What platforms measure whether AI cites your content?

The six, side by side

Every row below was read from the vendor's own pages on 27 September 2026, except Peec's prices, explained underneath.

Peec's own pricing page lists the same 50, 150 and 350 prompt tiers but did not render its prices for us, so the euro figures come from a third-party review updated on 26 September 2026. A February review quoting €89 for 25 prompts was out of date and we did not use it. Otterly and Peec both discount annual billing by roughly 15 percent.

What each one actually counts

"AI visibility" is not one number, and these tools do not all measure the same thing.

• A mention is your brand named in the answer text.
• A citation is one of your pages linked as a source.

They come apart all the time. An answer can quote your page without naming you, or name you while citing a directory. Semrush's own knowledge base describes its toolkit as measuring how often a brand is mentioned compared with the median of its competitors. Ahrefs describes Brand Radar as reporting mentions and citations separately. Signal AI reports mentions alongside which publishers and articles the engines relied on. We wrote about why the difference matters, and it matters here because two tools can give the same brand very different scores while both being right.

Engine coverage matters more than prompt count

A thousand prompts on the wrong engines tells you less than twenty on the right ones.

Google's AI Overviews and AI Mode are different surfaces that often cite different sources for the same question. Among the engines each vendor lists as standard, only Ahrefs and Profound's enterprise tier include both. Peec can, if you spend two of your three engine slots on them, and Otterly sells additional engines as add-ons. Claude appears as standard only in Profound's enterprise list. Semrush watches two engines, and one of them is AI Mode rather than AI Overviews. So before comparing price per prompt, write down which engines your buyers actually use, and discard any tool that cannot see them.

Per prompt, the self-serve plans land in a narrow band: about $1.22 to $1.93 at Otterly and about €1.21 to €1.70 at Peec, cheaper at the larger tiers. Price is rarely the deciding factor. We worked through the cost against revenue separately.

Who each one suits

Our judgement, from the documentation. We have not run the six against each other.

When a manual audit beats a dashboard

Dashboards count. They rarely tell you what your citation was used for, and that can be the whole story.

On 22 September 2026, one of our pages was cited in a Google AI Overview. Any tool counting citations would have scored that as a win. Reading the answer line by line showed something else: we were cited once, for a caveat, while another site was cited five times, for every step of the method the answer recommended. A buyer reading that answer took their procedure from someone else. No dashboard we looked at classifies citations by the job they do in the answer.

A manual audit is the better buy when:

1. You need to read the answer text, not just a count. Accuracy, tone and role all live in the text.
2. You are checking fewer than about twenty questions, where two or three hours a month does the job for nothing.
3. You have not yet shown there is a gap worth tracking. Our method is published, and one careful run tells you whether a subscription would measure anything.

A dashboard is the better buy when you need daily data on many questions, across several engines, for longer than anyone's patience lasts. Most teams eventually need both: the tool for the trend and a person for the reading.

Something to know before you buy

A tool tracking Google's AI surfaces has to run your prompt as a real Google search to see the answer. That appears to leave a trace. In our own Search Console, long persona questions have been turning up daily at one impression each, never clicked, including one that existed before we had a page for it. We wrote up how to spot them.

The practical point for a buyer: if your prompts include anything you would not want a competitor to read, a product name that has not launched, a client's name, a sensitive issue, assume the sites ranking for that prompt may be able to see it in their own reports. This is our inference from our own data, not something any vendor has confirmed.

Limits

• Prices and plans change often. Everything here describes 27 September 2026, and Peec's prices come from a third-party review rather than Peec's page.
• We compared documentation, not performance. We have not run the same prompts through all six to see whether they agree, and we would expect them not to, because each samples differently.
• What a vendor says it measures and what it actually counts can differ. Ask for a sample report on your own prompts before paying.
• This is not a complete market list. Other tools exist, and we included these six because they are widely used, publish enough to compare, or answer a distinct need.