FullStory and Hotjar were the only two tools named in all 6 of 6 AI answers we measured for heatmap and session replay tools, the single pair that showed up every time we asked.
We put six buyer questions about heatmap and session replay tools to ChatGPT, Gemini and Perplexity on 3 August 2026, one live web search per engine, and recorded every product named and every source cited. FullStory and Hotjar each landed in 6 of 6 answers (100%). No other tool matched them.
Microsoft Clarity came next, named in 5 of 6 answers (83%). One of those six answers called it simply "Clarity" rather than "Microsoft Clarity," and it is the same free tool, so in practice Clarity surfaced in every answer too, which puts it level with the two leaders once you treat the two spellings as one product.
UXCam was also named in 5 of 6 answers (83%). It missed only the bare "best heatmap and session replay tools" question and appeared in all five of the others, including both price-qualified ones.
LogRocket and Zoho PageSense tied behind them, each named in 4 of 6 answers (67%), and Contentsquare followed at 3 of 6 (50%). After those seven tools the market thins out fast: no other product was named in more than 2 of the 6 answers.
How the shortlist changed from question to question
Only FullStory and Hotjar were named under the same label in all six answers, with Microsoft Clarity joining them once its two spellings are merged. The other 15 tools moved in and out depending on how we phrased the question, which is the real story here: the shortlist an AI hands a buyer is not fixed, it rewrites itself around the qualifier in the question.
The bare "best heatmap and session replay tools" query pulled in two names that appeared nowhere else: Quantum Metric and Amplitude. Both are pricier, analytics-heavy platforms, and both showed up in exactly 1 of 6 answers, the unqualified "best" one. Add a single price word to the question and they disappeared.
"Best free heatmap and session replay tools" reshuffled the deck. Microsoft Clarity, which is genuinely free, was named first in that answer, and UXtweak entered the list. The full free set ran to 8 names: Microsoft Clarity, Hotjar, FullStory, UXtweak, Zoho PageSense, UXCam, LogRocket and Contentsquare.
"Most affordable heatmap and session replay tools" overlapped with the free set but not perfectly. Six names carried over (Microsoft Clarity, Hotjar, FullStory, UXCam, UXtweak and LogRocket), while Plerdy appeared here and in no other question, and PostHog showed up in the affordable and small-team answers. Zoho PageSense and Contentsquare, both present in the free answer, dropped out of the affordable one.
The remaining questions each had their own exclusives. "For small teams" uniquely surfaced Inspectlet and MIDA. "What should I use" uniquely surfaced Crazy Egg and Datadog. "Recommendations" uniquely surfaced Dynatrace. Counting them up, 8 of the named tools appeared in exactly one of the six questions, so nearly half the list is a single phrasing away from never being seen at all.
Where the answers came from
Two domains fed every single answer: quantummetric.com and reddit.com, each cited across all 6 of 6 questions. A vendor's roundup page and a community forum sat at exactly the same level, together forming the backbone of this category's answers.
Community threads were everywhere. Reddit fed 6 of 6 answers and Quora fed 2 of 6, so a real user discussion sat behind every answer we got. That matches what we keep finding: the engines lean on a forum thread full of practitioners as heavily as on any polished listicle.
The roundup and content sites did the rest of the work. zapier.com fed 5 of 6 answers, userpilot.com fed 4 of 6, and cpoclub.com fed 1. These are the classic "best tools" articles, and sitting inside them plainly counts.
Review directories barely registered. G2 fed 4 answers in total, but 3 of those came through its blog at learn.g2.com and only 1 through the main g2.com listing. Capterra, TrustRadius and Software Advice did not appear at all. If a vendor here has been investing in directory profiles, the engines were reading G2's articles far more than its star ratings and ignoring the other directories completely.
Here is the finding that should change how a vendor thinks about this. Several tools were named partly because their own domain fed the answer. mouseflow.com fed 5 of 6 answers and Mouseflow was named in 2. uxtweak.com fed 3 and UXtweak was named in 2. inspectlet.com fed 2 and Inspectlet was named in 1, while amplitude.com fed 3, datadoghq.com fed 1 and mida-app.io fed 1, each helping name Amplitude, Datadog and MIDA respectively. Publishing the comparison the model reads is a live path into the model's answer.
The counter-example keeps us honest. quantummetric.com fed all 6 answers, more than any other domain, yet Quantum Metric the product was named in only 1 of 6. Its content is trusted enough to be the source of every answer, but the roundups on that domain list FullStory, Hotjar and Clarity, so the page mostly promotes the leaders. Being the source is not the same as being the pick.
What we would do to get a tool named here
If we ran marketing for a tool in this category, we would work this list of 16 source domains directly, because the source list, not market share, is what decides who gets named. An engine can only recommend a tool that one of its sources mentions, so the whole job is to be in those sources.
We would start where the citations concentrate: the Reddit and Quora threads that fed 6 and 2 of the answers, the Zapier and Userpilot roundups that fed 5 and 4, and G2's blog that fed 3. Getting a tool genuinely discussed and listed on those exact pages does more for you than brand spend the model cannot see.
We would also publish our own comparison content, because vendor domains fed a large share of these answers. Mouseflow turned its own site into 5 of 6 citations. That is the gap between hoping to be mentioned and writing the page that gets read. Market share shows up nowhere in this data. Source presence shows up on every line.
The full list, counted
| Product | Named in | Share |
|---|---|---|
| FullStory | 6 of 6 | 100% |
| Hotjar | 6 of 6 | 100% |
| Microsoft Clarity | 5 of 6 | 83% |
| UXCam | 5 of 6 | 83% |
| LogRocket | 4 of 6 | 67% |
| Zoho PageSense | 4 of 6 | 67% |
| Contentsquare | 3 of 6 | 50% |
| Mouseflow | 2 of 6 | 33% |
| PostHog | 2 of 6 | 33% |
| UXtweak | 2 of 6 | 33% |
Counted strictly by the label used, Microsoft Clarity's sixth appearance sits under the shortened name "Clarity," which is why the raw data splits it into Microsoft Clarity at 5 and Clarity at 1. All told, the three engines named 18 distinct tools across the six questions, 19 if you keep those two spellings apart.
Why Glotier is not on this list
You will not find Glotier anywhere in these 6 answers, and that is correct. We do not make a heatmap or session replay tool, so no engine should name us here, and we would not want them to. We are not going to fake our way onto a list we do not belong on.
This page is the thing we actually do, though. We took one category, asked 6 real buyer questions across 3 engines, and logged all 18 tools named and all 16 source domains behind them. We can run the same measurement for your category and show you whether ChatGPT, Gemini and Perplexity name your product, who they name instead, and the exact pages feeding those answers. The check is free, takes about a minute, and needs no account.