← All articles

Measuring8 October 2026 · 5 min read

ChatGPT, Gemini and Perplexity don't agree. That's useful.

One buying question, three different answers. What the differences tell you about your website.

Oleksii· Co-founder

In short

  • ChatGPT, Gemini and Perplexity often name different businesses for the same question. That's normal.
  • Each engine searches in its own way and trusts different sources. The gaps show which source you're missing.
  • Measure all three. Being named by one is luck, being named by all three is a position.

One question, three answers

Ask "Which accountant in Utrecht is good for freelancers?" on three engines on the same day and you'll often get three different lists. One names you first. One names you third. One doesn't name you at all.

It's tempting to see that as noise. We think it's the most useful signal you have, because it tells you where the gap is.

Why they differ

Each engine looks things up in its own way before it answers. They use different search indexes and partners, they read different pages, and they weigh sources differently. Gemini and Google's AI Overviews lean on Google's own index. Perplexity searches widely and shows its sources very openly. ChatGPT searches the web too, with its own mix of sources.

On top of that, every engine also has what it learned during training, which can be months old. So an engine that doesn't search for a question answers from older knowledge.

What the differences tell you

EngineYou named?Main sources cited
ChatGPT#1Your own site, a listing site
Gemini#3Google reviews, two competitor sites
PerplexityNot namedA comparison article, a forum thread

Illustrative example, not real data

Read the table from right to left. ChatGPT finds your own site, so your page works. Gemini puts more weight on reviews, where you have fewer than your competitors. Perplexity quotes a comparison article you're not in. That's three different fixes, and only one of them is on your own website.

How to use it

  1. 1Measure the same questions on all three engines, every day, so one odd answer doesn't fool you.
  2. 2For each question where one engine leaves you out, look at the sources it cited instead.
  3. 3Ask what those sources have that your site doesn't: a price, a specialty, a place, a review count.
  4. 4Fix the one you control first. Usually that's a page on your own site.

Questions we get

Which engine matters most?

The one your customers use. For most small businesses in the Netherlands that's ChatGPT first, with Google's AI answers close behind because they show up in normal Google results. Perplexity is smaller but cites sources openly, which makes it a good teacher.

Why does my result change from day to day?

Engines don't give the exact same answer twice. That's why one check means little and a daily measurement over weeks means a lot.

Should I also track Claude or Copilot?

If your customers use them, yes. We start with the three engines most buyers use and add others when a plan needs them.

Oleksii

Written by

Oleksii

Co-founder of Mentiofy. Builds the engine that asks your buying questions every day and turns the answers into clear insights.

DannyOleksii

Want to see what AI says about you?

Book 20 minutes with Danny or Oleksii. We run your prompts live and show you who AI names instead of you, and where. Or start right away, 7 days free.