Three products tied at the top of voice of customer software: CustomerGauge, Medallia and SurveyMonkey were each named in 4 of the 6 buyer answers we measured. No tool was named in all six, and none reached five, so a 4 of 6 finish is as dominant as this category gets. On 4 August 2026 we put six buyer-style questions to ChatGPT and Gemini, let each run one live web search, and logged every product named and every source domain behind the answers.
The list has a long tail. Across the six answers the assistants named 22 distinct product labels, and 14 of those were named just once. Only three labels reached 4 of 6, two more reached 3 of 6 (Qualtrics and Zonka Feedback), and everything else landed at 2 or 1. If you sell here, the realistic goal is to be one of the eight names that were named in more than one answer, not to expect a clean sweep that no product achieved.
Two brands were recorded under two names each, which matters when you read the counts. The assistants named Qualtrics as "Qualtrics" in 3 of 6 answers and again as "Qualtrics XM" in 1 of 6, and named Verint as "Verint Voice of the Customer" in 2 of 6 and as plain "Verint" in 1 of 6. We keep those separate and never add them together, because the run recorded four distinct labels, not two combined totals.
How the shortlist changed across the six questions
The single most useful finding is that the shortlist reshuffled hard depending on how the buyer phrased the question. "Best voice of customer software" returned an enterprise-leaning set of 8 names: Qualtrics, Medallia, Verint, Forsta, Zonka Feedback, CustomerGauge, Crescendo.ai and Zenloop, with Zenloop appearing in this question alone (1 of 6).
Add one word and the category almost disappears. "Best free voice of customer software" returned only 2 names in the entire answer: SurveyMonkey and Crescendo.ai. Only 2 of the 22 products named across this run survived the word "free," and the enterprise leaders Medallia, CustomerGauge and Qualtrics were not among them.
"Most affordable voice of customer software" produced its own cohort of 8: Medallia, Qualtrics, Zendesk, SogoCX, CustomerGauge, Refiner, Zonka Feedback and SurveyMonkey. Three of those, Zendesk, SogoCX and Refiner, appeared in the affordable question and nowhere else (each 1 of 6). Asking for cheap surfaces cheaper tools that the bare "best" query never mentions.
"Best voice of customer software for small teams" surfaced a fourth, different cohort of 8: Zonka Feedback, Qualtrics XM, Medallia, CustomerGauge, Survicate, InMoment, Chattermill and SurveySparrow. Five of those, Qualtrics XM, Survicate, InMoment, Chattermill and SurveySparrow, appeared only in the small-teams question (each 1 of 6). "Voice of customer software recommendations" added yet another four names unique to it: Gainsight, Thematic, Alchemer and AskNicely (each 1 of 6).
SurveyMonkey is the clearest illustration of why phrasing matters. It tied for the top spot at 4 of 6, yet it never appeared for the bare "best" query or the "small teams" query at all. It showed up only for "what should I use," "free," "most affordable" and "recommendations." SurveyMonkey is the accessibility pick, not the "best" pick, and a share number on its own would hide that. The steady core is narrow: CustomerGauge and Medallia each appeared in 4 of 6 questions, and only the open "what should I use" question drew the reliable names together at once (Qualtrics, CustomerGauge, Medallia, Verint Voice of the Customer and SurveyMonkey).
The source map: who fed the answers
The answers were built from vendor blogs and roundups far more than from review directories. The most-cited domain was getthematic.com, Thematic's own site, which fed 5 of the 6 answers. Next was customergauge.com at 4 of 6, then a cluster at 3 of 6: enterpret.com, gartner.com, qualaroo.com, qualtrics.com, usersnap.com and zonkafeedback.com. Most of those are software vendors publishing "best tools" list posts, not neutral directories.
Community threads showed up, but narrowly and in a telling place. reddit.com fed 2 of 6 answers and quora.com fed 1 of 6, and every one of those citations landed on the "free" and "most affordable" questions. When a buyer asks what is cheap or free, the assistants reach for Reddit and Quora; when a buyer asks what is "best," they reach for vendor roundups and analysts instead.
The classic review directories barely featured. Neither G2 nor Capterra fed a single one of the 6 answers. The analyst-style sources that did appear were Gartner, cited in 3 of 6 answers, and SoftwareReviews, cited in 2 of 6. If your category playbook assumes G2 and Capterra reviews drive AI recommendations here, this run says they did not.
The key insight is in the overlap between the source list and the named list. CustomerGauge is the one vendor that won both games: its own domain fed 4 of 6 answers and its product was named in 4 of 6. Qualtrics did the same at 3 and 3, and Zonka Feedback at 3 and 3. Owning citable pages and getting named moved together for them. The counter-example is just as instructive: getthematic.com was the number-one source at 5 of 6, but Thematic the product was named only 1 of 6. Being the page the assistant reads is not the same as being the product it recommends, and only CustomerGauge, Qualtrics and Zonka Feedback managed both strongly.
What a vendor in this category would do to get named
The source list, not market share, decided who was named. The assistants recommended whoever the 23 domains that fed these answers recommended, so the practical task is to be present and quotable inside those exact pages. That means the vendor roundups that fed the most answers (getthematic.com at 5 of 6, plus enterpret.com, qualaroo.com and usersnap.com at 3 of 6 each), the two analyst sources that appeared (Gartner at 3 of 6 and SoftwareReviews at 2 of 6), and the Reddit and Quora threads that surfaced specifically on the free and affordable questions.
Own-domain content clearly counts. CustomerGauge's own site fed 4 of 6 answers and Qualtrics' own site fed 3 of 6, so a well-structured comparison or "best VoC tools" page on your own domain is something these assistants will read and cite. And because the shortlist changed with phrasing, coverage has to match the phrasing you care about: if you want to appear for "free" or "for small teams," you need pages that actually address those, since the 2 names that survived the free query and the 5 that were unique to small teams were almost entirely different from the "best" list.
The full list, counted
The eight products named in more than one answer, plus the two alternate brand labels the assistants used. Twelve further products were each named in 1 of 6 answers.
| Product | Named in | Share |
|---|---|---|
| CustomerGauge | 4 of 6 | 67% |
| Medallia | 4 of 6 | 67% |
| SurveyMonkey | 4 of 6 | 67% |
| Qualtrics | 3 of 6 | 50% |
| Zonka Feedback | 3 of 6 | 50% |
| Crescendo.ai | 2 of 6 | 33% |
| Forsta | 2 of 6 | 33% |
| Verint Voice of the Customer | 2 of 6 | 33% |
| Qualtrics XM | 1 of 6 | 17% |
| Verint | 1 of 6 | 17% |
Qualtrics XM and Verint are the second labels the assistants used for Qualtrics and Verint. We list them on their own rows at their measured counts of 1 of 6 each, and we do not fold them into the Qualtrics (3 of 6) or Verint Voice of the Customer (2 of 6) totals above.
Where Glotier fits
Glotier does not sell voice of customer software, so we were correctly absent from all 6 of these answers. We would not expect an AI assistant to name us in this category, and it did not, which is exactly right.
But this page is the thing we do. We take a category, put its real buyer questions, the same six we used here, to ChatGPT and Gemini, and record who gets named and which pages fed each answer. If you sell in a category like this, the same run tells you whether the assistants name you today, who they name instead across all 6 questions, and the exact source pages you would need to be on. The check is free, no account, and takes about a minute.