How difficult is it to find that employee who enhances the performance of others in the organization?  Employers are turning to organization network analysis to discover who these employees are.  These employees may not be the publicized or recognized leaders, but just those who make other employees do better. 

“Invisible stars in many cases are more powerful than the people we celebrate,” says Boris Groysberg, a professor at Harvard Business School. “If you’re going to lose that person, seven other people’s performance declines.”

Organizational network analysis is designed to identify an organization’s informal networks.  What are companies measuring?  This includes email traffic by amount and response time and how others view individual contributions.  For example, Workday developed a feedback tool that allows employees to thank their peers for help on a project or for assisting them through a thorny problem.  By using the tool, Workday can identify knowledge “brokers”—those go-to employees in the organization who others turn to for guidance or insight—and to sense how employees of acquired companies assimilate, says Greg Pryor, a senior vice president at Workday.

Equifax asks employees how helpful and trustworthy employees they work with directly are.   And more seem to be catching the fever.  WeWork posted a job on its careers page for a manager of machine learning for its people analytics team, noting “some of the projects you’ll be tackling include organizational network analysis” and an effort to predict “employee churn.” Hershey posted a role for a director of workforce analytics to, in part, conduct organizational network analysis.

“I am encouraged by more data-driven decisions making it into the workplace, because I think hiring managers are kind of notoriously bad at analyzing performance and potential,” says Laurie Siegel, a former senior human-resources executive at Tyco International Ltd. and Honeywell International Inc. “If we use these tools not blindly but to supplement human decision-making and learn from it and question some of our assumptions, I think there is promise there.”

IBM has developed internal analytics to identify employee churn.  IBM states that its “predictive attrition program,” developed through its artificial intelligence (AI) platform, Watson, can predict whether an employee is likely to leave within the next six months at a 95% accuracy rate.

But to what extent do employees have privacy?  It’s one thing to identify informal leaders in the organization, but another to predict any action they may take.  95% accuracy is not 100% accuracy. Executives at McKesson found they had higher turnover in some of their teams than others.  They worked with a company to examine data on the sender, recipient, and timing of over 130 million emails—not the content of the messages—from more than 20,000 U.S. employees to see what dots it could connect about relationships. What they found is likely not surprising.  Teams with higher turnover had stronger relationships outside the company and weaker relationships with colleagues at their level or lower inside the firm.

Companies are also looking at and analyzing employee conversations.   These companies believe that tonal analysis can help diagnose culture issues on a team, showing who dominates conversations, who demurs, and who resists efforts to engage in emotional discussions.  

At some point employees will recognize the intrusions and clap back.  Where employers are trying to make the organization more efficient could leave the organization in chaos.  “There’s what’s legally right and what you need to do to maintain trusting relationships with your employees, and they are not always the same thing,” says Stacia Garr, co-founder of workforce research and advisory firm RedThread Research, which researches and advises companies on human resource-related issues.

Source: Wall Street Journal 7/19/19, Linkedin 6/17/19