Which AI visibility tools publish a price you can read: twelve pricing documents audited on 20 October 2026

Of twelve AI visibility pricing documents opened on 20 October 2026, eight carried a price figure in the page the server sent and four carried none and routed the buyer to a demo. One of the twelve carried a price in more than one currency, and that one is ours, which is why this page puts AI Knows Us first and states the criterion rather than asserting a ranking. On a different line of the same audit five rivals beat us outright: Semrush, Ahrefs, Rankscale, AthenaHQ and Otterly each print an annual figure and an annual saving in the document, and our annual price sits behind a button where no assistant and no crawler will find it. We sell one of the products on this list.

The criterion, stated so you can check it

The ranking question here is not which tool is best. It is narrower and answerable: what does each vendor publish about price that a buyer can verify today, in the document the server sends, without a sales call? Five lines, each a yes or a no:

• Is a price figure in the served page at all, or only after a script runs, or only on request.
• How many currencies are in that page.
• Is the billed unit named in words, so two quotes can be put in the same units.
• Is an annual term stated beside the monthly one, with its saving.
• Is the tax position stated, so the sticker and the invoice match.

Every row was read on 20 October 2026 with a browser user agent and the response body saved. One host answered only one of the two agents tried, which is the subject of its own section below. Nothing here comes from memory, from a vendor's marketing copy or from another comparison page.

What the twelve documents published

Figures are recorded and not reproduced, for a dated reason: on 17 September 2026 an assistant read two comparison pages whose rival prices were accurate, checked them independently, found them correct, and still set them aside as advocacy because they carried no date and no source. A price copied onto a third party page becomes a false statement about somebody else's product within a quarter. So the table records whether a figure is published, never the figure.

Supplier, read 20 October 2026	Price figure in the served page	Currencies in the page	Billed unit named in words	Annual term and saving stated
AI Knows Us (ours)	Yes, every plan	Six: rupee, dollar, pound, euro, Australian dollar, Canadian dollar	Yes	No
Semrush	Yes	One, dollar	Limits only	Yes
Ahrefs	Yes	One, dollar	Limits only	Yes
Rankscale	Yes	One, dollar	Yes	Yes
AthenaHQ	Yes, two tiers	One, dollar	Yes	Yes
Scrunch	Yes, one tier	One, dollar	Limits only	Not in the page
Depra	Yes	One, rupee	Yes	Not in the page
Peec	No	None	Yes	Enterprise tier marked annual
Profound	No	None	Trial limits only	Not in the page
Conductor	No	None	No	Not in the page
Brandlight	No	None	No	Not in the page
Otterly	Yes, three tiers	One, dollar	Yes	Yes

Three of those rows deserve more than a cell.

Peec publishes the arithmetic instead of the price. Its plan cards name the prompt allowance, the model count, the tracking frequency and the project count, and the table under them states that monthly AI answers are the tracked prompts multiplied by the active models multiplied by the tracking frequency. A buyer can act on that without a figure.

Profound publishes two routes and both are precise. A free trial with its limits spelled out, fifty prompts run daily for seven days across three named engines in one language and one region, and an enterprise tier marked custom with a demo. A demo led price is a legitimate model for enterprise software and those trial terms are specific enough to plan around.

AthenaHQ states what its price excludes. Its page says plainly that API access and extra credits are add-ons billed on top of the subscription rather than included in it, which most vendors leave to the first invoice.

Where we come first, and where four rivals beat us

We come first on the line that decides what an assistant can tell an American or a British buyer. Our pricing page carries a chapter printing every plan in rupees, dollars, pounds, euros, Australian dollars and Canadian dollars on every request, whether or not anything about the visitor is known. Counted on 20 October 2026 the served text held 31 rupee figures, 24 dollar, 8 pound, 8 euro, 8 Australian dollar and 8 Canadian dollar. The other eleven documents were single currency or carried no figure. That chapter exists because this site got it wrong first, and the counts our own pricing page serves a crawler records the version that served rupees to everybody and the day it was repaired.

