All posts
SEO & AI search

What Is AI Share of Voice? A Plain-English Definition

What is AI share of voice? A plain definition of the AI visibility metric, how to calculate it by hand with a prompt list, and how to read the result honestly.

By Mehdi Hassan4 min read

The short answer

AI share of voice is the percentage of AI-generated answers, across a fixed set of relevant prompts, that mention or cite your brand compared with competitors. If ChatGPT, Perplexity and Google's AI features name you in 6 of 20 buyer prompts, your AI share of voice for that set is 30%.

Key takeaways

  • AI share of voice counts how often answer engines name you across a fixed list of buyer prompts.
  • You can calculate it by hand with a spreadsheet, 20–30 prompts and a monthly check.
  • Track mentions and citations separately. Being named and being linked are different wins.
  • Off-site mentions drive it more than backlinks, so it's as much a PR metric as an SEO one.

Part of Web Design, SEO and AEO Glossary: 50 Terms Explained Simply

What is AI share of voice?

AI share of voice measures how visible your brand is inside AI answers for the questions your buyers ask. Classic share of voice compared ad spend or rankings. The AI version compares how often you appear in the answer itself, against the competitors who appear instead.

It matters because more B2B buyers now start research in a chatbot, and the brands an answer names become the shortlist. A brand that never appears is missing before anyone visits a website. It's one of the newer terms in our SEO and AEO glossary.

How do you calculate AI share of voice manually?

You calculate AI share of voice by running a fixed prompt list through each engine, recording which brands each answer names, and dividing your mentions by the total answers. Tools automate this, but a manual version is accurate enough to start.

  1. Write 20–30 buyer prompts

    Use the questions real prospects ask: "best invoicing tool for agencies", "X vs Y", "how much does Z cost". Mix category, comparison and problem prompts.

  2. Run them in each engine

    Ask ChatGPT, Perplexity, Gemini and Google's AI features the same prompts, in fresh sessions, on the same day each month.

  3. Log every brand named

    In a spreadsheet, note for each answer whether you were mentioned, whether you were cited with a link, and which competitors appeared.

  4. Divide and compare

    Your mentions divided by total answers is your share of voice. Do the same for each competitor so the number has context.

What does a share of voice example look like?

A worked example makes the metric concrete. The numbers below are invented to show the arithmetic, not real results.

Hypothetical example: a B2B HR software company's monthly audit

For example, suppose an HR software startup runs 20 prompts in four engines, giving 80 answers. It's mentioned in 12 answers and cited with a link in 5. Its main competitor is mentioned in 30.

Its AI share of voice is 12 out of 80, or 15%, against the competitor's 37.5%. The team notices most of its mentions come from Perplexity and almost none from Gemini, which uses fewer sources. That tells it where the gap is, and which comparison prompts to target with new pages and outreach.

Metric

What it counts

Why track it

Mention rate

Answers that name your brand

Shortlist presence

Citation rate

Answers that link to your site

Potential traffic

Share of voice

Your mentions vs all brands named

Position against competitors

Sentiment

Whether the answer describes you accurately

Catching wrong facts early

What moves AI share of voice?

AI share of voice moves mostly with how often other sites mention you, and with how clearly your own pages answer buyer questions. In Ahrefs' study of 75,000 brands, web mentions correlated 0.664 with AI Overview visibility, against 0.218 for backlinks.

For the full measurement setup, including citations and referral tracking, read measuring AI search visibility. Acting on what the audit finds, with new pages and clearer copy, is the kind of ongoing design and development work our $6,000/month retainer covers.

Frequently asked questions

Is AI share of voice a reliable metric?
It's directional, not precise. AI answers vary between sessions, users and days, so one run can differ from the next. A fixed prompt list, checked on a schedule across several engines, smooths out much of that noise. Treat changes over a quarter as meaningful and changes from one week to the next as mostly random.
Which tools measure AI share of voice?
Several SEO platforms, including Semrush and Ahrefs, now offer AI visibility reports, and dedicated AI tracking tools exist too. They differ in which engines and prompts they cover, so numbers don't match across tools. A manual spreadsheet audit is a good baseline before paying for one, because it shows you which prompts actually matter.
How many prompts should I track?
Twenty to thirty prompts is a practical starting set for a startup. That's enough to cover your category, your main comparisons and the problems you solve, while staying small enough to check by hand each month. Add more only once you know which prompts bring real buyers, rather than tracking hundreds of loosely related queries.
Does AI share of voice affect revenue?
Indirectly. Buyers increasingly build shortlists from AI answers, so being absent can mean losing deals you never knew about. To connect it to revenue, add an AI channel group in your analytics and a "How did you hear about us?" field to your forms, then compare those leads with your share of voice over time.

· CEO & Design Director

Copywriter, designer and developer, and founder of Designify Studio. Over five years, 60+ projects across 14+ industries, for startups that have raised $300M+.

Want this done for you? Monthly retainer, $6,000.

Your ongoing design and development team, for products that keep shipping. $6,000, per month, flat fee. Book a free 30-minute call to see if it fits.

Schedule Call