Does AI name you, or your competitor?
Enter your domain. We work out what you sell, write five questions a buyer would actually ask, and put them to ChatGPT and Gemini — then show you exactly who got named.
Free · no signup · takes about a minute, because we actually ask the engines
One sentence decides whether you exist in your category.
A buyer asks an assistant about your category and gets back a paragraph naming two or three companies. There is no page of ten results to be sixth on. Either you are in that sentence or you are not — and that is the only thing this tool measures.
“best ats resume checker for tech roles”
Illustrative. Rival A–D and Yourbrand are placeholders — we do not publish invented quotes about real companies. Your check uses your real domain and real answers.
We ask what buyers ask
Not your brand name. An engine asked about you directly will of course say you exist — that measures nothing. The questions are category questions: 'best X for Y', 'which tool does Z'.
We store the answer
Every score traces back to text an engine actually produced, at the moment we asked. No index, no proprietary number you have to take on trust.
We report position, not just presence
Being named third in a hedged sentence is not the same as being the recommendation, and a score that treats them alike is not telling you anything.
Being mentioned once is luck. Staying is a position.
Ask the same question eight weeks running and you get eight slightly different answers. The check above is one of those eight. It is genuinely useful and it cannot tell you whether a result is a trend or noise — nothing that runs once can.
Named in 46% of cycles, this brand is emerging — climbing, but not yet a position. A tool that only counted mentions would have called week 7 a breakthrough and week 3 a crisis. Neither was either.
of prompts keep a stable core of 1–5 names that persist week after week. Those are positions, and they are winnable.
cycles before a Core vs Carousel verdict means anything. That is why the free account grants exactly enough for one.
surfaces on every paid plan — including Google's AI Overviews and AI Mode, which no free tool can afford to query.
Written down, rather than discovered later.
The free check is a smaller measurement, not the same one with parts hidden. Here is precisely what is missing from it — and none of it is blurred out in your result, because there is no number there to hide.
Four of the six surfaces
Perplexity, Gemini, Google AI Overviews and Google AI Mode. Each costs money per query, so they belong to plans rather than to a free tool that would later have to withdraw them.
Your questions, not ours
We generate five. On a plan you set them — 40 on Starter, 120 on Growth — because you know what your buyers ask better than a model reading your homepage does.
Anything over time
A single check has no previous reading to compare against. Trends, volatility bands and alerts on real change all need history.
Core vs Carousel
Whether you hold a durable position or rotate through. It needs at least four cycles, which is exactly what the free account grants.
Competitor share of voice
The free check names who came up. Which of them beats you, on which questions, with a confidence interval, needs more than five prompts to be worth printing.
The fix loop
Blindspots turning into briefs, drafts with live checks, publishing to WordPress, and a before/after that shows what a rewrite broke as well as what it gained.
What the free check does not do is lie to you. The four surfaces we did not run show as empty rather than blurred, because we have no number for them. Filling a locked row with plausible noise would be the fastest way to make every other figure on this site worthless — and it is the specific thing this product exists to be an alternative to.
Questions about the check
Is the free check really free, and what do I get?+
Free, no account, no card. You get the complete result of a small measurement: five questions run against ChatGPT and Gemini on their cheapest models, the real score, the questions we asked, who was named instead of you, and what to do about it. Nothing is blurred — the free check is a smaller measurement, not the same one with parts hidden.
Why only two engines, and which models exactly?+
Cost, and we would rather tell you the model names than let you assume. The free check runs gpt-5.6-luna and gemini-2.5-flash-lite — the cheapest model each provider offers. Claude's cheapest is about twelve times the price of Gemini's, and Perplexity charges a per-request web-search fee that no model choice avoids, so running all four free would cost us more than twice what these two do. Paid plans run the faithful models across all six surfaces.
Why does it take a minute?+
Because we genuinely ask the engines. We read your site, work out your category, write five questions a buyer would plausibly ask, run each one against both engines, and parse the answers. There is no cached score being looked up — that is also why a repeat check on the same domain returns the stored result for a week.
Where do the five questions come from?+
From your site. We read your homepage, derive what you do and who for, and generate questions in the shape a buyer would actually type — 'best X for Y', not your brand name. Searching your own brand name tells you nothing: of course an engine knows you exist if it is asked about you directly.
Does the free check use the same models as a paid plan?+
No, and that is worth knowing before you compare the two numbers. Free runs each provider's cheapest model; paid runs the faithful tier — the models closest to what a person actually gets in the app. A smaller model has different training data and different retrieval behaviour, so it genuinely names different brands. Expect your score to move when you upgrade, in either direction, without anything on your site having changed.
Is one check enough to know where I stand?+
No, and that is the honest answer rather than a sales line. Ask an AI the same question next week and you may get a different answer — that variance is the whole reason this category exists. One reading is a snapshot; telling normal drift from a real change takes several cycles, which is what the tracking does.
Will this make AI mention my brand?+
We measure what the engines say and tell you what to fix. Nobody controls the output of a model they do not own, and anybody promising you otherwise is selling something they cannot deliver. What you can influence is whether your pages are readable, quotable and attributable — and whether what exists about you elsewhere supports being named.
Do you store my domain or my results?+
We store the result so a repeat check within a week returns it instantly rather than costing another run. No account, no email, nothing that identifies you. Those stored result pages are not indexed — publishing a score about someone else's domain without their consent is not something we do.
My score is low. Is my site broken?+
Not necessarily, and it is worth separating the two. A low score can mean your pages are hard for a model to read — that is what the free page checker tests — or it can mean your pages are fine and there is simply not enough about you elsewhere for an engine to name you confidently. Roughly a third of what gets cited in most categories is a forum thread you do not control.
One reading, or the whole picture.
The free account adds 20 checks — enough for one complete Core vs Carousel verdict. No card, and nothing starts charging you when they run out.
Or make sure AI can read your pages first — free page checker.