How it works
Why we count, and how a study runs
Asking an assistant yourself shows you one answer on one day. A study runs at least 1,000 simulated buyer conversations, reads every mention and checks the sources, then hands you the fixes in order.
Supracited AI engine
The same buyer question, asked 20 times. Each card is one recorded answer and lists which of the five most-named brands it named, first at the top.
What should I take for a headache?
What is the best over-the-counter medicine for a tension headache?
Best over-the-counter pain reliever for back pain?
Are generic painkillers as good as brand names? Which should I buy?
What is the cheapest effective headache medicine?
- All but Aleve
- All five
- Tylenol, Advil and Motrin
- Tylenol and Excedrin
- All but Excedrin
- All five
- Tylenol and Excedrin
- All but Aleve
- Tylenol, Aleve and Excedrin
- Only Tylenol
- All but Motrin
- All but Excedrin
- Tylenol, Advil and Aleve
- Only Tylenol
- All but Excedrin
- Tylenol, Advil and Motrin
- Tylenol and Advil
- Tylenol, Advil and Aleve
- Tylenol and Advil
- Tylenol, Advil and Excedrin
- None of the five
- All but Motrin
- Only Tylenol
- Only Advil
- Tylenol, Advil and Motrin
- Tylenol and Aleve
Each card you draw lands on the pile for the brands it named. The figures count only the cards drawn so far.Asked 20 times, the same question gave five different answers; the most common came back in 25% of them.
Each card you draw lands on the pile for the brands it named. The figures count only the cards drawn so far.Asked 20 times, the same question gave six different answers; the most common came back in 65% of them.
Each card you draw lands on the pile for the brands it named. The figures count only the cards drawn so far.Asked 20 times, the same question gave three different answers; the most common came back in 80% of them.
Each card you draw lands on the pile for the brands it named. The figures count only the cards drawn so far.Asked 20 times, the same question gave four different answers; the most common came back in 45% of them.
Each card you draw lands on the pile for the brands it named. The figures count only the cards drawn so far.Asked 20 times, the same question gave eight different answers; the most common came back in 50% of them.
| Brands the answer named, of the five | Answers |
|---|---|
| All but Aleve | 25% |
| All five | 20% |
| Tylenol, Advil and Motrin | 20% |
| Tylenol and Excedrin | 20% |
| All but Excedrin | 15% |
| Brands the answer named, of the five | Answers |
|---|---|
| All five | 65% |
| Tylenol and Excedrin | 10% |
| All but Aleve | 10% |
| Tylenol, Aleve and Excedrin | 5% |
| Only Tylenol | 5% |
| All but Motrin | 5% |
| Brands the answer named, of the five | Answers |
|---|---|
| All but Excedrin | 80% |
| Tylenol, Advil and Aleve | 15% |
| Only Tylenol | 5% |
| Brands the answer named, of the five | Answers |
|---|---|
| All but Excedrin | 45% |
| Tylenol, Advil and Motrin | 25% |
| Tylenol and Advil | 15% |
| Tylenol, Advil and Aleve | 15% |
| Brands the answer named, of the five | Answers |
|---|---|
| Tylenol and Advil | 50% |
| Tylenol, Advil and Excedrin | 20% |
| None of the five | 5% |
| All but Motrin | 5% |
| Only Tylenol | 5% |
| Only Advil | 5% |
| Tylenol, Advil and Motrin | 5% |
| Tylenol and Aleve | 5% |
25%
The most common answer
The same headache question, asked 20 times: five different sets of the five brands came back.
95%
Aleve, for back pain
And 10% for the cheapest pick (2 of 20), from the same assistant.
Why not just ask an assistant yourself?
-
One answer is one draw.
Aleve was named in 55% of the 100 pain and headache answers in our pain relief census, swinging by 85 percentage points across the five scenarios (the rail beside the figure above shows where). Asked “What should I take for a headache?” 20 times, the assistant gave five different combinations of these five brands. The figure at the top deals those 20 answers as cards to draw one at a time.
-
Assistants read other sites first.
None of the twelve sources cited most in the pain relief census is a maker’s or a retailer’s own site. goodrx.com was cited in 28% of its answers, mayoclinic.org in 11%. That is where a fix list starts.
-
Counting is reading.
Store brands came up 39 times in the pain relief census. We read every mention: 97% were a place to buy or one example among several, and one was a product. A keyword count would have reported 39.
Three steps, then the same count again
Measure
At least 1,000 simulated buyer conversations a study, each simulated across buyer types and across free and paid versions of the leading AI assistants. Every brand’s share is printed with the count it came from.
Find the sources
Every domain the assistants cite gets three checks: popularity rank, whether its homepage fits the topic, and signs of a recent ownership change. A domain that fails is marked flagged, in words.
Fix, in order
At the Reveal you get a signed Shelf Report and a fix list graded by evidence. The Re-check at day 90 runs the same buyer scenarios again and shows whether the number moved; the Category and Market Studies include it.
Before you ask
Do you write health claims for us?
No. We measure and source. We do not write, edit or suggest any statement about what a product does; that stays as your label states it. The fix list covers where you are missing and what stops an assistant reading your pages, never what your product says it does.
Why pay to be measured?
You are not paying for the count. You are paying for the fix list it produces: which sources AI trusts in your category you are missing from, what on your own site stops an assistant reading it, and what to do first. The count shows the list was needed and, at the Re-check, whether the work moved the number.
Is a brand in your census a recommendation?
No. A census reports which names the assistants gave, how often, and which sources they cited. The names are measurement results, not recommendations.
Which assistants do you measure?
Free and paid versions of the leading AI assistants. Supracited’s AI engine asks leading AI assistants, including ChatGPT, Gemini, Grok and Google’s AI answers, with more being added. 6,000+ data points measured by the Supracited AI engine. A study’s report names the assistant behind its figures to the client it is for; public pages read “a leading AI assistant,” or, where a comparison is shown, “Assistant A” and “Assistant B.”
Do you promise a ranking?
No promised rankings, no bought reviews, no hidden text, no incentivised posts, with no guarantee of results, stated in the statement of work.
We measure and source. We do not write, edit or suggest any statement about what a product does; that stays as your label states it. Brand names in our censuses are measurement results, not recommendations.
What a study includesStart with three buyer scenarios
We send back how often AI named each brand, which sites it cited and one source worth a closer look.