Buyer question
Can you see how many people ask AI a question?
Not from the engine we use. The Perplexity API returns an answer and the sources behind it, with no count of how often anyone asks that question. Three of the eight AEO trackers we checked on 19 October 2026 list a prompt volume figure of their own on their pages, and search summaries credit two more. Bungad does not show one.
Published 20 October 2026 · From eight vendors' public pages, read 19 October 2026 · 4 minute read
In search, "volume" means roughly how many times a month people search a phrase. Buyers ask for the same number for AI questions, so they can track the questions that matter most. Some trackers sell it under the name prompt volume. The engine we query does not supply such a number, so a tracker that shows one has worked it out some other way.
Who sells prompt volume, vendor by vendor
Swipe sideways to see every column
| Vendor | Prompt volume | What the page says | Where we read it |
|---|---|---|---|
| Profound | Listed | Prompt volumes, among its features | Profound's own pricing page |
| Peec AI | Listed | Prompt volume, among its features | peec.ai/pricing |
| Ahrefs Brand Radar | Listed | Volume for each AI platform. A search summary also describes "AI adjusted volume" and 30M prompts; unverified | ahrefs.com/brand-radar |
| Semrush | Unverified | A search summary describes a prompt research database of 317M+ queries; we did not confirm it on the page | semrush.com/prices |
| Scrunch | Unverified | A search summary of its Q3 2026 post names a Keyword Volume Estimator; we did not open that page | scrunch.com/blog/ai-search-q3-2026 |
| Otterly.AI | Not seen | Not among the features we noted on its pricing page | otterly.ai/pricing |
| AthenaHQ | Not seen | Not among the features we noted on its pricing page | athenahq.ai/pricing |
| HubSpot AEO | Not seen | Not among the features we noted on its product page | hubspot.com/products/aeo |
| Bungad | Not built | No volume figure on any plan | Our own code and plans |
Read this as what each vendor says, not what we tested. Every row except ours comes from that vendor's public page, read on 19 October 2026 without signing in, or from a search summary where the row says so. We have used none of these products. "Not seen" means we did not find it on the page we read, not that the product lacks it.
Why to treat every figure as the vendor's own estimate
None of the vendor pages we read said how its volume figure is made. A search summary gives Scrunch's tool the name Keyword Volume Estimator; we have not opened that page. Until a vendor tells you its method, treat the number as its estimate, not as a count from the engine, and ask three things:
- Where does the number come from? A panel of users, search data converted into AI questions, or something else. The answer should be specific.
- Which engine and which country? A figure for one engine in one market says little about another.
- How close is a rewording? Buyers phrase the same question many ways. Ask whether the figure is for your exact wording or for a group of similar ones.
How to choose questions without a volume figure
You do not need a volume number to pick good questions. Start from what buyers already ask you: sales calls, support tickets, the searches that bring people to your site. Write each one the way a buyer would type it to an assistant. Then keep the list fixed from month to month, so the results can be compared.
Our own self-audit works this way: eight fixed buyer questions about our category, chosen with no volume figure. In all three stored runs of our self-audit (19 September and 1 October 2026) Perplexity named us in none of the eight. The full method is in which questions to track, and the result is on our own score.
When Bungad is the wrong choice
- You need a volume figure to rank your questions. We do not have one. Buy a tracker from the table that lists it, and ask the three questions above.
- You want to find new questions, not track known ones. A prompt research database does that job; we only measure the questions you give us.
Sources
- Vendor pages, each read on 19 October 2026: Profound's own pricing page, peec.ai/pricing, ahrefs.com/brand-radar, semrush.com/prices, otterly.ai/pricing, athenahq.ai/pricing and hubspot.com/products/aeo. Scrunch's Q3 2026 post through a search summary only.
- The research note behind the table:
docs/research/rival-refresh-2026-10-19.mdin our repository, which marks each line as read first-hand or taken from a search summary. - Our self-audit: the three stored runs in
web/data/self-audit/, written up on our own score.
Common questions
Is there search volume for AI prompts?
Not from the engine we use. The Perplexity API returns an answer and the sources behind it, with no count of how often anyone asks that question. Of the eight AEO trackers we checked on 19 October 2026, Profound, Peec AI and Ahrefs Brand Radar list a prompt volume figure of their own, and search summaries credit Semrush and Scrunch with one. None of the vendor pages we read said how the figure is made.
Does Bungad show prompt volume?
No. Bungad does not show any volume figure for a question. It is not built. We measure whether Perplexity names you for the questions you choose, and we leave the choice of questions to you.
How do I choose questions to track without volume data?
Start from what buyers already ask you: sales calls, support tickets, the searches that bring people to your site. Write each one the way a buyer would type it to an assistant, and keep the list fixed from month to month so the results can be compared.
See whether Perplexity names you, free
Enter your domain. We ask Perplexity three of your buyers' questions and show you whether you were named and which domains the engine cited.
3 questions on Perplexity, free. A domain checked in the last week is served from that stored run rather than asked again, and the daily free allowance resets at midnight UTC.
More guides