Which AI visibility metrics matter, and which tool reports them?

A marketing manager who picks a tool asks which metrics it must report. An agency owner asks which number goes into the client report. This guide lists the seven metrics that matter and the decision each one supports. It then sets four of them side by side on our own project, with the numbers of its last 28 days. It also shows what each tool's own pages say it reports. For the formula behind one visibility score, read our guide on the visibility score. For what a tool does before it counts, read what an AI visibility tool is.

What is the best platform for AI visibility metrics?

The best platform for AI visibility metrics names the denominator of every number and opens each number into the answers it counts. In the 28 days to 30 October 2026, on 7 AI platforms, our most visible rival trailed the second by 12 points on recommended mentions. Visibility is the share of analyzed answers that name the brand at least once. Share of Voice is the brand's mentions divided by the mentions of every brand you monitor, your own included. Our guide on what an AI visibility tool is defines sentiment and recommendation status.

Citation share is the share of every cited URL in the window that points to one domain. Position is the place of a brand among the brands one answer names, first to last, counted over the answers that name it. Perception is the set of themes and objections that the answers attach to a brand, read from questions about that brand alone. Each theme scores 0 to 100 by how early it appears, averaged over every analyzed answer about the brand.

The table lists the decision each metric supports.

Which metrics look alike and count differently?

Four pairs look alike: visibility and Share of Voice, sentiment and recommendation status, citations and mentions, position and visibility. Read the denominator before you compare the two numbers of a pair.

Visibility and Share of Voice. Our guide on Share of Answer and Share of Voice states the denominator of each and works both out by hand on 20 answers. It then shows both for every brand in our project over 28 days. Our guide on why numbers differ shows how a share moves when the rival list changes and no answer does.

Sentiment and recommendation status. Sentiment reads the tone of a mention. Recommendation status reads the advice. In the 28 days to 30 October 2026, on 7 AI platforms, one rival in our project had 38% positive mentions and 31% recommended mentions. Another had 48% and 43%.

Citations and mentions. A citation is a link that the answer shows. A mention is a name in the text. In the same 28 days, the answers cited fullreach.ai 45 times, in 24 answers, in 174th place among the cited domains. Visibility counts none of those citations, only the answers whose text names the brand. The second most cited domain in our project was semrush.com, with 1,807 citations, 3.6% of all citations. The project does not monitor Semrush as a competitor, so that domain adds nothing to any Share of Voice.

Position and visibility. Position counts only the answers that name the brand. A brand that appears in few answers can lead every list it appears in. Our guide on why numbers differ counts one pair of brands five ways on the same answers. Two of the ways rank by position: first-named answers, and a Share of Voice weighted by position.

Perception reads a separate set of weekly questions that name the brand, so it shares no answer with the four pairs. Only an answer that describes the right company counts. In the 28 days to 30 October 2026, the platforms answered the question about FullReach AI 216 times. 120 answers described our company, 84 described another company, and 12 named no company or could not be judged. The 96 neither score nor dilute.

What do the metrics show side by side on one project?

The same five rivals rank differently on sentiment, on recommendation and on visibility. Our own project tracks the category we sell in, AI visibility tools, and monitors five rivals. The table lists them in the order of their visibility. The table pools the 28 days to 30 October 2026 on 7 AI platforms. Visibility divides by the analyzed answers. The two shares divide by each brand's classified mentions, and the last column counts mentions. Otterly AI's 51% is 1,670 recommended of its 3,270 classified mentions. The Share of Voice of the same brands stands in the table of our guide on Share of Answer and Share of Voice.

Profound, the most visible rival, leads Otterly AI on visibility by 4 points, 41% to 37%. Profound ranks fourth of five on positive mentions and on recommended mentions. Otterly AI leads both columns. On positive mentions it is 2 points ahead of AthenaHQ and 9 points ahead of Profound. On recommended mentions it is 7 points ahead of Peec AI and 12 points ahead of Profound. Profound also collected the most mentions that advised against it, 33 against 7 for Otterly AI. A report that shows visibility alone ranks Profound first. A report that shows recommendation alone ranks Otterly AI first. The two reports answer different questions: how often an answer names the brand, and what the mention says.

Perception adds a further ranking that none of these columns predicts. For Scrunch, the objection "Pricing" scored 83 of 100, from 72 mentions in the same 28 days. That is the highest pricing objection of the six brands, for the brand that ranks last on recommended mentions. For Peec AI, the attribute "AI Search & Visibility Tracking" scored 89, from 125 mentions. That is the highest attribute score of the six brands, for the brand that ranks third on visibility.

Brand Visibility›Competitors

Example

Top 5 brands

#	Brand	Visibility	Share of Voice	Sentiment	Recommended
1	HubSpot	57.8%▲0.7 pts	31.0%	66%	58%
2	PipedriveYou	41.1%▲5.8 pts	22.0%	73%	52%
3	Salesforce	39.0%▲0.2 pts	20.9%	71%	49%
4	Zoho CRM	27.3%▼0.3 pts	14.6%	58%	37%
5	monday CRM	21.4%▼0.7 pts	11.5%	62%	41%

Which tool reports which metric?

The table below quotes what each tool's public pages say it reports, and each row names the day we read it. We read the rival pages between 24 September 2026 and 13 October 2026. Our own row names visibility, Share of Voice, sentiment, recommendation status, cited sources and perception, and no position. Compare each row with the seven metrics above before a demo. A page that names no metric may still report it, so the table reads the pages, not the products. Our roundup of AI visibility tools compares the same tools on platforms, fit and limits.

When two tools report the same metric, the guide on why numbers differ measures how far their counting choices move one brand's figure.

Which plan carries each metric?

On FullReach AI, every plan reports visibility, Share of Voice, sentiment, recommendation status and cited sources. Perception and objections come from the Pro plan. Every number opens into the answers it counts, on every plan. Our pricing guide prices every plan we compare for ChatGPT, Gemini and Perplexity together. FullReach AI costs 65€, 165€ and 360€ a month, billed yearly, for 50, 150 and 350 prompts, with every platform included.

What to do next

1. Write the decision first: whether buyers see you, whether you beat the rivals you chose, or what the answers say about you.
2. Pick the one metric for that decision. Write its denominator beside it in every report.
3. Before you compare two weeks, read our guide on how accurate the numbers are.

For how FullReach AI collects the answers, read the methodology.