how to predict

employee turnover.

Almost nothing is as frustrating — or costly — for companies as having a recently hired staff member leave despite a lengthy recruiting and screening process. Maybe the salary wasn't competitive enough or it wasn't a good cultural fit. Whatever the individual's reasons for leaving were, one thing is always true when employees quit: you've got to scramble to backfill them, and that's the last thing you've got time for. But what if there was a way to know ahead of time that a key member of your team is looking for an exit, allowing you to intervene?

As it turns out, there is. It sounds like science fiction, but it's very, very real.

Take the guesswork out of compensation. Get our 2018 salary guide today.

how people analytics can predict your next vacancy.

By combining the latest in artificial intelligence (AI), machine learning and predictive analytics, machines can predict when employees are likely considering opportunities elsewhere with some degree of accuracy. Within the industry, these data points are collectively called "people analytics." The way this works is ultimately not so different from the way Netflix is able to recommend a new show for you to binge on. One people analytics platform called Workday examines factors like employee activity, recent promotions, regional factors and changes within a given industry to deliver insights on who's likely to quit next.

Of course, the keyword above is the right data points, which will be a stumbling block for many companies. After all, how do you know what factors to include, what to leave out and how much weight to give each data point?

Plus, the predictive model that's perfect for one company is unlikely to be the right fit for the next, since so many factors — including the thousands of hard-to-quantify components that constitute company culture — go into determining retention. Even within a single organization, the extent to which each data point actually influences turnover is liable to change over time, so the predictive model will need to be dynamically updated and the weight given to each data point continually rejiggered.

As Bloomberg points out, these solutions work best at tech companies like Google and Microsoft that already collect substantial amounts of employee data. But today's technologies develop and evolve at breakneck speeds, so it's only a matter of time before these apps can be successful within other industries. In the meantime, there's a simpler approach to people analytics, or the inputs that you can use to predict employee behaviors — and it doesn't necessarily require tech wizardry.

it doesn't take a machine to spot the signs of turnover — and do something about it.

If you want to avoid the risk of making a bad hire by not having to hire at all, here are three easy ways to identify at-risk employees and intervene before it's too late.

1. check in with employees — all the time.

Constantly ask questions. How was your morning? How did you feel after Sam said that in the meeting? What are you excited about? What are you working on right now? Besides getting a valuable feel for the pulse of the business, you'll be surprised by how easy it is to spot sour notes and identify those who feel frustrated or undervalued. Studies show that 56 percent of not engaged employees and 73 percent of actively disengaged employees are hunting for jobs elsewhere. Regularly polling employees is an easy way to measure engagement and identify fluctuations over time.

2. be extra attentive to your hardest workers.

Employees who put in the longest hours are probably among your most valuable and best-compensated employees. But those long hours can also increase the risk of turnover. In fact, at the extreme end of the spectrum, studies show that when people work insanely long hours — say, 290 hours in a single month — turnover is nearly 100 percent. With employees who stay at the office late and come in early, check in often and try to provide perks and praise whenever possible.

3. keep an eye on the social profiles of your most valuable employees.

This one is a bit sneakier, but when you consider the costs to your organization of losing top talent, the ends justify the means. Recently refurbished professional profiles are definitely a red flag. Followed by sick days, it could be a sign that the employee is actively taking interviews. Of course, you shouldn't jump to any conclusions. Talk to senior leadership, develop a game plan and consider whether a preemptive raise is in order. Surveys show that 81 percent of employees would consider leaving their current role for a better offer, so money may be an important lever — and one that you still have control over. By following your intuition — and these three simple tips — you should be able to stay one step ahead, strategically intervene and avoid employee churn, even without AI. Remember the mighty predictive power of the pre-cogs in the movie "Minority Report?" After all, that was nothing more than the human brain.

Of course, some turnover will always catch you by surprise. When — not if — that happens and you need to backfill fast, it can pay off to work with a partner who knows how to find quality talent quickly.