So much depends upon managers. For example, a Gallup study found that at least 70% of the variance in employee engagement scores is driven by who the boss is. This is disconcerting because the same research found that about 70% of people in management roles are not well equipped for the job. This state of affairs is hurting not just employee engagement and quality of life, but also corporate performance.
Most companies understand the importance of having highly effective managers, but few invest heavily in training to help them get there. One reason is that it’s difficult to measure and quantify what good management actually looks like. While there has been a lot of great work done to identify qualitative traits of great managers — they create trust, focus on strengths, instill accountability, avoid politics, etc. — these traits don’t provide much insight into how great managers spend their time on a day-to-day basis that differentiates them.
But there’s new data that can help. Microsoft’s Workplace Analytics product analyzes metadata from the digital breadcrumbs of a customer’s millions of de-identified email and meeting interactions to generate an objective and granular set of behavioral KPIs across the organization (for example, how much time managers spend in one-on-ones with employees, how quickly they respond to emails from each direct report, how large and diverse there networks are, etc.). Among other things, these KPIs can then be combined up with other data sets to understand what behaviors differentiate sub-populations of employees.
We recently had the opportunity to combine behavioral KPIs with employee engagement survey results for two Fortune 100 clients comprising thousands of knowledge workers. Inspired by Gallup’s findings about the influential role managers play on employee engagement, we wanted to understand what made managers of highly engaged employees different than the rest on a day-to-day basis. The results were illuminating.
Managers lead by example when it comes to working hours. Two metrics we use to provide a proxy of active working time per week are utilization and after-hours time. ”Utilization” essentially looks at the average amount of time between your first and last email or meeting of the day across several months of data and estimates total weekly working time for each employee. It’s an imperfect metric, but does provide a good directional sense of working norms. “After hours” is the amount of time spent in email or meetings outside of an employee’s normal work hours, which are typically 9 a.m. to 5 p.m.
The data shows that managers in the top quartile of utilization — a.k.a. those who work the longest hours — end up with employees who work up to 19% more hours relative to their colleagues who report to less highly utilized managers. This is perhaps unsurprising. What might be more surprising is that even though they are working more hours, the engagement scores of these employees are actually 5% higher than their lower utilization colleagues. It’s also true that employees of managers in the lowest 25% of utilization have lower than average engagement scores (2-4% lower). This suggests that people are more engaged if they work for a manager who is working at least as much as they are.
However, managers need to ensure even allocation of work. Using the same metrics as above, we found that employees who put in more hours than the rest of their team are more likely to be disengaged. More specifically, highly utilized individual contributors that work 120% longer hours than their peers are 33% more likely to be disengaged and twice as likely to view leadership unfavorably as highly utilized employees working similar hours as their team.
This intuitively makes sense in that it would be frustrating to be staring at hours of additional work on your desk while watching all of your teammates — or your boss — happily go home at 5 pm. While in some cases employees may volunteer to take on extra workloads on their own, it is a core function of a manager to allocate work across their team. This finding clearly shows that uneven allocation leads to disengaged employees.
Effective managers maintain large internal networks across their company. We measure the size of a person’s network based on the number of connections to other employees that they actively maintain. The primary algorithm we use to define a connection has both a frequency and intimacy threshold. Put more simply, in order to qualify as a connection, one must interact with another person at least twice per month in an email or meeting with five or fewer participants. This allows us to get a reasonably accurate view of the number of people an employee actually works with on a regular basis. We have consistently found that larger networks are correlated with a number of different positive business outcomes.
In this case, we found that employees who report to a manager with a relatively large internal network — in the top quartile of all managers, more specifically — have engagement scores up to 5% higher. Additionally, these employees had networks up to 85% larger than those of their colleagues reporting to managers with smaller networks.
We also found that managers with small networks can have a significantly negative impact on their teams. Employees who had networks 110% or more larger than their manager are 50% more likely to be disengaged and twice as likely to view leadership unfavorably. One interpretation of this is that employees rely upon their manager to provide a coordination role with other teams across the company and they are unable to do that effectively if they don’t have a big enough network. Employees who already have larger networks than their manager may simply see little value in the relationship and feel unnecessarily constrained by the hierarchy.
One-on-ones remain vital. We can quantify actual time managers spent in one-on-one meetings with direct reports based on calendared meeting invitations. In the companies we analyzed, the average manager spent 30 minutes every 3 weeks with each of their employees. Perhaps unsurprisingly, employees who got little to no one-on-one time with their manager were more likely to be disengaged. On the flip side, those who get twice the number of one-on-ones with their manager relative to their peers are 67% less likely to be disengaged. We also tested the hypotheses that there would be a point at which engagement goes down if a manager spends too much time with employees, but did not find such a tipping point in these datasets.