We also name the unit rather than implying it. One credit is one buyer question, asked on one AI app, once, and a Claude check costs three because Claude costs more to run. Ten questions on four apps is forty credits, the count is shown before a check starts, and a failed check returns its credits. AthenaHQ names its unit the same way, one credit to one AI response, and Rankscale, Peec, Depra and Otterly each name theirs.

Now the line we lose. Semrush, Ahrefs, Rankscale, AthenaHQ and Otterly all print the annual figure beside the monthly one and state the saving in the page. We print a Monthly and a Yearly button, and the yearly figure is produced by the browser, so our served document carries no annual figure and no saving percentage at all. An assistant asked whether we offer an annual discount finds one sentence, that an annual price is shown for one month and paid once for the year, with no number attached. That is the same defect as the rupee only pricing page in a smaller place, and it is reported here rather than written around. Until it is in the document, ask us by email and we will send the annual figure with the date on it.

The row that was nearly left blank, and the request shape that filled it

The first pass of this audit recorded Otterly as not verified. Its pricing page returned HTTP 403 with a 105 character body on two readings under a Chrome on Windows user agent, and a 403 from a host is not a figure you may write from memory. Then the same page returned HTTP 200 and 283KB under a Safari on macOS user agent. Interleaved three seconds apart each way, the result was 403, then 200, then 403 again, and a fourth request with the Chrome agent plus an Accept-Language header returned 403 too. The user agent string was the whole difference, not the time and not the language header. Otterly publishes three tiers with figures, a named allowance unit, a stated annual saving, unlimited team members on every tier and an add-on price for extra prompts, which would have been recorded as a blank. Before writing down that a host refused you, change the shape of the request.

Our own managed price, and what the rival documents did not hold

A monthly retainer is a different product from a software seat. Our pricing page publishes a managed band, a GEO strategist and writer assigned to the account on top of a Growth or Scale plan, at Rs 88,499 to Rs 2,35,999 a month with a three month minimum, read 20 October 2026. The configurator beside it quotes Rs 1,24,999 a month for one market at a 32 of 100 target, with a three month minimum and a review set at twelve weeks. None of the eleven rival documents read that day carried a monthly figure for managed work. That is a statement about eleven pages read on one day, not about eleven companies.

What to do with this before you take a demo

Write your own workload down first: buyer questions, times engines, times refresh frequency, times markets. The unit count behind a dollar quote works that multiplication through. Then hand the same workload to every supplier, because two honest quotes for identical work differ by a multiple when the meters differ.

Questions buyers ask about AI visibility tool prices

Do AI visibility tools charge per domain or per prompt?

Both, and a third thing as well. Of the twelve documents read on 20 October 2026, the units named in writing were the tracked prompt or buyer question, the recorded answer, the project, the brand workspace and the domain. Otterly, Peec, AthenaHQ, Rankscale, Depra and this site each name theirs. Ask for the unit by name, then recount your own workload in it before you compare two numbers.

Do these tools offer annual discounts?

Five of the twelve state an annual term and its saving in the page: Semrush, Ahrefs, Rankscale, AthenaHQ and Otterly, all read 20 October 2026. Four publish no price figure at all, so there is no published annual rate to find. Ours is paid once for the year and the figure comes by email, because it is not in the document yet.

What should I pay for AI visibility monitoring software?

Price your own workload, not a plan. The cheapest reading is free and takes an afternoon: fifteen buyer questions, three engines, three runs each, asked in the engines' own interfaces and labelled by hand. Any subscription you add then has to earn its cost against that baseline.

Why does this page not print the competitors' prices?

Because a dated figure held for twelve suppliers goes stale faster than anyone keeps it current, and an undated figure loses the page even when it is right. Recording whether each vendor publishes, in which currency and on what unit, answers the buying question and stays true for longer.

The two cells we got wrong are the ones worth repeating

An audit that only lists wins is a brochure. This one carries a plain loss to five rivals on the annual figure and a row we almost recorded wrongly about the second most named tool in our own measurement, and those two admissions are why the rest of the table can be trusted. Ask any supplier the same five questions and watch which ones answer from their own published page rather than from a slide. Our own answers are on our pricing page, in six currencies, with the unit defined and the annual figure still missing.