Minuttia alternatives, measured

Between 22:30 on 26 July and 01:15 on 27 July 2026 we put ten buyer questions about AI search visibility agencies to Claude, ChatGPT, Perplexity and Google AI Overviews, five times over. Forty answers per run, 200 in total. Six of the ten questions named no agency, which leaves 120 answers where the engine had to choose without a prompt.

An alternatives-to page is normally written from the totals, because the totals are flattering to everybody. The name-free column is the one that decides whether a firm enters a deal it was not already invited to.

Minuttia sits third of five on totals and third of five on the name-free count, on one naming out of 120.

Fourteen of Minuttia's fifteen namings came from questions that had already put an agency name in front of the engine. Those questions took the shape of alternatives-to and one firm versus another, and they reward an engine for repeating the name it was handed. One naming in 120 open questions is the number that describes reach. The other fourteen describe a question set, not a market position.

The same firm looks like a contender on one column and a near-absence on the other.

This is not a criticism of Minuttia. It is the ordinary shape of this category. Four of the five firms we measured were named three times or fewer across 120 open questions, and two were named zero times. One firm took 15 of the 19 total namings. That concentration is the finding.

The 200 answers returned 663 citations across 223 distinct domains, 129 of them cited exactly once. minuttia.com was cited 24 times, fourth among agency-owned domains, behind beomniscient.com at 41, hamstergarage.com at 39 and rampiq.agency at 25.

So the engines are reading Minuttia. They opened minuttia.com two dozen times across 200 answers, and then named the firm once when nobody prompted them to. Those two facts sit together and they are the most useful thing on this page for anyone comparing vendors.

A citation means an engine found your page relevant enough to read while composing an answer. A naming means it decided you belonged in the answer. Improving the first does not automatically produce the second, and a dashboard that counts only citations will report progress that your pipeline never sees.

If you are shortlisting alternatives, the honest summary of what the engines returned is short. One firm dominates the open questions. Three firms appear rarely. One appears not at all unless prompted. Every one of the eight most-cited domains is owned by the agency it describes, so there is little third-party validation underneath any of these names.

We hold ourselves to the same standard rather than exempting ourselves from it. Broadcastwell scored 0 of 40 when we ran our four-engine method on our own category, and the run is published at our own visibility page. The broader category work is at GEO agencies, measured.