Saleshandy was the only cold email tool named in all 6 of 6 buyer answers we measured, the single product that survived every way we phrased the question to ChatGPT, Gemini and Perplexity. We asked six versions of the question a real buyer types, from the bare "best cold email software" through "best free" and "most affordable," and pooled every product the three engines named on a live web search. Saleshandy hit 6 of 6. Nothing else did.
Three tools tied just behind at 5 of 6 answers, 83% each: GMass, Lemlist and Woodpecker. Each missed exactly one framing. Woodpecker was named in every version except "best free," and Lemlist in every version except "most affordable." So the top four are near locks, but the wording of the question already decides which of them falls away.
Below that the field thins quickly. Instantly and Smartlead were each named in 4 of 6 answers (67%), and Apollo in 3 of 6 (50%). Then Snov.io, Mails AI and PlusVibe at 2 of 6 each (33%), and six more tools, Reply.io, Apollo.io, Klenty, Mailforge, Mailshake and Frostbite, named in a single answer apiece (17%). Seventeen distinct product names showed up across six questions, and more than a third of them appeared for one phrasing and then disappeared.
One honest wrinkle before the shortlist: the engines spelled two vendors two ways. Instantly appeared as "Instantly" in 4 answers and "Instantly.ai" in 2; Apollo as "Apollo" in 3 and "Apollo.io" in 1. We report them exactly as the models returned them, unmerged, because the words an assistant uses are inherited from its sources, not tidied up afterward.
How the question changed the shortlist
The bare "best cold email software" query returned 8 names, led by Lemlist, Apollo and Saleshandy, and it was the only one of the six questions where Reply.io appeared at all. Ask the plain question and you get the consensus incumbents, nothing more.
Add one word, "free," and the list re-sorts. "Best free cold email software" also returned 8 names but pushed GMass to the front, a Gmail-native tool, and pulled in two names that barely register elsewhere: PlusVibe (2 of 6 overall) and Frostbite, which appeared in this one answer and nowhere else. Woodpecker, named in 5 of the six framings, dropped out of "free" entirely.
Ask "most affordable cold email software" and the shortlist actually shrinks to 6 names, the shortest of the six questions; the other five each returned 8. Lemlist falls out completely here, the single framing this 5-of-6 tool missed. Mailshake shows up for its only time, and Snov.io and Instantly.ai climb into a list they are absent from when you just say "best." Same category, one adjective, a different answer.
The remaining two framings each surfaced their own one-hit names. "Best cold email software for small teams" was the only answer to include Apollo.io, and it is where the .ai spellings and Mails AI cluster. "What cold email software should I use" was the only answer naming Mailforge and Klenty. Add it up and six tools, Reply.io, Apollo.io, Klenty, Mailforge, Frostbite and Mailshake, owe their entire presence to one specific phrasing.
The takeaway for anyone selling here: there is no single AI shortlist to get onto. There are at least six, and they overlap only at the very top. Saleshandy is the one name in all six; everyone else is in or out depending on whether the buyer typed "free," "affordable," "small teams," or nothing at all.
Where the answers came from
Three domains fed all 6 of 6 answers: emailtooltester.com (a review and roundup site), reddit.com (community threads), and saleshandy.com (a competitor's own website). The most-cited source set in this category is one review site, one forum, and one vendor blog, in equal measure.
Community was in every answer. Reddit was cited in 6 of 6 answers and Quora in 3 of 6, so a community thread fed every single answer we measured. The roundup side was led by emailtooltester.com at 6 of 6 and Zapier's roundup at 4 of 6, with leadsmonky.com in 2 of 6.
Here is the finding a vendor should not miss. saleshandy.com, Saleshandy's own domain, was cited in all 6 answers, and Saleshandy was the product named in all 6. The company whose own content the models pulled most often is also the company they recommended most often. The pattern repeats down the list: Snov.io's domain fed 3 answers and Snov.io was named in 2; GMass's domain fed 2; and Frostbite's own site, frostbite.ai, was cited in the exact "free" answer where Frostbite got its single mention. When your domain is in the citation set, your name tends to be in the answer.
Now the absence. Not one of the 6 answers was fed by G2 or Capterra. The review directories that dominate "software" search, and that vendors pour budget into, did not appear in a single answer for this category. The models were reading Reddit, Quora, two or three roundups, and vendors' own comparison pages instead.
What a cold email vendor would do about it
If you sell in this category, the source list, not your market share, is what decides whether an AI names you. The three domains that fed every answer, emailtooltester.com, reddit.com and saleshandy.com, are the map. Being on those exact pages is worth more here than brand size that lives somewhere the models never read.
Three moves fall straight out of the data. Get into the roundups the engines actually pull (emailtooltester.com at 6 of 6, Zapier at 4 of 6). Be discussed on Reddit and Quora, which between them fed all 6 answers. And publish your own comparison content, because a vendor's own domain fed all 6 answers here and earned the top spot. It does cut both ways: mailforge.ai's domain fed 3 answers but Mailforge itself was named in only 1, so its comparison posts mostly handed citations to rivals.
Frostbite is the cleanest proof that presence beats size. It is a tool most buyers have never heard of, it was named in exactly 1 of 6 answers, and that one answer, "best free," was fed by its own domain frostbite.ai. It did not need market share. It needed to be on the page the model read for that one question. That is a repeatable play, not luck.
The full list, counted
| Product | Named in | Share |
|---|---|---|
| Saleshandy | 6 of 6 | 100% |
| GMass | 5 of 6 | 83% |
| Lemlist | 5 of 6 | 83% |
| Woodpecker | 5 of 6 | 83% |
| Instantly | 4 of 6 | 67% |
| Smartlead | 4 of 6 | 67% |
| Apollo | 3 of 6 | 50% |
| Snov.io | 2 of 6 | 33% |
| Mails AI | 2 of 6 | 33% |
| PlusVibe | 2 of 6 | 33% |
Instantly was also returned as "Instantly.ai" in 2 more answers and Apollo as "Apollo.io" in 1; we left the models' own spellings unmerged. Six tools were named once each (17%): Apollo.io, Frostbite, Klenty, Mailforge, Mailshake and Reply.io.
Glotier does not sell cold email software, so we are not named in any of these 6 answers, and that is exactly right: the models should not recommend us for a category we are not in. But measuring a category this way, every product named, every source counted, across the six ways a buyer phrases the question, is the whole of what Glotier does, pointed at your category instead of this one. If you want to see who ChatGPT, Gemini and Perplexity name when a buyer asks for the best tool like yours, and the exact source pages behind each answer, the check is free, takes about a minute, and needs no account.