# How Ahrefs Brand Radar Measures AI Visibility

Brand Radar compresses several different observations into one dashboard, but each metric has a different denominator and business meaning.

Canonical URL: https://truborankai.com/blog/ahrefs-brand-radar-ai-visibility-metrics

Author: TruboRankAI
Published: August 16, 2026
Updated: August 16, 2026

## Quick Answer

Brand Radar counts a mention when a tracked entity appears in an AI response and a citation when a page or domain is cited at least once. Its impressions are modeled from Google search volume associated with the prompt, not measured views inside the AI assistant. AI Share of Voice is the brand's percentage of those modeled impressions compared with the configured competitors, so it is directional and depends on entity setup.

## AI Summary

This guide defines the four headline Brand Radar metrics, shows which questions each can answer, explains the search-backed impression model, and provides a scorecard that keeps response visibility, source evidence, referrals, and conversions separate.

## Main Explanation

Mentions answer a presence question: did the configured brand appear at least once in the response? Ahrefs documents that repeated uses of the same brand inside one response still count as one mention for that response. This is useful for filtering answers and comparing coverage. It does not reveal whether the description was positive, accurate, prominent, or persuasive without reading the answer.

Citations answer a source question: did an AI response cite the page or domain? At domain level, multiple pages from the same domain in one response are counted as one domain citation. Ahrefs also distinguishes pages that were found during answer generation but not selected as inline citations. That distinction can reveal source discovery, but neither state proves that a human clicked or that the source influenced a purchase.

Impressions require the most care. Ahrefs says it sums Google search volume for prompts where the brand appears in an AI answer. For a prompt linked to People Also Ask, it uses the highest-volume keyword whose search results show that question. This creates a consistent demand-weighted model, but it is not telemetry from ChatGPT, Gemini, Perplexity, Copilot, or Google AI experiences. Label it “modeled impressions” in internal reports.

AI Share of Voice divides the brand's modeled impressions by the comparison set. Across multiple platforms, Ahrefs uses a weighted average where platforms with more impressions receive more weight. The number can move because the brand gained coverage, a competitor lost coverage, the entity rules changed, the prompt index changed, or the selected platforms changed. Preserve the report configuration before interpreting a trend.

No single metric represents AI visibility completely. A useful measurement stack has four layers: access evidence from crawler and server checks; response observations from a fixed prompt sample; citation and source evidence; and business outcomes from identifiable referrals, assisted conversions, leads, or sales. Search Console adds Google search demand and click context but does not currently turn every AI answer into a distinct traffic channel.

Use Brand Radar metrics to prioritize questions, not to award a vanity score. A high-share query with inaccurate brand description may need entity clarification. A cited competitor page may expose missing original evidence. A mention with no referral may still help awareness, but its value should not be invented. Tie each finding to a hypothesis and a page-level action that can be verified.

## Metric interpretation scorecard

| Metric | Good question | Required context | Separate outcome |
| --- | --- | --- | --- |
| Mentions | Which tracked responses name us? | Entity rules, prompt, platform, date | Accuracy, sentiment, referral, conversion |
| Citations | Which pages or domains are selected as sources? | Exact response, cited URL, platform, date | Click and assisted conversion |
| Modeled impressions | Which mentioned prompts carry more search-backed demand? | Keyword mapping and Ahrefs method | Observed assistant views |
| AI Share of Voice | How do configured brands compare under one setup? | Competitor set, platforms, dates, filters | Absolute market share |

## A reporting sentence that avoids overclaiming

Prefer: “In the fixed Brand Radar report for these entities, platforms, and dates, our share of modeled impressions increased from X to Y; identifiable AI referrals and conversions are reported separately.” Avoid: “AI traffic grew by Y” unless analytics actually measured that traffic.

## Practical Steps

1. Document entity and competitor rules.
2. Freeze platforms, filters, dates, and prompt groups.
3. Label impressions as modeled search-backed demand.
4. Read the answer behind important mentions.
5. Open and classify cited and found-but-not-cited pages.
6. Track identifiable referrals and conversions separately.
7. Explain configuration changes beside every trend.
8. Connect each material gap to one verifiable action.

## FAQ

### Are Brand Radar impressions real AI assistant views?

No. Ahrefs documents them as a model based on Google search volume associated with prompts where the brand appears.

### Can AI Share of Voice change when competitors are edited?

Yes. It is a comparative metric, so the configured entities and competitor set affect the denominator.

### Does a citation mean someone visited my website?

No. A citation shows source selection in an observed response. Referral analytics are required to identify a visit.

### What is the best metric for AEO?

Use several: response presence and accuracy, citations, source gaps, crawl and page readiness, referrals, and conversions. The right mix depends on the business question.

## Editorial Methodology

Definitions follow Ahrefs' current metric documentation. Editorial interpretation tests each metric against its documented numerator, denominator, and collection method, then separates modeled visibility from observed referrals and conversions.

## Sources

- [Ahrefs Help: mentions, citations, impressions, and AI Share of Voice](https://help.ahrefs.com/en/articles/15501968-ai-visibility-metrics)
- [Ahrefs: Brand Radar prompt and data methodology](https://ahrefs.com/blog/brand-radar-methodology/)
- [Ahrefs Help: supported reports and data sources](https://help.ahrefs.com/en/articles/11064852-what-is-brand-radar-and-how-to-use-it)

## Related Internal Links

- [Ahrefs Brand Radar Guide](/blog/ahrefs-brand-radar-guide)
- [Set Up a Brand Radar Report](/blog/ahrefs-brand-radar-tutorial)
- [Measure AI Search Visibility](/blog/measure-ai-search-visibility)
- [Mentions vs Citations vs Referrals](/blog/ai-mentions-citations-referrals)
- [AI Bot Traffic Tracking](/blog/ai-bot-traffic-tracking)
- [Audit AI Visibility Foundations](/ai-visibility-tool)