More than 70% of the Fortune 500 companies offer some form of mentoring to their employees, hoping to boost performance and bolster retention, among other things. However, hard evidence of firms’ accruing those benefits has been scarce. New research finds that mentorship programs can indeed produce valuable gains—for employees and their firms—but only when they are mandatory. That’s because if mentoring is optional, the people most in need of it tend to decline the opportunity.
The two-pronged experiment involved 603 newly hired sales reps at a U.S. inbound call center. In the first part of the study, 110 randomly selected reps were assigned to a mandatory four-week mentorship program consisting of structured discussions with their mentors in which they shared their responses to standardized questions and obtained feedback. During their first two months on the job they generated 19% more daily revenue, on average, than 171 unmentored counterparts did, and the increase was durable, with more than 90% of the revenue gains sustained over six months. And the mentored reps were 14% more likely to stay with the company for at least a month—a significant gain given that sales forces suffer from notoriously high attrition among recent hires.
In the second part of the experiment, training personnel at the call center asked another group of sales reps whether they wanted a mentor. Those who said yes were randomly divided into two subgroups. People in the first were given mentors, whereas those in the second were told that none were available. The researchers found that mentoring did not produce lasting benefits. There was no difference in average productivity between the two subgroups, and although those in the first were 14% more likely than those in the second to stay with the company for at least a month, the retention difference disappeared after three months. Importantly, both subgroups significantly outperformed the reps who had opted out of the program, generating 30% more daily revenue, on average. “It seems that the people least in need of mentoring are the ones who seek it,” says Christopher Stanton, an associate professor at Harvard Business School and one of the study’s authors.
The reasons for this paradox are not clear. “It could be that low performers are insecure and shy and so are reluctant to participate,” Stanton says. “Or it could be just the opposite: They might have an inflated sense of their own abilities and believe that they don’t need help.”
How to Optimize Your Program
Companies that offer mentoring should make participation mandatory if at all possible, Stanton says. “It might feel wasteful and inefficient to set up a universal program, but mentoring is a bit like advertising,” he explains. “Half the money might be wasted—but it’s impossible to know which half.”
The researchers estimated the call center’s ROI for the mandatory mentoring conducted in the study. Accounting for administrative expenses along with the costs of taking mentors and mentees away from answering calls during their meetings, they calculated that the program generated an ROI of a whopping 870%. “The payoff was quite significant despite the inefficiency and waste,” Stanton says. By contrast, the optional mentoring program’s costs outweighed its revenue gains.
It isn’t always feasible to offer mentoring across the board, of course. The challenge for companies seeking to target limited resources is that it may be impossible to identify who needs mentoring. Stanton and his colleagues analyzed demographic characteristics and the Big 5 personality traits of the low performers in the study who chose to forgo the program, but they uncovered no statistically significant pattern. “We were very surprised,” Stanton says. “We had assumed that gender and race would play a role in who opted out, but that’s not what we found.” Reps who had gotten low scores from the recruiter during the hiring process were much more likely than others to decline to participate, but they accounted for just a small share of those who made that choice.
“We Wanted to Democratize Mentoring”
Organizations looking to deploy a limited mentoring program could analyze how new hires do in the first week or two on the job and aim their efforts at the weakest performers, Stanton suggests. Or they could run small-scale mentoring programs as an experiment, analyze participants’ performance and determine who is benefiting, and offer more-intensive mentoring to those individuals.
A couple of caveats are in order, says Jason Sandvik, an assistant professor at Tulane University and a coauthor of the study. Companies should be cautious about interpreting the results of the study too broadly. The call center experiment was limited to new hires and did not examine the effects of mentoring on more-experienced employees. And the guidance provided was highly structured and limited to sales-related activities. “Some mentoring involves informal chats about navigating office politics or discussions about the protégé’s career ambitions—but that’s not what we studied,” Sandvik notes.
In addition, managers in organizations that rank employees relative to one another should take care not to disincentivize mentors, whose compensation could suffer if mentees improved beyond a certain level. The mentors in the experiment were given a small bonus for participating and told that doing so would enhance their prospects for a promotion. Just as important, Stanton says, companies should avoid making overly fine distinctions in their sales-department incentive structures. “Top salespeople might not want to share their knowledge if doing so could threaten their standing in the department and thus reduce their pay,” he explains. “But if you have a fairly coarse incentive structure—say it offers the maximum compensation to the top 20% of performers, not to the top 5%—it’s very unlikely that a mentor would drop into a lower bonus pool because of mentored employees’ improved performance.”
Although the study was conducted prepandemic at a single call center in which everyone worked on-site, Stanton believes its findings are applicable to a post-Covid hybrid or dispersed workforce. “With less organic mingling, it becomes even more important to make sure that new employees are interacting with top performers,” he points out. Finally, he emphasizes that new hires would be smart to seek out mentorship even if their company doesn’t require it. “After all, the people who didn’t think they needed to sign up for the call center’s program are the ones who fell behind,” he says. “You never know how much you might benefit.”
About the research: “Treatment and Selection Effects of Formal Workplace Mentorship Programs,” by Jason Sandvik, Richard Saouma, Nathan Seegert, and Christopher T. Stanton (NBER working paper)