How Zapier pioneered the conversational application and saved hundreds of recruiter hours

Key results
- 97% of candidates rated the experience as excellent
- Saved an estimated 228 hours of recruiter time without narrowing the funnel
- Ran 273 conversational applications across 75 days on senior roles, with a low opt-out rate of 6%
About Zapier
Zapier is a fully remote AI automation company whose platform connects thousands of apps, helping teams automate work without code. Its strong employer brand draws large, global candidate pools across technical and go-to-market roles. As an AI-first company, Zapier applies the same philosophy to its own hiring, using AI to gather and organize information while keeping every hiring decision with a human on its team.
The challenge
For years, the job application served one purpose: it was the filter that protected a recruiting team’s most precious resource, human time. You narrow the field on paper, then invest recruiter energy in the people most likely to fit. That logic only holds when the filter is reliable, and for Zapier it had stopped being reliable.
The early signal was failing. Zapier’s signal has always come from applications, with a resume or LinkedIn profile optional. But surging volume, AI-generated submissions and increasingly polished applications meant early-funnel decisions were being made on signal that was often inflated, fabricated or simply thin.
The best candidates looked like everyone else. Career changers, people from non-traditional backgrounds and those whose strengths show up in conversation rather than formatting were flattened into the same bullet points as everyone else, and easy to miss.
Great candidates were lost to bandwidth, not fit. Because deep evaluation required scarce recruiter time, qualified people were being screened out for lack of capacity rather than lack of ability. The constraint was the process, not the talent.
The information we actually needed, a candidate’s real experience, their motivations, how they think and communicate, only ever surfaced in a live recruiter screen. But we don’t have the capacity to screen everyone, and frankly, we shouldn’t have to.
– Tracy St.Dic, VP, Global Head of Talent, Zapier
The solution
Zapier had already been using Voice AI interviews, but in the usual order: a recruiter reviewed applications first, then invited a subset to a Voice AI interview. Then, over a 75-day trial, the team decided to flip that sequence and made the interview part of the application itself.
From screener to application. Instead of a step candidates unlock after review, the voice interview became part of applying. Zapier’s job posts tell candidates plainly that a resume or LinkedIn profile does not tell the whole story, so after submitting, they complete a 25-minute Voice AI conversation to finish applying. It also reframes the process to signal openness from the start.
Instead of ‘prove you’re worth talking to,’ the message becomes ‘we want to hear from you,’ from the very first step.
– Tracy St.Dic, VP, Global head of Talent, Zapier
A deterrent to bad actors. Because the conversation asks structured, role-specific questions in real time, it naturally discourages less serious applicants, including those using bots or AI-generated submissions, helping restore trust in the funnel.
Human judgment is kept at the center. As an AI-first company, Zapier uses AI to gather and organize information, but every application is reviewed by a Zapier recruiter and every hiring decision is made by a human. Candidates who prefer not to use Voice AI can opt out and receive the same consideration.
The results
Moving the conversation to the front of the process let Zapier reach more candidates, save significant recruiter time and win over the candidates least expected to embrace an AI interview.
Candidates are heard, and hiring teams get real signal. Applicants experience the conversation as a chance to represent themselves more fully than a document ever could. Every candidate gets the same opportunity to tell their story, and every hiring team gets signal that resumes structurally cannot provide.
What I liked about this format is that it gave me a chance to explain my background and thinking in a bit more depth, rather than feeling like I was simply submitting a resume into a black box.
– Sr. Applied AI Engineer candidate, Zapier
228 hours of recruiter time saved. Over the 75-day trial, Zapier saved an estimated 228 hours of recruiter time without narrowing the funnel. Rather than reserving recruiter hours for a pre-filtered few, the team gave every candidate a real conversation while spending less human time overall.
97% rated the experience excellent, and participation stayed high. Zapier ran the experiment on various technical roles, including senior positions like Sr. Applied AI Engineer, experienced and in-demand candidates you might expect to resist an AI interview. Instead, 273 completed the conversational application, opt-out held steady at roughly 6% and satisfaction was near universal.
Deeper conversations with senior candidates. Senior candidates spoke for an average of 31 minutes versus 22 for junior candidates, using the time to explain their thinking rather than recite their background. This is proof that this new way of hiring works, for companies and candidates alike.
