When a Mentor Helps — and When It Doesn't

In 1999, psychologists Belle Ragins and John Cotton surveyed 609 professionals across three occupations and found an uncomfortable result: those who had been assigned an official mentor by their employer showed no measurable difference in salary or advancement compared to colleagues with no mentor at all. Every advantage belonged to the people who had found their mentor on their own.

"Find a mentor" appears in every career advice book, and the logic is obvious: an experienced person who knows how the system works shortens the road. That's true. But the shortcut only materialises under specific conditions — and when those conditions are absent, a mentoring relationship can consume time, raise stress, or in the worst case, hold you back. In 2004, Tammy Allen and her colleagues published a meta-analysis of 43 mentoring studies in the Journal of Applied Psychology. The effects were real — and considerably more modest than the career books promise.

The question isn't whether you have a mentor or not. It's how you ended up together.

What a mentor actually does: two functions

Kathy Kram at Boston University laid out a framework in Mentoring at Work (1985) that still holds: mentor functions split into two categories. Career functions — opening access to the right people, assigning challenging work that builds your visibility where you can't yet reach on your own, protecting you from beginner mistakes. Psychosocial functions — acceptance, honest feedback, a role model at close range.

Both matter, but they produce different outcomes. Allen and colleagues' meta-analysis shows that career functions correlate more strongly with objective measures: promotions and pay. Psychosocial functions link more closely to satisfaction and sense of meaning. Most people searching for a mentor want the first category. Many end up with only the second.

What the numbers show

When Allen, Eby, Poteet, Lentz and Lima pooled 43 studies, the picture was mixed. Mentoring works: correlation with career advancement — r=0.31; with salary — r=0.12; with job satisfaction — r=0.18. But even the strongest of these effects is small to moderate. And it doesn't materialise in every mentoring relationship.

The difference between types of mentoring relationship turned out to be larger than the difference between having a mentor and having none.

The surprising finding: formal mentoring barely works

Ragins and Cotton compared three groups: those who had found a mentor on their own (informal mentoring), those whose company had assigned one (formal mentoring), and those with no mentor at all. The first group consistently outperformed both others — in pay, advancement, and satisfaction. The formal group, by objective measures, was indistinguishable from the third: an assigned mentor delivered exactly as much as no mentor.

The explanation is straightforward: a person told by their organisation "you are now a mentor" often lacks any internal motivation for the role — they are meeting an obligation, not investing in another person. That is what kills the effect.

The practical implication: if your company runs a mentoring programme, joining it guarantees nothing. What determines the outcome is the specific person — and why they showed up.

When a mentor makes things worse

Allen and Eby, writing in Personnel Psychology (2004), studied what they called negative mentoring — and found it is not an edge case. Some mentors take credit for their mentee's work. Others grow competitive when the mentee begins to outgrow them. Others create dependency: the mentee learns to ask for answers rather than to think.

The conclusion is stark: a bad mentor is worse than no mentor. People who had experienced toxic mentoring reported lower job satisfaction and higher intention to leave than those who had worked without a mentor at all.

Four questions before you start looking

What exactly do you want? Career acceleration — access to the right people, challenging projects, visibility — or feedback and orientation in your field? Different goals lead to different people.

What is this person willing to give? Being a mentor is not a position or an obligation. If genuine interest isn't there on their side, pushing the relationship is pointless.

Are they working toward your independence? A good mentor assigns tasks that gradually make their advice unnecessary. A bad one builds a habit of returning.

Do your expectations align? Ragins and Cotton identified misaligned expectations as one of the strongest predictors of failed mentoring. A direct conversation about goals at the outset saves months.

Sources

1. Kram, K. E. (1985). Mentoring at Work: Developmental Relationships in Organizational Life. Scott Foresman.
2. Allen, T. D., Eby, L. T., Poteet, M. L., Lentz, E., & Lima, L. (2004). Career benefits associated with mentoring for protégés: A meta-analysis. Journal of Applied Psychology, 89(1), 127–136.
3. Ragins, B. R., & Cotton, J. L. (1999). Mentor functions and outcomes: A comparison of men and women in formal and informal mentoring relationships. Journal of Applied Psychology, 84(4), 529–550.
4. Eby, L. T., Butts, M., Lockwood, A., & Simon, S. A. (2004). Protégés' negative mentoring experiences: Construct development and nomological validation. Personnel Psychology, 57(2), 411–447.

This article is for educational purposes and does not constitute professional advice.