How ChatGPT Recommends Lever

Across 100 applicant tracking software questions, Lever appeared in 39 answers (rank 3 among all brands named), was recommended in 38 (rank 3) and was ChatGPT's own pick in 28 (rank 4). That is a pick rate of 72% of its appearances, close to the 12-brand median of 76.5% for brands named 10 or more times in this teardown.

When ChatGPT picked Lever, it named it first among the brands it picked in 3 of those 28 answers. On the 20 direct advice questions ("which should I use"), ChatGPT picked it in 9 and recommended it in 11.

These are the labels ChatGPT attached to Lever in its applicant tracking software answers, taken from "best for" table columns and scenario lists and quoted as written. We found 24 such labels for Lever in this category; the most frequent are below.

• "Recruiting + talent CRM" (2 answers)
• "Recruiting-heavy tech startup"
• "Growing companies"
• "Sourcing/passive candidates and CRM are particularly important"
• "High-volume + sourcing/CRM"
• "Remote startup / scale-up"

If this is not how you describe your product, that gap is the first thing to fix: AI answers repeat the positioning they find most often across the sources they read.

In 7 head-to-head applicant tracking software questions that named Lever against at least one competitor, ChatGPT picked both Lever and at least one named rival in 7, only Lever in 0, and only a rival in 0. In every one of them ChatGPT made both Lever and a rival its picks instead of declaring a single winner.

To see what ChatGPT suggests instead of Lever, we asked 4 questions about alternatives to it ("lever alternatives", "cheaper alternative to lever", "open source applicant tracking system alternative to lever", "lever alternative for startups").

The brands ChatGPT picked most often in those answers, with how many of the same answers recommended each one:

• Ashby: picked in 2 of 4, recommended in 2
• Breezy HR: picked in 2 of 4, recommended in 2
• Greenhouse: picked in 2 of 4, recommended in 2
• Workable: picked in 2 of 4, recommended in 2
• JazzHR: picked in 1 of 4, recommended in 2
• Teamtailor: picked in 1 of 4, recommended in 2
• Gem: picked in 1 of 4, recommended in 1
• OpenCATS: picked in 1 of 4, recommended in 1
• SmartRecruiters: picked in 1 of 4, recommended in 1
• iCIMS: picked in 1 of 4, recommended in 1

The 28 applicant tracking software answers in which ChatGPT picked Lever carried 79 source citations in total. The most-cited domains in those answers:

• lever.co (vendor site): 20
• greenhouse.com (vendor site): 8
• workable.com (vendor site): 6
• ashbyhq.com (vendor site): 4
• hireduce.cloud (other third party): 3
• yardstick.team (other third party): 3

Across all 38 answers that recommended Lever, picked or not, there were 105 citations.

Lever's own site (lever.co) was cited 26 times across all 100 answers in the category, in 26 different answers.

Google check: for 9 of the 28 questions where ChatGPT picked Lever, lever.co also ranked in Google's top ten organic results for the same question. So most of these picks came on questions where its own site did not rank.

These are the brands ChatGPT named alongside Lever most often in applicant tracking software answers, and in how many of those answers ChatGPT picked them. This is the competitive set ChatGPT has built for it, which is not always the one a vendor would choose.

• Greenhouse: in 34 of Lever's 39 answers, picked in 29
• Workable: in 29 of Lever's 39 answers, picked in 24
• Ashby: in 28 of Lever's 39 answers, picked in 23
• SmartRecruiters: in 9 of Lever's 39 answers, picked in 6
• BambooHR: in 8 of Lever's 39 answers, picked in 6
• Breezy HR: in 8 of Lever's 39 answers, picked in 6

Lever is on ChatGPT's applicant tracking software shortlist but is not one of its top three picks: it was recommended in 38 answers and picked in 28, against 47 picks for Workable, the most-picked brand. Its weakest showing is on questions asking for alternatives to other brands, where ChatGPT picked it in 4 of 16.

lever.co was the most-cited source in the answers that picked Lever, so its own pages are doing real work in shaping how it is described.

In 19 of the 28 answers that picked Lever, lever.co did not rank in Google's top ten for the same question, so Google rankings alone do not explain where ChatGPT sends buyers.

The ChatGPT data on Lever comes from Teardown 09: applicant tracking software. Each teardown runs 100 buyer questions (category, head-to-head, alternatives and advice phrasing) through ChatGPT with web search on, logged out, United States, English, one run per question.

Every answer was coded under codebook v2.0 by two independent AI readers working blind from the published reader protocol. Each reader coded every brand mention as picked, recommended, listed, passing or anchor; where the two disagreed, a third AI reader settled the code. Across all answers coded under v2.0, the two readers agreed on 97.8% of brand codes.

"Picked" means the answer made the brand its own verdict: "my pick", "choose X if", a shortlist it tells the buyer to act on, the #1 of its ranking, or the direct answer to a "which should I use" question. "Recommended" means the answer assigned the brand to a stated case or fit ("best for small teams") or picked it, so every picked brand also counts as recommended. "Appeared" counts answers where the brand was recommended or listed. Every count is a count of answers. Use-case labels are quoted from the answers as written. Google checks use the top ten organic results for the same question collected the same day. Corrected October 23, 2026: Google answered some automated searches in late September with swapped results pages full of the wrong sites, so where the page collected on September 26 was swapped, the check now uses Google's page for the same question on October 23, 2026.

Limits: one run per question, one country, one point in time. AI answers vary between runs, so treat these as frequencies for this sample, not fixed positions. High Salience has no commercial relationship with any brand on this page. If you work for this brand and see an error, write to hello@highsalience.com and we will check it against the dataset.

Full per-answer datasets (CSV) are on each teardown page.

These FAQs are the quick version of how Lever fares in ChatGPT, using the same data as the sections above.

Does ChatGPT recommend Lever?

In our applicant tracking software teardown (web search on, United States, September 26, 2026), ChatGPT named Lever in 39 of 100 buyer questions, recommended it in 38 and made it its own pick in 28.

What does ChatGPT say Lever is best for?

The label ChatGPT used most often for Lever was "Recruiting + talent CRM". Other labels included "Recruiting-heavy tech startup", "Growing companies", and "Sourcing/passive candidates and CRM are particularly important".

Which brands does ChatGPT compare Lever with?

In applicant tracking software answers, the brands ChatGPT named most often alongside Lever were Greenhouse, Workable, Ashby, and SmartRecruiters.

To see how ChatGPT describes your brand, as shown for Lever on this page, start with the Category Salience Brief. It runs your category's buyer questions across Google and the AI surfaces, shows how each answer positions you against competitors, and ranks what to fix first.