Profound prices on answer engines and seats, not on prompt volume. Cost per tracked prompt actually rises between its two self-serve tiers, from $1.98 to $3.99, because the step up buys coverage rather than capacity.
Whether that is expensive depends entirely on which unit you divide by. Per prompt it gets dearer; per prompt-and-engine pair it gets cheaper.
Last updated: 17 August 2026. Read from Profound’s own pricing page on that date, and rechecked quarterly.
Profound is the best-known name in this category, and its pricing page is unusually specific about what each tier includes. That makes it easy to work out what you are actually buying, provided you divide by the right thing.
What does Profound cost?
Three tiers, two of them self-serve. Every figure below is from its published pricing page as read on 17 August 2026.
| Plan | Price | Prompts | Answer engines | Seats |
|---|---|---|---|---|
| Starter | $99/mo billed yearly | 50 unique prompts | ChatGPT only | 1 |
| Growth | $399/mo billed yearly | 100 unique prompts | 3: ChatGPT, Perplexity, Google AI Overviews | 3 |
| Enterprise | Custom | Tailored | Up to 9 | Custom |
Both self-serve tiers are quoted as billed yearly with two months free, so the monthly figure carries an annual commitment. Starter also includes 100 agent credits a month and Growth includes 400. Enterprise adds multiple companies tracked, SSO and SAML, SOC 2, API access and dedicated support.
What are you actually paying for between the tiers?
Not prompts. The jump from Starter to Growth is four times the price for twice the prompts, so on a per-prompt basis the more expensive plan is worse value.
| Unit | Starter | Growth | Direction |
|---|---|---|---|
| Cost per tracked prompt | $1.98 | $3.99 | Rises, roughly doubles |
| Cost per answer engine | $99.00 | $133.00 | Rises |
| Cost per seat | $99.00 | $133.00 | Rises |
| Cost per prompt-and-engine pair | $1.98 | $1.33 | Falls, by about a third |
The last row is the only one that improves, and it is the one that describes what the product does: 100 prompts across 3 engines is 300 tracked combinations against Starter’s 50. If you need three engines, Growth is the cheaper plan. If you need one engine and more prompts, it is a large bill for capacity you are not getting.
$1.98 becomes $3.99. On this reading the upgrade is poor value and the entry tier is the efficient one.
$1.98 becomes $1.33. On this reading the upgrade is the efficient one and Starter is the expensive way to track a single engine.
Is the entry tier usable on its own?
For one engine, yes. Starter tracks ChatGPT and nothing else, so it answers “how do we look in ChatGPT” and cannot answer “how do we look in AI search.” Those are different questions and only one of them is usually the one being asked.
That single-engine limit is the specific reason meaningful coverage begins at $399 rather than $99. It is also why a per-engine breakdown matters when you compare vendors: a tool covering four engines at $69 and a tool covering one at $99 are not competing on price, they are competing on scope, and the reasoning behind keeping them separate is in mention rate vs citation rate.
What do the agent credits buy?
Starter includes 100 a month and Growth 400, which is a separate budget from the prompt allowance. They meter the automated work the product does on top of tracking rather than the tracking itself, so a plan can run out of credits while its prompt count is untouched.
Treat them as a second capacity line when you size a plan. Four times the credits arrives with four times the price at Growth, which is the one ratio on the page that scales exactly with cost.
Where does the seat limit bite?
Sooner than most teams expect. Starter is a single seat, Growth is three, and the pricing page states plainly that “If you have more than 3 people on your team, you are better suited for a custom enterprise plan tailored to your needs.”
So a four-person marketing team is an enterprise conversation regardless of how few prompts it tracks. That is a legitimate way to price, and it is worth knowing before the evaluation rather than after, because it moves the decision from a credit card to a procurement cycle.
How does that compare to the rest of the market?
At the $99 mark specifically, the comparison is stark on scope rather than on price. Trakkr quotes £79 a month, about $100, for 50 prompts with all eight of its models included. Rankscale quotes $99 for 1,200 credits, which it states is up to 4,800 engine answers across ten named engines.
None of this makes Profound expensive in the abstract. It buys SOC 2, SSO, multi-company tracking and the strongest brand in the category, and for an enterprise buyer those are the line items that matter. It does mean the entry tier is priced for a single-engine question, and most teams arrive with a multi-engine one.
The wider field, including the vendors that publish a free plan, is in Profound alternatives and the full twelve in best AI visibility tracking tools.
One structural difference is worth naming because it survives any price change. Profound’s tiers gate engines behind price, so coverage is what you buy as you climb. Several competitors gate prompts instead and include every engine at entry, which inverts the upgrade path: you move up for volume rather than for surfaces. Neither is better in the abstract, and the right one depends on whether your constraint is how many questions you track or how many places you track them.
What should you check before signing?
Two things this pricing page does not state. First, the run count: how many times each prompt is queried per engine per day, because a prompt tracked once a day and one tracked five times produce different numbers from identical reality. Second, whether you get the raw answer text or only the score, which decides whether you can diagnose a drop or only observe one.
Both are covered in the reading discipline in how to interpret an AI visibility report, and the same two questions are worth asking of every vendor in Semrush AI toolkit alternatives, ours included.