Analytics used to predict who will leave a job!

WSJ – March 14, 2013, 3:36 PM ET

Joel Schectman

Book: HP Piloted Program to Predict Which Workers Would Quit

Hewlett Packard Co. tested a predictive scoring system that attempted to grade the likelihood that individual workers would quit the company, according to a new book.

HP piloted the scoring system in 2011 aimed at lowering attrition through a better understanding of which workers were most likely to leave, according to Predictive Analytics: The Power To Predict Who Will Click, Buy, Lie Or Die by Eric Siegel. The analytics model, Mr. Siegel says, looked at factors such as salaries, promotions and job rotations, and scored the likelihood that particular employees would leave. HP data scientists believed a companywide implementation of the system could deliver $300 million in potential savings “related to attrition replacement and productivity,” according to a November 2011 company presentation.

Data scientists made the presentation at a Predictive Analytics World conference in London, an event series founded by Mr. Siegel, a former assistant professor of computer science at Columbia University. “The scarcest resource any company has is human resources,” Mr. Siegel said. Predictive analytics offers the possibility to “preemptively intervene” in employee attrition, and “that’s the holy grail,” Mr. Siegel said.

An analysis of which factors made employees more likely to quit yielded some surprising results: “Those employees who had been promoted more times were more likely to quit, unless a more significant pay hike had gone along with the promotion,” Mr. Siegel wrote.

The “flight risk” scores, not divulged to employees, were intended to give managers a heads up if an employee was predicted likely to leave, according to Mr. Siegel, who provided CIO Journal with emails between him and HP data scientists describing the program. To try to lower a flagged worker’s probability of quitting, managers could consider raising the employee’s salary or rotating his work assignment, Mr. Siegel said.

Mr. Siegel’s says the flight risk score was assigned to HP employees worldwide. And so far, Mr. Siegel says, scores have been used to guide management decisions in at least one company team. The current extent of the program is not known, Mr. Siegel said. HP did not respond to requests for comment.

Advertisements
This entry was posted in Analytics, Human Resources. Bookmark the permalink.

11 Responses to Analytics used to predict who will leave a job!

  1. Pingback: Analytical Worlds Blog – Predictive Analytics and Text Analytics – by Eric Siegel, Ph.D. » HP Piloted Program to Predict Which Workers Would Quit

  2. ecocinar says:

    Reblogged this on Enriconomics and commented:
    HP ran a pilot program to use analytics to predict if employees might quit. Employers use exit interviews to figure out why and when employees leave for improvement (hopefully). Quantifying the sources and finding a companies risk factors before losing talent would benefit any business.

    In my own research I find that most voluntarily exits by employees occur in July though September. See the chart below for an example of data from the Bureau of Labor Statistics. Look for the highlighted spike in late summer. December on the other hand is one of the lowest periods of the year for turnover. Therefore January through May would be a great time to start analyzing the historic data and implementing changes to beat off those late Summer exits, or for whenever your high turnover period was. It is predictive analytics you can do without special software or a PHD in data science or statistics.

  3. Eric TF Bat says:

    So they leave in summer when everything is optimistic, but not in December because nobody wants to be unemployed for Christmas? It would be intriguing to see how the patter is different in the southern hemisphere then!

    • I would bet the answer is more in when bonuses pay out. Sometimes unhappy employees in the fall wait until their end of year bonus kicks in before leaving. That would explain why there is a lower number in December.

  4. john says:

    i’m wondering if how the supervisor/direct superior treats his/her employees are taken into consideration as a factor to predict who will leave the company?

  5. lucian303 says:

    Those employees who had been promoted more times were more likely to quit, unless a more significant pay hike had gone along with the promotion,”

    It’s not a promotion without a payhike; it’s a demotion.

  6. Interesting concept, I am a fan of data driven decisions or at minimum, including data analysis when making a decision. My experience has been that large companies don’t have any issue with gathering data. Understanding, at times that they don’t have a choice but analyzing and taking actionable steps to use the data is the hurdle that I have seen more often than not.

    I wonder why this data has only been used to guide decisions in a single team at HP?

  7. Bala says:

    would be interesting to see how the environment and culture factors will have a bearing on this experiment.

  8. Kate says:

    I suspect that giving preferential treatment (e.g. salary increase) to those with a high ‘flight risk’ would negatively impact the remaining employees. With this policy in place, the result may be that those who don’t get preferential treatment become disenchanted and leave instead. I’m not surprised that HP doesn’t want to comment; if this is true then the system of rewards in the company is not transparent and is not based on achievement.

  9. Avinash says:

    We at Sapience dot net are doing similar predictive analysis based on the employee engagement levels. We believe there is at least 1-3 months of gap between when an employee decides to leave in his mind and the time when he declares it. During this period his work pattern will change, he is less engaged with work, comes late and leaves early, spends more time on non-work activities and probably spending time searching for next job. This analysis compares work patterns of current engagement with previous work patterns when he was fully engaged on his job.

  10. Pingback: Big Data and HR: Everything you might want to knowAshley Gilbert's Financial Coaching Institute

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s