---
title: "AI Visibility Score: What It Is and How It Is Calculated"
description: "An AI visibility score is the share of AI answers that name your brand. See the exact formula, a worked example, what a good score looks like and why it moves."
source_url: "https://brantial.ai/blog/what-is-an-ai-visibility-score-how-is-it-calculated/"
lang: "en"
updated: "2026-10-20T00:00:00.000Z"
---

Guides October 20, 2026

# What Is an AI Visibility Score and How Is It Calculated?

An AI visibility score is the share of AI answers that name your brand. See the exact formula, a worked example, what a good score looks like and why it moves.

## Key Takeaways

5

- An AI visibility score is the share of AI answers that name your brand: answers that mention you divided by all answers in the sample, as a percentage.
- The denominator is answers, not prompts. Every prompt asked on every engine on every run adds one answer, so the engine mix and the run count shape the score.
- A visibility score only says whether you were named. Average position and sentiment say where and how, and a single blended number hides which of the three moved.
- Good is relative to the category. In the October 2026 AI Sector Index, the most visible brand in 16 of 18 global sectors appeared in fewer than 14% of answers.
- Compare scores only when the prompt set, engines and time window stay the same. Change any of them and the number moves without the brand changing at all.

[Brantial Brantial editorial team](https://brantial.ai/authors/brantial/) 9 mins read

Share [LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fbrantial.ai%2Fblog%2Fwhat-is-an-ai-visibility-score-how-is-it-calculated%2F) [X](https://x.com/intent/post?text=What%20Is%20an%20AI%20Visibility%20Score%20and%20How%20Is%20It%20Calculated%3F%20https%3A%2F%2Fbrantial.ai%2Fblog%2Fwhat-is-an-ai-visibility-score-how-is-it-calculated%2F) [Facebook](https://www.facebook.com/sharer/sharer.php?u=https%3A%2F%2Fbrantial.ai%2Fblog%2Fwhat-is-an-ai-visibility-score-how-is-it-calculated%2F)

[Summarize with ChatGPT](https://chatgpt.com/?q=Summarize%20Brantial%20content%3A%20https%3A%2F%2Fbrantial.ai%2Fblog%2Fwhat-is-an-ai-visibility-score-how-is-it-calculated%2F)

![What Is an AI Visibility Score and How Is It Calculated?](https://brantial.ai/images/blog/what-is-an-ai-visibility-score-how-is-it-calculated/cover.webp) ![What Is an AI Visibility Score and How Is It Calculated?](https://brantial.ai/images/blog/what-is-an-ai-visibility-score-how-is-it-calculated/cover-dark.webp)

A marketing lead opens a dashboard on Monday and sees “AI visibility: 17%”. Last week it was 21%. Is that a problem, a fluke or a change in how the number was measured?

The answer depends on what the score actually counts. “Visibility score” is used for very different formulas across tools, and some of them blend three or four signals into one number that cannot be traced back to an answer. This guide explains the simplest and most useful version, how Brantial calculates it, how to read it and what makes it move.

## What is an AI visibility score?

An AI visibility score is **the percentage of AI answers in a sample that name your brand**. If ChatGPT, Gemini, Claude and other answer engines were asked 1,000 relevant questions and 170 of the answers mentioned you, your visibility score for that sample is 17%.

It answers one question: when people in your category ask AI for help, how often does your brand come up at all? It does not say whether you were recommended first, described positively or linked as a source. Those are separate measurements, covered below.

The [glossary entry for visibility score](https://brantial.ai/glossary/visibility-score/) covers the general concept. This article is about the arithmetic.

## The formula

```
AI visibility score = answers that name the brand ÷ all answers in the sample × 100
```

Three parts of that formula decide what the number means.

### 1. The unit is an answer, not a prompt

Each time a prompt is asked on one engine, it produces one answer. A prompt set of 40 prompts, measured on 5 engines once a day for 7 days, produces:

| Component | Value |
| --- | --- |
| Prompts tracked | 40 |
| Engines | 5 |
| Runs in the window | 7 (daily) |
| **Answers in sample** | **1,400** |

If 238 of those 1,400 answers name the brand, the visibility score is 238 ÷ 1,400 = **17%**.

This is why the engine mix matters. A brand that ChatGPT names in most answers but Claude rarely mentions will score differently depending on how many engines are in the sample. Brantial pools all answers in the window rather than averaging a score per prompt, so every answer counts the same.

### 2. A mention means the brand is named in the answer text

Brantial counts a mention when the brand is named in the answer with some context: a description, a comparison, an evaluation or a recommendation. A source link in the citations list on its own is not a mention; citations are tracked separately because a page can be used as evidence without the brand being named in the answer.

Each answer counts once per brand. An answer that names the brand five times still adds one to the numerator, so a single long answer cannot inflate the score.

### 3. The window sets the sample

The score always belongs to a time window. Brantial shows 7, 14, 30, 60 and 90-day views, with the last 7 days as the default. A 7-day score reacts quickly to change; a 90-day score is steadier and better for reporting to leadership. Comparing a 7-day number with a 30-day number is comparing two different samples.

## The two numbers that belong next to it

A visibility score treats every mention the same. Being named first as the recommended option and being named last as an alternative both count as one. Two more measurements fill that gap.

**Average position** is where the brand appears in the answer. Brantial reads the answer from top to bottom and numbers brands in the order they are first named, so the first brand mentioned is position 1. The average is taken only over the answers where the brand appears; answers that leave it out do not count as a low position.

**Sentiment** is how the brand is described. Each sentence about the brand is scored from 0 to 100, where 50 is neutral, and the scores are averaged.

