AI Share of Voice: How to Measure Your Brand Against Competitors in AI Answers
AI share of voice is the percentage of relevant AI answers that mention your brand compared with competitors. How to build the prompt set, score results and report it without guesswork.
On this page
Last updated: 30 October 2026
Quick answer: AI share of voice is the share of answers to a fixed set of buyer prompts that mention your brand, compared with named competitors, across the AI engines your buyers use. Measure it by running the same 30 to 100 prompts each month in ChatGPT, Gemini, Perplexity and Google AI Overviews, recording who is mentioned, in what order and which pages are cited, then dividing your mentions by all brand mentions.
Traffic from AI assistants is still small for most sites, and many AI answers produce no click at all. That makes rankings and sessions incomplete measures of AI visibility. Share of voice answers a more useful question: when buyers ask AI about your category, how often are you part of the answer, and how does that compare with the brands you compete with?
Key takeaways
- Use a fixed prompt set, so month-to-month changes mean something.
- Track several engines separately; results differ a lot between them.
- Record mentions, position, sentiment and cited URLs, not just yes or no.
- Report share of voice next to AI referral traffic and enquiries.
- Expect variation: run prompts more than once and use averages.
Step 1: build the prompt set
Start from real language. Sales calls, support emails, Search Console queries and community threads tell you how buyers phrase questions. Group prompts by intent so you can see where you are strong and weak. Our guide to AEO prompt research covers this in detail.
| Prompt group | Example |
|---|---|
| Category discovery | “Best AEO agencies in India” |
| Problem led | “How do I get my brand mentioned by ChatGPT?” |
| Comparison | “Agency A vs Agency B for SaaS companies” |
| Brand check | “What does Digital Tatva do?” |
| Local or segment | “AEO consultant for clinics in Ahmedabad” |
Step 2: decide what to record
- Mentioned: is your brand named at all?
- Position: first, among the top three, or later in the answer?
- Accuracy: is what the engine says about you correct?
- Sentiment: positive, neutral or negative framing.
- Citations: which URLs are linked, yours or third parties.
- Competitors: every other brand named in the answer.
Step 3: calculate share of voice
For each engine and prompt group, divide the number of answers that mention you by the total number of answers run, which gives a mention rate. For share of voice, divide your mentions by the total mentions of all tracked brands. Weight top-three mentions more heavily if you want the score to reflect prominence.
| Brand | Mentions (60 answers) | Share of voice |
|---|---|---|
| Your brand | 18 | 24% |
| Competitor A | 27 | 36% |
| Competitor B | 21 | 28% |
| Competitor C | 9 | 12% |
Illustrative example only, to show the calculation.
Step 4: report it monthly, with context
A useful report shows share of voice by engine and by prompt group, the answers where a competitor appears and you do not, the third-party pages most often cited, and any incorrect statements about your brand. Pair it with AI referral sessions from GA4 and enquiries that mention AI, as described in our AEO KPIs guide.
Manual tracking or a tool?
A spreadsheet works for 20 to 30 prompts across two or three engines. Beyond that, a tracker saves time and reduces human error. Whichever you use, keep prompts, locations and settings consistent between runs.
A simple tracking sheet layout
If you start manually, one row per prompt per engine per run is enough. Keep the columns the same every month so results can be compared.
| Column | Example value |
|---|---|
| Date and run number | 2026-10-15, run 2 |
| Engine and mode | ChatGPT, search on |
| Prompt and group | “Best AEO consultants in India”, category discovery |
| Brands mentioned, in order | Competitor A, Your brand, Competitor C |
| Your position | 2 |
| Accuracy of description | Correct, partly correct, wrong |
| Cited URLs | The pages linked in the answer |
| Notes | Anything surprising or worth fixing |
Turning the numbers into actions
- Low share in discovery prompts: you need more outside mentions on the sites engines cite for those questions.
- Low share in comparison prompts: build honest comparison and alternatives pages.
- Mentioned but described wrongly: fix your entity data and the profiles engines are reading.
- Strong in one engine, weak in another: check crawler access and the sources that engine prefers.
- Competitor cited from one page again and again: study that page and the questions it answers better than yours.
Want a baseline of your AI share of voice? See our AI Search Visibility service.
Frequently asked questions
How many prompts do I need?
Thirty to fifty well chosen prompts give a useful picture for most businesses. Larger brands with many product lines may need a hundred or more.
Why do results change when I ask the same question twice?
AI answers vary by wording, model version, location, personalisation and time. Run each prompt more than once and look at averages over months.
Which engines should I track?
The ones your buyers use. For most markets that means ChatGPT, Google AI Overviews and AI Mode, Gemini and Perplexity, with Copilot for B2B audiences.
Can share of voice be guaranteed to rise?
No. It can be improved by fixing the reasons engines leave you out, but no one controls what an AI system says.
Sources
- AI features and your website, Google Search Central
- Introducing AI Performance in Bing Webmaster Tools, Microsoft Bing, February 2026
Product details, prices and features are checked at the time of writing and can change. Sources are linked in the text so you can check them yourself.
How does your brand appear in AI answers?
We test the questions your buyers ask across the major AI assistants and search engines, then show you what to fix first.
Get an AI Visibility Audit