When it comes to scouting out workforce talent, some employers these days do a lot more than just rustling through a bunch of resumes looking for the right person. Instead of doing things the old-fashioned way, smarter companies these days are resorting to collecting massive amounts of data to vet potential employees first.

This practice is justified because employers can study data about people to predict specific outcomes, for example if a prospective hire is likely to quit the job within the first 12 months, steal from their employer, file claims against them and so on. Big Data has other uses too, allowing employers to analyze their most successful employee’s characteristics, factoring in such things as prior experience and educational achievements to hone in on the most suitable qualities for any position they need to fill.

With more employers slowly discovering the benefits of a Big Data-fuelled recruitment process, it’s no surprise that specialist companies have stepped in to sell these services directly. One of the leaders in this emerging niche is Evolv Inc., which has come up with a number of surprising insights that dispel several myths about what we thought we knew about recruiting.

Criminals can be an asset to your team

It’s well known that having a criminal record generally goes against you when walking into that interview room, and indeed, many former jailbirds will be unlikely to even get that far. But companies that discriminate against those with a criminal background could actually be shooting themselves in the foot – Evolv’s data shows that workers in this category are just as likely to stick at the job, and do it just as capably, as anyone with a squeaky clean past. Moreover, there are even certain industries where ex-cons outperform their do-gooder contemporaries, such as in call center work, claims Evolv.

Always ask a prospect, “What web browser do you use?”

One of Evolv’s more curious findings relates to software – specifically the kind of internet browser that people use. Anyone that admits to using Internet Explorer should set off some alarm bells, for apparently people who use a browser different to the one originally installed on their machine are likely to be a much better worker than those who don’t. Evolv says that out of a group of 30,000 employees it studied, Chrome and Firefox users not only did better at their jobs, but were more likely to hang around too. As to why this is, Evolv isn’t quite sure, but speculates that it may indicate an ability to “reach informed decisions”.

Friendship matters

Healthy relationships among workers can have a seriously good impact on their profitability. While some employers might be concerned that having too many “friends” in the workforce might be distracting, the data suggests that in actual fact, workers with close relationships are far more productive and more likely to stay in their jobs.

“Our studies found that rather than just a pleasant happenstance, work friendships significantly affect employee attrition and productivity,” said Dr. Michael Housman, Evolv’s Director of Analytics.

“It is this simple: employees referred by their friends are less likely to quit and are more productive; employees who are trained in a generally ‘friendly’ culture stay twice as long; and if you have friends at work, you are more likely to stay at that job.”

Long-term unemployment should be ignored

Most employers are of the opinion that there’s a reason why certain people haven’t been able to hold down a job in the last few years or months – because they’re unreliable, not as a productive and unlikely to stay that long. But the data shows that this simply isn’t true. On the contrary, workers who land a job after being unemployed for six months or more are just as likely to stick at as someone that hasn’t suffered the misfortune of being out of work for so long.