Put together, the three can tell very different stories with the same headline number:

| Brand | Visibility | Avg. position | Sentiment | What it means |
| --- | --- | --- | --- | --- |
| A | 30% | 1.4 | 72 | Named often, usually first, described well |
| B | 30% | 4.8 | 51 | Named as often, but as a late, neutral option |

_Illustrative example._

Brand A and Brand B have identical visibility scores. Only one of them is winning the answer.

## What about single “AI scores”?

Many tools, including Brantial’s free [BrandScore](https://brantial.ai/brandscore/) check, also show one blended number for a quick read. BrandScore weights visibility at 65% and sentiment at 35% from a 10-prompt sample. A blended score is useful as a headline, but when it changes you cannot tell from the number alone whether the brand was mentioned less often or described less favourably.

For ongoing measurement, keep the components separate and treat any composite as a summary of them, never a replacement. If a tool cannot show the answers behind its score, the score cannot be acted on.

## What is a good AI visibility score?

There is no universal benchmark, because the score depends on the prompt set. A set full of questions that name your brand will produce a score close to 100% and say very little; this is why [branded and unbranded prompts should be reported separately](https://brantial.ai/blog/how-many-prompts-track-ai-visibility/).

What a real category looks like is easier to show. The [AI Sector Index](https://brantial.ai/ai-sector-index/) asks AI 500 unbranded questions per sector every month and counts the brands named in the answers. In October 2026 (one model, Claude Opus 5.5, without web search) the most visible brand in each sector scored:

| Sector | Most visible brand (global) | Visibility | Most visible brand (Türkiye) | Visibility |
| --- | --- | --- | --- | --- |
| Business software | Microsoft | 31.8% | Microsoft | 25.6% |
| Electronics and home appliances | Apple | 20.2% | Apple | 15.6% |
| Food delivery and restaurants | Uber Eats | 13.0% | Getir | 6.8% |
| Parcel delivery and logistics | UPS | 12.6% | PTT | 16.0% |
| E-commerce and marketplaces | Amazon | 11.6% | Trendyol | 9.4% |
| Beauty and skincare | La Roche-Posay | 11.6% | La Roche-Posay | 19.6% |
| Automotive | Toyota | 9.0% | Toyota | 14.8% |
| Banking | Revolut | 7.4% | İş Bankası | 10.0% |
| Telecoms | Starlink | 3.0% | Türk Telekom | 19.4% |
| Insurance | Allianz | 2.0% | Allianz | 3.8% |

_Source: Brantial AI Sector Index, October 2026. 500 answers per sector and market._

Two patterns stand out. First, category leaders rarely pass 20%, because broad questions spread answers across many brands: the global fashion answers named 341 different brands. Second, the same category can look very different by market. Türk Telekom appears in 19.4% of Turkish telecom answers, while no global telecom brand reaches 4%.

For your own brand, the useful benchmark is not an absolute number but your score against the competitors measured on the same prompts, engines and window.

## Why your score changes

A visibility score can move for reasons that have nothing to do with your brand. Before reacting to a change, check these in order:

1. **The prompt set changed.** Adding or removing prompts changes the sample. Brantial flags trend comparisons when the prompt set changed during the period.
2. **The engine mix changed.** Adding an engine that rarely names you lowers the pooled score. Compare engine by engine before drawing conclusions.
3. **The window is different.** A short window is sensitive to a few days of unusual answers.
4. **Normal answer variation.** The same prompt can return a different answer an hour later. A [stable prompt set and enough repetitions](https://brantial.ai/blog/understanding-prompt-volatility-why-ai-search-visibility-changes-over-time/) separate real movement from noise.
5. **Engines answer differently.** Each engine retrieves and weighs sources in its own way, which is [why AI platforms show different answers](https://brantial.ai/blog/why-ai-platforms-show-different-answers/) to the same question.

If none of these explain the change, look at the answers themselves: which competitors replaced you, and which sources the engine relied on.

## How to improve it

The score is an outcome. It improves when engines have better reasons to name you:

- Find the prompts where competitors are named and you are not, and read [why ChatGPT recommends competitors](https://brantial.ai/blog/why-chatgpt-recommends-competitors/) in those answers.
- Strengthen the pages and third-party sources engines cite for your category; see [strategies for getting cited by AI](https://brantial.ai/blog/how-to-get-cited-by-ai-strategies-for-maximizing-content-visibility/).
- Check how engines describe you, not just whether they do, with an [AI brand perception](https://brantial.ai/blog/ai-brand-perception-how-chatgpt-describes-your-brand/) review.

See the answers behind your score

Brantial measures how often AI answers name your brand and your competitors, where you appear and how you are described, by prompt and by engine, with every answer and source one click away.

[Measure your AI visibility](https://app.brantial.ai/register)

## Frequently asked questions

### Is an AI visibility score the same as share of voice?

Not quite. A visibility score is the share of answers that name your brand, so several brands can each have a high score at once. Share of voice divides mentions among brands, so the shares add up to 100%. Both are useful; they answer different questions.

### Does a citation count as visibility?

In Brantial, no. A brand is visible in an answer when the answer names it. Citations are measured separately, because an engine can use your page as a source without naming your brand, and it can name your brand while citing someone else.

### How many answers do I need for a reliable score?

More than a handful. A score from 10 answers moves 10 points with every mention. A few hundred answers across a stable prompt set gives a number that moves only when something real changes. The [prompt-count guide](https://brantial.ai/blog/how-many-prompts-track-ai-visibility/) covers how to size the set.

### Why is my score 0% on one engine and high on another?

Engines rely on different sources, training data and retrieval. A brand that is well represented in the sources one engine prefers can be missing from another. Look at the score per engine and at the sources each one cites.

### Can I compare my score with a competitor’s score from another tool?

Only if both were measured on the same prompts, engines and window. A score is only meaningful within the sample that produced it.

↑ Back to top