The Machine Readable Workforce

The ideal. What that means in an employee differs to some degree depending on the job and the organization. It has always been thought of as a philosophical construct, not an achievable reality. But that is changing.

Companies are starting to use Big Data to run statistical models that ostensibly identify the characteristics most suited to the task and the enterprise overseeing it.

Ironically, one of the first institutions to embrace this concept is Xerox, the company best known for photocopies of an original...

A benevolent interpretation of this approach suggests that by diagnosing those factors crucial to optimal performance in a specific function, the best match can be made between the task and the people available to do it. The implication is that the work will be done more efficiently, effectively and productively while, at the same time, the worker's job satisfaction will increase because the person in that role will be more successful and fulfilled.

A less sanguine perception might be that companies are attempting to weed out the individuality that makes for a vibrant workplace. That by attempting to pre-judge what success looks like in a workplace, the company is setting itself up for failure. It does so by reducing the possibility of serendipitous interaction between colleagues and the devices they use. As well, it does so by fostering a clone-like environment in which - based on the principles of evolution - the likelihood of dysfunction, under performance and coun terproductive behavior are more likely to occur.

As a civilization we support experimentation to improve institutional genetics, whether individual or organizational. The danger, of course, is that such hubris leads not to the Model Man but to Frankenstein. JL

Jessica Leber reports in MIT Tech Review:

Companies are analyzing more data to guide how they hire, recruit, and promote their employees

Xerox is screening tens of thousands of applicants for low-wage jobs in its call centers using software from a startup company called Evolv that automatically compares job seekers against a computer profile of the ideal candidate.
According to these data, culled from studying job records of many similar workers, past experience working in call centers isn’t a good predictor of success. Instead, a person should be a “creative” type, though not too inquisitive. Participating in one social network like Facebook is a plus, but involvement in too many is a negative. A short commute is a must—that means a person is less likely to quit before Xerox can recoup its cost to train them.

While personality exams aren’t new to business, large employers like Xerox are beginning to embrace a concept called “workforce science” that promises to make performance reviews and judging résumés far more data-driven. One of the best-known attempts to hire and fire by the numbers is at Google, whose HR department, called “People Operations,” has turned hiring into a kind of engineering project, using computer models to determine how many times each candidate should be interviewed, how larges raises should be, and nearly every other personnel decision.
Evolv, based in San Francisco and founded in 2007 bases its advice on data gleaned from tens of thousands of employee files on hourly workers, who also make up 60 percent of the United States workforce. Applicants have to take a half-hour online test that ranks them against a profile of a successful call-center worker. Evolv has raised $42 million from investors. Another startup, Gild, has begun using software to score computer programmers who place their work in public repositories, locating job candidates whose résumés might otherwise end up in a trash bin (see “A Startup That Scores Job Seekers, Whether They Know It or Not”).
Lawyers who practice anti-discrimination law are watching these trends. While it’s legal to give aptitude tests, hiring based on a computer’s assessment of seemingly unconnected factors—like how many social networks you join—could raise new questions. “They’re creating these big databases of people,” says Christopher Moody, an employment lawyer in Los Angeles. “More and more companies are doing pre-employment testing. Whether this really indicates some job-related quality in the applicant is questionable.”
It’s easy to see why Xerox wants to turn to automated methods. Although it still sells photocopiers, Xerox has also become one of the world’s largest outsourcing companies (see “Q&A: Ursula Burns” and “The Empire Strikes Back”). It provides services like running customer service centers, handling health claims, and processing credit-card applications that brought in $11.5 billion in revenue last year.
That business relies on a huge workforce of 54,000 customer service agents, and because of high attrition in hourly jobs (pay in the U.S. ranges from $9 to $20 an hour), Xerox will have to replace 20,000 of them this year, says Teri Morse, vice president for recruiting at Xerox Services. Morse says employees that stay less than six months cause a loss for Xerox, due to the expense of training them.
Since the company began pilot tests of Evolv’s analytics software two years ago, Morse says employees are on average staying longer at Xerox and their performance is 3 to 4 percentage points better, as measured by factors like how many complaints they resolve or how long it takes to handle a call. The software has also started to influence other subtle factors, like what time of year Xerox hires people.
Morse says basing decisions on data means Xerox has been able to broaden the base of people it will consider for hourly jobs, including those who have been unemployed for long periods. But the data also rules people out. Morse says Xerox today won’t even look at résumés of those who score in the “red” category of Evolv’s initial behavioral assessment, a 30-minute online exam that workers fill out at home. “Individuals that test strongly perform better and survive longer,” she says. Early on, while piloting the system, Morse says Xerox still hired against the advice of the data. Now, she says, “people who do poorly we no longer hire.”