Untangling the Social Web

socialwebThe 2014 HRO Today Forum iTalent Competition winner aggregates real-time data to predict job-seeking behaviors.

By Audrey Roth

Seventy-one percent of the labor force, including those both employed and unemployed, is on the job market, according to the 2014 Jobvite Job Seeker Nation Study. The question of how to retain top talent has persisted, and is even more pressing with nearly three-quarters of the workforce open to alternative offers. Machine learning platform and 2014 iTalent Competition winner, Joberate, is taking the question head on by providing insight into job- seeking behaviors of the current workforce.

Joberate’s CEO Michael Beygelman explains that the initial form of Joberate was essentially conceived at #truHelsinki, the recruiting unconference. The original concept focused on recruitment by tracking activity on social media profiles in an attempt to be able to understand when employees were ready to leave their positions.

“There’s really nothing out there that gives companies insight into what their workforce is thinking. And there’s no way for them to know when they should intervene, when they should not intervene, and most of it’s purely reactive,” he says.

That’s where Joberate—in its form today—comes in. “We saw this unique opportunity to actually create a technology platform that would give HR leaders and corporate socially responsible companies an opportunity to better understand the job-seeking behaviors of their workforce. And then use that information to begin to do things like intervention, workforce planning, true retention, and employee engagement programs,” Beygelman says.

It was in first month of 2014 when the focus of the technology began to fully form in the retention direction. Says Beygelman, “We realized that an unanswered problem in the marketplace today is around workforce analytics: What your people are doing, how likely are they to leave, and really helping companies with data around their employee workforce and how they can retain them.”

But just how can Joberate help organizations recognize these patterns? Social media channels. ”You kind of had this advent of something called the social web, and then the transformation of society to adapt to the social web,” Beygelman says. “And what’s happening today is that most of the information about people actually sits on the social web in myriad of little places.”

Information from these new sources can deliver actionable insight about an employee’s volatility. The technology aggregates and validates this data, which is occurring constantly.

“All the data that we have is 24/7 live, fresh data, in data feeds that we get from various data service providers,” says Beygelman. The approximately 6,000 data points that Joberate gets per individual can be as simple as knowing when someone changes their status on various websites
to meetings that they have, and much more. As this data
is continuously fed to Joberate, their technology takes the reigns. “We built a machine learning platform that actually takes that data and measures what people do, and then scores their job-seeking behaviors based on their digital footprint,” he says.

This translates into a J-Score. The J-Score is a number between 4 and 70 that represents an employee’s job- seeking behavior, quantified as a result of the data about the individual. The lower the J-Score, the lower the chance for change.

A Joberate dashboard shows the number of employees at risk for leaving based on real-time data.

A Joberate dashboard shows the number of employees at risk for leaving based on real-time data.

The technology also produces the J-Index, which is an up- to-date score for organizations based on the average of their employees’ J-Scores. This provides organizations with an inside look into their employees’ voluntary attrition, satisfaction, and engagement. Joberate even provides a public J-Index of Fortune 100 companies, updated in Joberate’s system constantly and posted on their website publicly once a day.

“If you look at the top of the list, those companies have been in the top 10 for a while. They have the lowest volatility—meaning that the least amount of people in those companies are actively looking for work outside the company,” explains Beygelman. “That’s at the macro-level. You could then also do this in micro-level, so you can see every individual.”

Employees with similar J-Scores can then be grouped and examined to find out the issue causing the high score, and forecast attrition rates can be made. “What’s really cool is once you have the data, you can then do workforce planning. You can look at over the next 12 months: How many people am I going to lose? Where am I going to lose them? If I need to do a staffing plan, how many new people do I have to hire?” By gathering this data and insight, companies can make action items of how to intervene successfully.

Another way companies can leverage the technology is by benchmarking themselves against other companies with anonymity. This can be based on industry or even against other industries. “It gives HR insights about what they’re doing, how does it compare to what others are doing, who do we have that’s at risk, how can we intervene for those people that are at risk,” says Beygelman.

The key is: The data provides actionable insight. It could be as simple as discovering that employees from the same town have high J-Scores caused by new construction that’s causing major headaches and stress during their commute. For this group, HR could temporarily allow some work from home hours or flex commuting. Problem solved, J-Scores drop.

“Within 60 days, the target attrition rate went down by half,” says Beygelman. “(This is) a real-world scenario of something that’s (not) working that had nothing to do with a bad manger or being a bad company, something as fundamental and simple as these people didn’t want to spend five hours a day commuting.”

Predictive analytics is a hot market right now, so what sets Joberate apart from the rest? It comes down to the type of data. Much of the technology out there leverages existing static data that sits in disparate company systems and posts it to dashboards, says Beygelman. Joberate works with real-time, constantly changing data to create actionable insights. Since the information is up-to-date rather than from last year’s employee survey, organizations have the opportunity to tackle current workforce challenges before they develop into major issues.

What’s next for Joberate? Unfortunately, its analytics haven’t predicted anything specific yet, but its potential will continue to be bountiful. “I think that there is some moral good that’s going to come of this. Our vision, our guiding light is to help society better understand job- seeking behaviors of people,” says Beygelman. “And to be honest, we also want to make the labor and the employment market more transparent.”

Between the insight provided by the J-Index and the iTalent Competition win under its belt, Joberate is well on its way to these changes being made.

Posted October 14, 2014 in Enabling Technology

Leave a Reply