediate data.

A platform line is missing

Confirm the platform is selected in the monitoring configuration.
Confirm completed responses exist for the selected period.
Confirm Your Brands appears under the selected Provider Position.
Check whether platform jobs are still pending, running, saved as raw responses or in error.

The chart changed after editing Your Brands

This is expected because current aliases are matched against historical structured provider records. Add an annotation at the settings-change date.

Responses could not be loaded

Reselect the project or filters.
Refresh the page.
Confirm your session is still active.
Try a shorter date range.
If the problem continues, record the selected filters and time of the error for support.