Skip to content

How are you being referenced by SIs?

See what ChatGPT, Google AI and Perplexity say about your brand

sirefs is an AI visibility tool. It asks ChatGPT, Google AI Overviews, AI Mode, Gemini, Perplexity, Copilot, Claude and Grok the questions your buyers ask, shows who gets mentioned and cited, and tells you what to publish to close the gap.

Free GEO audit plus three buyer questions on ChatGPT, Google AI Overviews and Perplexity. No sign-up.

The demo shows real answers about one of our own products, collected weekly.

“best AI chatbot for a webshop”

United Kingdom, English

Illustration
  • ChatGPT

    ui

    Mentioned · #2

    Named instead: Brand A, Brand B

  • AI Overviews

    serp

    Not mentioned

    Named instead: Brand A, Brand C

  • Perplexity

    ui

    Mentioned · #4

    Named instead: Brand B

  • Claude

    api

    Not mentioned

    Named instead: Brand A, Brand B, Brand D

Visibility, 28 days

n = 170

33%

95% interval: 26% to 40%

Every AI engine we track, on every paid plan.

  • ChatGPTui
  • Google AI Overviewsserp
  • Google AI Modeserp
  • Geminiui
  • Perplexityui
  • Microsoft Copilotcoming
  • Claudeapi
  • Grokapi

Each prompt is checked on every engine your plan includes, in its own market. No per-engine add-ons, no credits. Microsoft Copilot is next: it joins every plan as soon as our data vendor can collect it.

How does sirefs work?

You choose the questions your buyers ask. We ask every engine, measure who they mention and cite, and show you what to publish.

  1. 01

    Pick the prompts

    Enter your domain and we suggest the questions buyers ask AI about your category, like "best AI chatbot for a webshop", and the competitors to track. Edit them or add your own, in any of 14 markets.

  2. 02

    Every engine answers

    Each prompt goes to ChatGPT, Google AI Overviews, AI Mode, Gemini, Perplexity, Copilot, Claude and Grok, weekly or daily, the way the public apps answer a logged-out visitor.

  3. 03

    Read honest numbers

    Visibility, share of voice, rank and citation rate per engine and market, each with its sample size and a confidence interval, so you can tell a real change from noise.

  4. 04

    Know what to publish

    See the pages each engine cites for every prompt: yours, your competitors' and third parties like Reddit or review sites. That's your list of pages to write and sources to get onto.

What does an AI visibility gap look like?

When we first ran sirefs on our own products, one SaaS brand appeared in 1 of 132 answers across seven engines. Its top competitor appeared in 70% of them.

  • The engines also named four competitors nobody had thought to track.
  • The brand's site scored 91 of 100 in the GEO audit. A well-built site can still be missing from the answers.
  • The fix is content: answer pages in the format the engines quote, and a place on the sites they already cite.
answers mentioned the brand

1 of 132

answers mentioned the brand

of answers named its top competitor

70%

of answers named its top competitor

Why can you trust the numbers?

AI engines rarely give the same answer twice. A tool that reports one sample as a trend is reporting noise, so sirefs is built to show what's real.

  • Real app answers

    We collect what the consumer apps show a logged-out visitor, from our servers. Where an app can't be read logged out (Claude, Grok), we use the official API and say so.

  • The method on every number

    Every answer is labelled ui, serp or api. Answers from different methods never share a trend line.

  • Intervals, not just percentages

    Every rate comes with its sample size and a 95% confidence interval, so 2 of 6 never looks like a solid 33%.

  • Alerts on real changes

    You get an alert when a change passes a significance test, not every time a number wobbles.

  • No browser extension

    Nothing to install and nothing that stops when your laptop sleeps. Checks run on schedule, spread across the day.

  • No personal chat memory

    Answers aren't shaped by your own ChatGPT history or account, so you see what a new buyer sees.

How we collect and measure

Is your site ready for AI engines?

The GEO audit reads your site the way most AI crawlers do, without running JavaScript, and runs 26 checks in five groups. Every failed check comes with a concrete fix.

  • 8 checks

    Access

    Can AI engines reach your site?

  • 3 checks

    Readability

    Can they read it without JavaScript?

  • 5 checks

    Identity

    Do they know who you are?

  • 9 checks

    Content

    Is it easy to quote?

  • 1 checks

    Presence

    Are you known elsewhere?

Frequently asked questions

What is an AI visibility tool?

An AI visibility tool checks what AI answer engines say when people ask about your category. sirefs asks ChatGPT, Google AI Overviews, AI Mode, Gemini, Perplexity, Copilot, Claude and Grok your buyers' questions and records whether your brand is mentioned, where it ranks, which competitors appear instead and which pages are cited.

Which AI engines does sirefs track?

ChatGPT, Google AI Overviews, Google AI Mode, Gemini, Perplexity, Claude and Grok, all on every paid plan. Microsoft Copilot is next: our data vendor's Copilot collector currently hits a sign-in wall, and Copilot joins every plan as soon as it works. The free plan covers the consumer-app engines; Claude and Grok are collected through their APIs and start on Starter.

How do you collect the answers?

Our servers collect the answers the consumer apps give a logged-out visitor, through data vendors, with no personal account and no chat history. AI Overviews and AI Mode come from Google's results pages. Claude and Grok can't be read logged out, so we use their official APIs and label those answers api.

What is GEO?

GEO, generative engine optimisation, is the work of making a brand show up in AI answers: letting AI crawlers read your site, making pages easy to quote, and getting onto the sources the engines cite. The free GEO audit shows where your site stands.

Why do your numbers come with intervals?

AI engines rarely give the same answer twice, so one answer says little. sirefs samples each prompt over time and shows every rate with its sample size and a confidence interval. If visibility moves from 40% to 33% on fifteen answers, the intervals show it's noise. Alerts fire only on changes that pass a significance test.

Can sirefs guarantee my brand shows up in AI answers?

No. AI engines choose their own sources, and no tool can guarantee a mention, a citation or traffic. sirefs measures where you stand, tells you when that really changes, and shows what the engines cite, so you know what to publish.

What counts as a prompt?

A prompt is one question in one market, for example "best AI chatbot for a webshop" in the United Kingdom, in English. Each prompt is checked on every engine your plan includes, at your plan's frequency. There are no credits and no per-engine multiplier.

Is there a free plan?

Yes, and it keeps running: 5 prompts checked weekly on the consumer-app engines for one brand, with the GEO audit. New workspaces also get 14 days of Starter without a card, then fall back to Free.

Can I track other countries and languages?

Yes. Every prompt carries a market: a country and a language. You can track prompts in 14 markets, including the United States, the United Kingdom, Germany, France, Spain and the Nordics, and local competitors and sources show up in the results.

How are you being referenced by SIs?

Run the free check on your domain, or start tracking your own prompts on every engine.

14 days of Starter free. No card needed.