Venture

Recruitment Startups Realize, With The Help Of AI, Less Is More

Hiring is time-consuming and expensive. And from Munich to Santa Monica entrepreneurs recognize the need to help companies hire experienced talent faster.

As our discussions with investors and entrepreneurs indicate, labor market conditions are changing hiring practices and the future of work. Those on the front lines of talent acquisition technology see the wake of AI mitigating the need for companies to source qualified candidates. Others view AI as helpful but likely overstated—even software engineers appreciate talking to a human when making a job leap.

Follow Crunchbase News on Twitter

Anu Hariharan, a partner at the Y Combinator Continuity Fund, has been following the recruiting space for years. She told Crunchbase News that changes in the HR startup market have evolved alongside the need of companies, across industries, for more skilled candidates.

“Pre-2009 I would classify the recruiting market into three types of models: one is job search, which is purely listings. That was largely Monster and CareerBuilder,” Hariharan recalled.

The second two types of companies, Hariharan explained, were aggregators, like Indeed, which became popular in the early 2000s. And then there was LinkedIn, later acquired by Microsoft, which allows companies to recruit outwardly to both active and passive job seekers. The strategy? Get as many leads as possible and worry about the details later.

Ian Siegel, the CEO and co-founder of ZipRecruiter, which was founded in 2010 as a job listing site for candidates and employers, said that his company started off focusing on sourcing.

“When we started, and for really the first four to five years we were in business, it was all about volume. So everybody wanted to get a bunch of candidates as efficiently as possible,” Siegel said. However, the volume approach has its limits, noting that “sending even more candidates becomes a negative.” And recruiting startups are responding to the change.

“All of us at the top tier in the category, aspire to send… a short list of great candidates, fast,” according to Siegel. And it’s this migration from “searching to matching” that ZipRecruiter has turned its attention to, bolstered by the accessibility of machine learning technology. ZipRecruiter now uses an algorithm which incorporates 64 different criteria to match job listings submitted by companies with qualified candidates. Other startups have also incorporated machine learning and artificial intelligence techniques into their platforms.

“You have entered the dawn of robot recruiting, where computers are doing a much better job than humans ever did,” Siegel expressed. “[They’re] finding the right jobs for a job seeker to apply to and highlighting in a list of candidates—the right candidates—for an employer to pay special attention to.”

Even so, Siegel believes that humans add something to the recruiting process that cannot, at the moment at least, be mimicked by AI.

“Having somebody recruit you is intoxicating,” Siegel said. “There’s still a high value in having a human being be the one to reach out and do the engagement with that candidate to try and entice them to apply to that company’s job.”

Having someone recruit you is intoxicating

And it’s not as if algorithms are better than humans at being bias-free. Back in October, Reuters reported that Amazon ended its AI-powered hiring project because it downgraded candidates with certain language in their applications, which led to an algorithmic bias against women applicants. The consequences of unfettered AI implementation is not lost on ZipRecruiter, according to Siegel.

“We have to police what our algorithms are doing and decide what information we’re going to include and what information we’re not going to include,” he explained. Regardless, he remains very optimistic about AI in the field and its potential to remove unconscious bias in the recruiting process. Yet removing bias successfully with AI doesn’t mean a recruit will turn into a successful hire.

Sara Menke, the CEO and founder of recruiting firm Premier, believes that it is important to the survival of her company to use the best technology available. However, she also said there are aspects of the process, like ensuring a candidate will work well within a team, that require human involvement.

“The challenge with technology doing that is that it’s going to miss many of the intangibles. It is possible that at some point it won’t, and that’s great. But today, it still does,” Menke said.

Further, the interview and technical screening process that follows lead generation remains a time-consuming part of the recruiting process.

To tackle that problem, Triplebyte, which was founded by former Y Combinator partner Harj Taggar, has designed a recruiting platform and service that incorporates an adaptive coding assessment. Applicants who pass its coding test and are cleared following interview assessments with Triplebyte are pushed to the front of the line for final onsite interviews. That reduces the many hours companies spend on the interview process. The assessment platform has raised $13.1 million in total known funding, including a $10 million Series A in February 2018.

Other startups, like HackerRank, which has raised a known total of $58.2 million, are also focused on streamlining the technical recruiting interview process through coding tests as well.

The potential to save valuable time in the hiring process, particularly when it comes to interviewing, is an attractive feature for companies. As Masha Drokova, founder and general partner at Day One Ventures, told Crunchbase News in an email: “[If] you have the best candidates, you’ll have top clients.” Of course, attracting the best candidates is easier said than done. While smaller startups may find the technical talent they need with the help of AI-powered recruitment startups, companies may still not achieve the targeted result: a great hire.

“The greatest challenge for our recruiters is where we’re transparent, we communicate about the opportunity, and people say they’re really interested,” Menke explained. “We get them all the way to the altar, and then they can go dark—or all of a sudden there’s something else.”

“The needs of top candidates are only increasing,” Day One’s Drokova wrote. “There was a joke on Twitter recently that workers in Silicon Valley are sending lists of the things that they’re not going to do in the workplace prior to the interview. I’m not sure if it’s actually a joke.”

Many startups, particularly those in the Bay Area, feel that they can’t compete with some of the top salaries and benefits packages offered by deep-pocketed alternatives. However, Wes Winham, the founder and CEO of the Indianapolis-based Woven, told Crunchbase News that he believes startups have a responsibility to offer competitive rates. To attract talent, Woven also helps companies adjust their job descriptions and find developers from non-traditional educational backgrounds and regions. Recruiting in this way, he said, often shifts hiring outcomes in favor of startups.

“Github, Automatic, and Zapier—these super high growth companies—are remote, so that’s changing quickly,” Winham said. “You can hire the person in Oklahoma that loves where they’re living, you pay them what’s great for there, and it’s better for you.”

So although the labor market may be tight, it appears companies have never been more equipped to tackle the talent shortage facing them today. And amid the giants, like Linkedin and Monster, lie a number of startups who claim to be faster and smarter at finding the right candidates. What happens to these startups when the labor market slackens is anyone’s guess. For now, there’s a lot of talent in talent.

Illustration Credit: Li Anne Dias