Skip to content
seenrixFree check

Guide, updated 31 October 2026

How to benchmark your AI visibility against competitors

Knowing whether ChatGPT names you is half the picture. The useful question is who it names instead, and how often. Here’s a method you can run in an afternoon, with a worked example from real data.

The method

Six steps. The first round takes about two hours by hand.

  1. 1

    Pick 3 to 5 competitors

    Use the businesses you lose deals to, plus anyone AI assistants already name for your category. Our free check lists who gets named instead of you, which is often a surprise.

  2. 2

    Write 10 buyer questions

    Questions a customer would ask before choosing, without any brand names: “best X for Y”, “X near Z”, “alternatives to the market leader”. Mix broad and specific ones. Keep them fixed so later rounds are comparable.

  3. 3

    Ask every question in every assistant

    ChatGPT, Gemini and Claude at minimum, in a fresh chat each time. Ten questions across three assistants gives 30 answers, enough to see a pattern.

  4. 4

    Record who is named, and where

    For each answer note every brand named and its position. A simple sheet with one row per answer and one column per brand works.

  5. 5

    Calculate share of answers

    For each brand: answers that name it, divided by all answers. This is the number to compare. Also note average position and which assistant favours whom.

  6. 6

    Repeat monthly

    Same questions, same assistants. Answers vary from run to run, so trust the trend over several months, not a single round.

A worked example

From our case study: three questions about monitoring brands in AI search, asked to ChatGPT, Gemini and Claude on 28 October 2026. Nine answers in total.

Share of answers naming each brand

  • Semrush7 of 9 · 78%
  • Ahrefs5 of 9 · 56%
  • Google Search Console4 of 9 · 44%
  • Brandwatch3 of 9 · 33%
  • Otterly1 of 9 · 11%

Otterly is a specialist AI visibility tracker, yet a general SEO suite was named seven times as often. That gap, not the 11% on its own, is what a benchmark shows you.

What to do with the result

  • Find where the leaders are mentioned. Search for the top-named competitor plus your category and note the review sites, lists and articles that come up. Those are the sources to get onto.
  • Compare assistants separately. If one assistant names you and another never does, the gap tells you which sources each relies on.
  • Look at the questions you lose. Being absent from “best for small businesses” but present for “cheapest” tells you how assistants see your positioning.
  • Fix how you describe yourself. Your homepage should say plainly what you do, for whom and where. Assistants can only repeat what is clear.

For the full playbook, read how to get your business recommended by ChatGPT.

When to use a tool instead

Doing this by hand every month is fine for a handful of questions. If you want daily data, more questions, or share of voice charted over time, a tracker does it automatically. Make sure the plan includes competitor tracking. Prices change, so confirm on each site.

Some are affiliate links. Compare all tools.

Questions

What is AI share of voice?

The share of AI answers to your buyers’ questions that mention your brand, compared with your competitors. If ChatGPT, Gemini and Claude name you in 6 of 30 answers and your main competitor in 15, your share of answers is 20% against their 50%.

How many questions do I need?

Ten questions across three assistants, 30 answers, is enough to see who dominates. Fewer than that and a single unusual answer can swing the result. Paid trackers typically run dozens or hundreds of prompts daily.

Is there a platform that benchmarks AI visibility against competitors?

Yes. Otterly.AI and Peec AI both list competitor tracking and share of voice on their pricing pages, and Promptwatch says it shows your visibility compared with any competitor you add. Check the plan you choose includes competitor tracking before you buy.

Why do competitors appear in AI answers more than I do?

Usually because they’re mentioned more often on the pages assistants learn from: review sites, comparison articles, directories and forums. Our case study found long-established brands named far more often than newer specialists, even for questions about the newer category.

Start with your first round, free

Our check asks ChatGPT, Gemini and Claude three buyer questions and lists every competitor they name. About a minute, no sign-up.

Run the free check