Profound pricing, the trial limits it publishes and the agency plan named only in its FAQ, compared with six priced rivals

Read on 31 October 2026, the Profound pricing page publishes no currency figure at all. Not a dollar, not a pound, not a euro, in about 234 KB of HTML. What it does publish is a free trial with hard limits, an Enterprise column marked Custom, and one sentence in the FAQ naming a self serve agency plan that the plan cards themselves do not show. So the honest answer to what Profound costs is that you cannot price it from the page, and the honest preparation is to use the two numbers it does give you.

This is a report on the document, not a complaint about it. Four of the ten vendors we read the same morning published no figure either. The difference is what each unpriced page leaves you to work with, and Profound leaves you more than most.

The two numbers Profound does publish

Fifty prompts, run daily, for seven days. That is the free trial, on ChatGPT, Gemini and Google AI Overviews, in one language and one region. The page is explicit that trial users receive a recommended prompt set based on their industry and cannot customise their prompts, so the trial measures Profound's idea of your market rather than your own buying questions. It also publishes what the trial withholds, in a grid: history none, exports none, API none, support none, single sign on no, prompt volume searches limited to 5.

Four hundred credits a month per client workspace. That figure appears in one place only, an FAQ answer about how agent pricing works, and the plan it belongs to is called Agency Growth and described as self serve. The plan cards in the served page show only Trial and Enterprise, with a For brands and For agencies toggle whose agency view is not in the HTML the server sends. If you are an agency, that one sentence is the most useful thing on the page, and you would never see it by reading the cards.

What Enterprise publishes, which is where the real detail sits

The Enterprise column is unpriced and specific, which is a better combination than priced and vague. Read on 31 October 2026 it carries up to nine answer engines, named as ChatGPT, Perplexity, Google AI Mode, Gemini, Microsoft Copilot, DeepSeek, Anthropic Claude, Google AI Overviews and Exa Search. It publishes daily frequency, custom languages and regions, ChatGPT Shopping, all time history, CSV and JSON exports, API access, SSO, SOC 2 compliance and a support option described as a dedicated specialist with a 24 hour SLA. Seats are unlimited on both plans, which the FAQ states twice.

That list is a procurement checklist you can hold a quote against. The nine engine names in particular are worth taking into the call, because a quote for five engines and a quote for nine are not the same quote and the page makes the maximum explicit.

Who published a figure the same morning

Vendor	Published on 31 October 2026	Unit stated
Evertune	Starter 60 dollars a month, Pro 800 billed annually	5,000 and 100,000 prompt credits a month
Scrunch	Core 250 dollars a month	125 prompts, 5,000 responses a month
AthenaHQ	Starter 295 dollars a month, free tier 300 credits	3,600 credits, one credit is one AI response
Goodie	Core 295 dollars a month, Pro 999	110 and 250 prompts, daily
Ahrefs	29 to 1,499 a month, AI index from 199	one prompt on one platform in one location is 1 check
AI Knows Us, ours	Rs 4,999 to Rs 74,999, in six currencies	one credit is one question on one app once, Claude costs 3
Profound	no figure	50 prompts for 7 days, 400 credits a month per client workspace

Six published prices and units you can divide into each other. Profound published neither, and published its limits instead. Both of those are information; only one of them lets you build a budget before a meeting.

How to run the demo so the hour is worth something

1. Convert the quote into a unit you already hold. Ask what one tracked prompt, on one engine, on one day costs under the proposal. Every priced vendor above can be expressed that way, so a Profound quote can be too.
2. Ask which of the nine engines are in the number. The maximum is published. The quote is not.
3. Ask who writes the prompt set after the trial. The trial set is theirs by design. On Enterprise the page says you can edit, disable or add prompts, so confirm that in writing.
4. Ask what Agency Growth costs. It is self serve according to the FAQ, which usually means a published price exists somewhere, and 400 credits a month per client workspace is the figure to size against your client count.
5. Ask what leaves the platform. Exports are none on the trial and CSV plus JSON on Enterprise, so the question is whether the plan you are quoted sits at the Enterprise end of that line.

Where Profound beats us, said plainly

Three lines, applying the criterion this page uses throughout, which is what a vendor publishes on its own live page that a buyer can verify. Profound publishes unlimited seats on both plans, where every one of our plans includes one login with full access and we publish no single sign on at any tier. It publishes up to nine engines against our five apps. And it publishes SOC 2 compliance and an SSO row, which a security review will ask for and which we cannot answer with a published page today.

Where we are the better buy on this specific question is the question itself: our price is in the served document, in six currencies, with the unit written in words, and our free scan needs no account before any of that. A buyer who wants to know the cost of being measured can know ours today. Our page on Profound alternatives for a small team or a small budget takes the replacement question further, and the review page handles what Profound publishes in place of ratings.

Questions buyers ask about Profound pricing

How much does Profound cost a month?

No figure is published in the page the server sends, read 31 October 2026. Any number you find elsewhere is somebody's quote, so treat it as a data point about one contract rather than a list price, and get your own in writing with the engine count attached.

Is the free trial enough to evaluate it?

It shows you 50 prompts a day for a week on three engines, which is a real look at the product. It will not tell you how your own buying questions perform, because the trial prompt set is chosen for you and cannot be edited.

Does Profound sell a self serve plan?

Its FAQ names Agency Growth as self serve with 400 credits a month per client workspace. The plan cards in the served page show only Trial and Enterprise, so ask about Agency Growth by name rather than waiting to be offered it.

What happens when the trial week ends?

The published route is a demo and an Enterprise package, with ongoing daily tracking. Before you take it, write down what seven days of readings told you and what you would want measured for the next twelve months, because the second number is what the contract is sized on.

Is an unpriced page a bad sign?

It is a sign about the sales model rather than the product. Enterprise software has always been priced in conversations. What it costs you is the ability to compare, which is why the sensible order is to price the six vendors that publish figures, decide what the work is worth, and then take the call.

The page tells you what to ask, if you read the grid rather than the cards

Profound hides the price and publishes the limits, which is the useful half to publish if you are only going to publish one. The nine engine names, the seven day trial shape, the exports grid and that single FAQ sentence about 400 credits a client are four things most unpriced vendors leave out entirely, and together they are enough to walk into the call with your own arithmetic. Ours is on the opposite plan: the ladder, the unit and the Claude exception are in the document, in six currencies, so the comparison happens before anybody picks up a phone. Whichever you end up buying, do the conversion into one unit first, because that is the only way two offers in this market can be read against each other.

How we know this

We read the assistants rather than guessing what they do. In our own audit of 1,260 saved answers from ChatGPT and Gemini across the United States, the United Kingdom and India, 82.0% of every citation landed on a site the business itself owned, and review directories took 0.7%. Counted as one domain per reading, 16,516 of 20,143.

That is the whole asymmetry: the pages you own are the pages an assistant quotes back to a buyer. See the dated readings and the method on our proof page.

Find out whether assistants name your business, on your own domain and your own buying questions.

By Aditya Goel, Founder, AI Knows Us. Reviewed 1 November 2026.