Mendix was the low-code platform ChatGPT and Gemini named most often, showing up in 5 of the 6 answers we logged (83%). We ran this on 2026-08-06: six buyer questions about low-code platforms, put to ChatGPT and Gemini, one live web search per question, with every product named and every source domain recorded.
Two names tied just behind it. Appian and OutSystems each appeared in 4 of 6 answers (67%). Below them sat a cluster of four platforms at 3 of 6 (50%) each: Quick Base, Retool, WaveMaker and Zoho Creator. That is the whole group named in at least half the answers, seven products out of the 24 distinct names the two assistants produced.
The other 17 names were thin. Fourteen of the 24 products showed up in exactly one of the six answers, and only three more (m-Power, Microsoft Power Apps and Oracle APEX) reached two. So most of the "low-code platform" names ChatGPT and Gemini returned were one-off mentions, not a stable consensus. The consensus is small: Mendix, then Appian and OutSystems, then a four-way tie.
The shortlist changed with the question
This is the part that matters most, because it decides which answer you are trying to win. The six questions did not return one ranking six times. They returned different lists depending on the exact words.
The "best free low-code platforms" answer shared almost nothing with the rest. It named Budibase, Appsmith, ToolJet, Retool, UI Bakery, TeleportHQ, Windmill.dev and Five.co. Seven of those eight, everything except Retool, appeared in no other answer in the entire run. Mendix, our overall leader at 5 of 6, was absent from the free question completely.
"Most affordable low-code platforms" went the other way. Instead of open-source tools it brought the enterprise names back: Quick Base, Zoho Creator, Mendix, OutSystems, Appian and Oracle APEX, plus Kissflow and a bare "Power Apps" listing. Five of those eight also show up under the plain "best" question. So ChatGPT and Gemini treat "affordable" as "discounted enterprise platform" and "free" as "open-source project," two separate shortlists from two words a buyer might think mean the same thing.
The bare "best low-code platforms" question returned eight enterprise-leaning names: Mendix, OutSystems, Microsoft Power Apps, Appian, Oracle APEX, WaveMaker, Pega and Quick Base. Pega appeared here and nowhere else across the six answers.
Plenty of names lived in a single question and then vanished. "Small teams" was the only place Base44 and Google App Maker came up. "What low-code platforms should I use" was the only answer to name Salesforce App Cloud. "Recommendations" was the only one to say Salesforce under that plain label. "Most affordable" was the only home for Kissflow and for the bare "Power Apps" label. And the seven free-tool names above, Budibase, Appsmith, ToolJet, UI Bakery, TeleportHQ, Windmill.dev and Five.co, each surfaced only in the "best free" answer. That is how fan-out works: one phrasing pulls one source page, and the source page decides the names.
Where the answers came from
Two domains fed every single answer. reddit.com and weweb.io each appeared in all 6 of 6 answers. Reddit is community discussion; weweb.io is a vendor's roundup blog. Behind them, thectoclub.com fed 5 of 6, and gartner.com and pcmag.com fed 4 of 6 each.
Community threads carried real weight. Reddit sat behind all 6 answers and Quora behind 1, so a discussion-forum link was under every answer the two assistants gave. The rest of the top of the list is roundup and listicle content: weweb.io (6 of 6), thectoclub.com (5 of 6), pcmag.com (4 of 6), integrate.io (3 of 6), zapier.com (3 of 6), appbuilder.dev (2 of 6) and superblocks.com (2 of 6).
Review directories barely registered. G2 appeared in 2 of 6 answers and Capterra did not appear at all. The analyst angle beat the directory angle: Gartner fed 4 of 6, twice G2's reach. If you assumed G2 and Capterra rankings drive AI answers in this category, this data says otherwise, one showed up twice and the other never.
Here is the insight we care about most. A vendor's own domain fed an answer, and that vendor got named. appsmith.com sat in the sources for the "best free" question, and Appsmith was named in that same answer. quickbase.com sat in the source pool for the "what should I use" question, and Quick Base was named in 3 of 6 answers across the run. Publishing on your own domain in a form these assistants pick up is not a vanity exercise, it is one of the real paths into the answer.
The reverse is just as telling. weweb.io fed all 6 answers and superblocks.com fed 2, yet neither WeWeb nor Superblocks was ever named as a product. Being the page an assistant reads is not the same as being the product it recommends. Those vendors won the citation and lost the mention, because their content ranked the other names.
What gets a low-code vendor named
If you sell a low-code platform and want to be in these answers, the data points at one thing: be present in the exact pages ChatGPT and Gemini read. The recurring sources across the six answers were reddit.com (6 of 6), weweb.io (6 of 6), thectoclub.com (5 of 6), gartner.com (4 of 6) and pcmag.com (4 of 6). Those five domains, not your market share, put the names on the list.
Market size clearly did not decide it. Microsoft's Power Apps, one of the largest products in the category, was named in only 2 of 6 answers as "Microsoft Power Apps," plus 1 more under a bare "Power Apps" label. Salesforce landed once as "Salesforce" and once as "Salesforce App Cloud." Meanwhile Mendix, at 5 of 6, was simply mentioned in more of the pages the assistants pulled. Presence in the sources beat brand size in every one of these answers.
The practical read: get cited in the community threads, since Reddit sat behind all 6 answers; get into the roundups that rank platforms, like weweb.io, thectoclub.com and pcmag.com; and publish your own material in a form these tools quote, the way appsmith.com made it into the "best free" answer. Then pick which question you want to win, because "best free" and "most affordable" pulled almost entirely separate lists.
The full list, counted
Every product named in at least 2 of the 6 answers, with its share of the six. Names appear exactly as the assistants gave them.
| Product | Named in | Share |
|---|---|---|
| Mendix | 5 of 6 | 83% |
| Appian | 4 of 6 | 67% |
| OutSystems | 4 of 6 | 67% |
| Quick Base | 3 of 6 | 50% |
| Retool | 3 of 6 | 50% |
| WaveMaker | 3 of 6 | 50% |
| Zoho Creator | 3 of 6 | 50% |
| m-Power | 2 of 6 | 33% |
| Microsoft Power Apps | 2 of 6 | 33% |
| Oracle APEX | 2 of 6 | 33% |
One caution on the counts: the assistants split two brands across labels, and we are not summing them. Microsoft's product appears as "Microsoft Power Apps" (2 of 6) and again as plain "Power Apps" (1 of 6). Salesforce appears as "Salesforce" (1 of 6) and as "Salesforce App Cloud" (1 of 6). We keep each label on its own line because that is what the measurement recorded. Merging them would invent a number the data does not show.
We are not on this list, and that is correct
Glotier does not sell a low-code platform, so we are absent from all 6 answers, exactly as we should be. We did not appear, and we would not expect to. But this run is the precise thing we do for a customer's own category: ask ChatGPT and Gemini the questions your buyers actually type, record every product named and every source behind each answer, and show you why the named names got named. If you want to see where you stand in your category, the check is free, needs no account, and takes about a minute.