Published entry pricing across the category runs from about $20 to $295 a month, but the number that predicts your bill is prompts multiplied by engines multiplied by requery frequency. One vendor prices on exactly that formula and says so.
Every figure below was read from the vendor’s own pricing page on 17 August 2026 and is linked.
Last updated: 17 August 2026. Rechecked quarterly.
Entry prices in this category are close to meaningless on their own, because the vendors are selling different quantities under the same word. The useful exercise is not comparing the prices. It is working out what unit each price buys and converting them.
What does entry actually cost across the category?
Ten vendors publish a self-serve entry price. Here is the spread, ordered low to high.
| Vendor | Published entry price | What it includes |
|---|---|---|
| Rankscale | From $20/mo | Essentials tier, allocations not detailed on the pricing page |
| Otterly.AI | $29/mo | 15 prompts, 4 core engines |
| Mentions.so | $49/mo | 25 prompts, 3 LLMs, 1 site |
| Knowatoa | $59/mo | 3 services, prompt allowance not published |
| Rivarise | $69/mo | 50 prompts per day, 4 engines |
| Semrush AI Visibility Toolkit | $99/mo per domain, billed annually | 25 prompts, 1 domain, 4 engine families |
| Profound | $99/mo billed yearly | 50 prompts, ChatGPT only, 1 seat |
| Trakkr | £79/mo, about $100 | 50 prompts, 1 brand, all 8 models |
| AthenaHQ | $295/mo | 3,600 credits, 9 models |
| Peec AI | Not published | Credits and features listed; no price shown publicly |
That is a fifteen-fold spread at entry, and it does not correlate with coverage. Trakkr at about $100 includes all eight of its models; Profound at $99 includes one engine. The price is not telling you what you are buying.
What actually predicts the bill?
Three multiplied quantities: how many prompts you track, how many engines you track them on, and how often each combination is re-run.
This is not an analogy. Peec’s own credits documentation states the formula outright: “1 prompt x 1 model x 1 day = 1 credit” and “1 prompt x 1 model x 1 month (30 days) = 30 credits.” A vendor has written the cost driver down for you.
The only quantity most pricing pages show clearly. Ranges from 15 to 400 at published tiers.
Sometimes bundled, sometimes tiered, sometimes an add-on. The quantity most likely to be hidden.
The quantity almost nobody publishes, and the one that decides whether the number is a rate or a coin flip.
The third quantity is the one to ask about, because it changes what the data means and not just what it costs. That argument is made in full in how to interpret an AI visibility report.
How do you normalise two quotes?
Convert both to observations per month, then divide the price by it. An observation is one prompt, on one engine, once.
Worked through. Otterly Standard is $189 for 100 prompts on four core engines. If each runs daily, that is 100 times 4 times 30, so 12,000 observations, or about 1.6 cents each. Profound Growth is $399 for 100 prompts on three engines, which at daily frequency is 9,000 observations, or about 4.4 cents each. Neither vendor publishes its requery frequency, so both figures assume once a day and both would fall if the true frequency is higher.
The comparison you want is cents per observation. Almost nobody sells on it, and every quote can be converted to it.
Why is the cheapest option rarely the cheapest?
Because the low tiers are sized to prove the product works rather than to run a programme. Otterly Lite is 15 prompts and Mentions.so Starter is 25, and a set that small is dominated by which prompts you happened to pick.
The practical effect is that teams buy the entry tier, find it does not answer their question, and upgrade within a quarter. The real cost of the cheap plan is the tier above it, arrived at three months later, plus the quarter of data collected on a prompt set too small to trust.
That is not an argument for buying the expensive plan. It is an argument for sizing the prompt set before choosing the tier, because the prompt count is the constraint that binds first in almost every case.
What is not in the sticker price?
Three things, in descending order of how often they surprise people.
Engine add-ons. Otterly’s Claude add-on runs from $29 to $439 a month and its Google AI Mode and Gemini add-ons from $9 to $149 each, on top of the tier. Full coverage there cannot be quoted as a single number, which is unpacked in Otterly pricing explained.
Seats. Profound includes one seat at Starter and three at Growth, and its pricing page states that teams larger than three are better suited to a custom enterprise plan. Semrush charges from $45 a month for additional users.
Domains and brands. Semrush prices per domain. Mentions.so includes five sites at $99 and ten at $199. If you run more than one brand, this line moves the ranking more than the headline price does.
So what should you budget?
Decide the three quantities first, then shop. Most B2B teams tracking one brand land on 25 to 50 unbranded prompts and want four or more engines, and at that shape the honest range is roughly $70 to $200 a month depending on whether engines are bundled.
Below about $50 you are buying a sample small enough that prompt selection decides your results. Above about $300 you are usually buying seats, compliance or multi-brand support rather than more measurement.
The full field with per-vendor detail is in best AI visibility tracking tools, the sub-$200 band in AthenaHQ alternatives, and what a bundled suite module does differently in Semrush AI toolkit alternatives.