job hopping

A new tool tells companies when they’re about to lose their best people

July 31, 2014
July 31, 2014

Forget the traditional job hunt. “People are now increasingly being found for jobs, versus having to find their next job,” said Jon Bischke, CEO and co-founder of recruiting platform Entelo, which focuses on tech careers and highly skilled workers.

To serve this end, the company just launched a tool that uses predictive analytics and an algorithm to identify which workers have the right skills as well as “a higher propensity to leave” their current job.

The company made headlines in April when it announced a new paid search tool that would identify job candidates who are women or people of color by pulling data from individuals’ online profiles. At the time, the company said it was 95% accurate—and would help its 130 corporate clients make more diverse hires.

The new tool, which has been in the works since the company was established in 2011, “narrows the lens, narrows the filter” for recruiters and companies seeking in-demand engineers or experienced sales staff among others, Bischke told Quartz. The application is used in tandem with Entelo’s search engine and database of more than 25 million workers (mostly in the US) in highly skilled jobs such as data scientists and high-level sales people that are difficult to land.

Candidates flagged by the tool are six times more likely to leave their job within 120 days than their peers, according to the company. The algorithm is based on more than 70 variables, partly inspired by what recruiters look for when approaching potential candidates, but also include layoff notices or changes at the candidate’s current employer that might lead to turnover. It also measures how long employees have been at their current job, and changes in their social media profiles. Entelo uses a web crawler that searches public posts, comments, and tweets, and collects them into data files for professionals. Workers tend to become more active on social media and in professional communities just before they start a job search, Bischke noted. When testing the tool, about one-third of the people Entelo predicted would leave their jobs actually did.

But it might not matter.

“To a good recruiter there is no difference between passive and active candidates,” said Jeff Zinser, principal of Right Recruiting, who has worked in hiring for 25 years. “It all depends on the job you put in front of them… I have seen out of work people have no interest in a specific job and happily employed people very interested in the same job.”

Zinser is skeptical that Entelo can make a formula to predict reliably all candidates’ perspectives. There are “too many variables to break down into an algorithm,” he told Quartz.

The tool could potentially be used by employers to determine if current staff is getting restless and eager to change employers. But Bischke concedes that people that the tool surfaces may be exhibiting characteristics that fit the criteria but really have no intention to leave.

The best way to find out is still probably to talk over lunch or drinks—and so far no one has created an app for that.

Follow Vickie on Twitter @WorkingKind. We welcome your comments at ideas@qz.com.

Top News

Powered by WordPress.com VIP
Follow

Get every new post delivered to your Inbox.

Join 23,780 other followers