Mentorships Have Expiration Dates
Picture a mentee years into the relationship that made his career. He goes to the senior partner who taught him everything and asks for room to build his own client book. The answer is some version of this: it isn't really in my interest, and you are more useful where you are. The relationship has been good for a long time, which is exactly why the answer is so hard to argue with. The good ones rarely feel like they should end.
That moment, the ask and the soft refusal, is where a mentorship quietly stops developing one person and starts capping his career. Nothing breaks, and no one is obviously wronged. The arrangement that built a career keeps running long after it became a ceiling, and it usually takes a shock from outside the relationship to make anyone notice.
CEO Brief: Kathy Kram's foundational research on mentoring at work, published in the Academy of Management Journal in 1985, identifies four phases every mentor relationship moves through: initiation, cultivation, separation, and redefinition (Kram & Isabella, AMJ 1985). The relationship is not built to stay in cultivation forever. Redefinition, where mentor and mentee become peers, is where the structural shift in authority is supposed to happen. When that redefinition never arrives, both parties usually feel it. Neither knows what to do about it.
Why does the relationship usually last too long?
A useful mentor opens doors, compresses learning curves, and provides political cover at moments when political cover is what makes a career survive. For years, the relationship works. The mentee gains skill and standing. The mentor gets to see their judgment reflected in someone else's growth. The structural arrangement holds because everything inside it is going well.
These are exactly the conditions under which the relationship becomes hardest to end. Kram's four phases describe a healthy lifecycle. In closely-held firms, the cultivation phase often extends well past its useful length. Separation gets postponed because nothing has visibly broken. Redefinition into a peer relationship, the move that should formalize the mentee's emergence, never happens.
Herminia Ibarra at London Business School has shown that identity transitions require working in networks beyond the familiar ones, the people who knew the previous version of you and tend to reinforce it (Working Identity, Ibarra, 2003/2023). Long mentor-mentee relationships often produce the opposite condition: a tightly bonded circle where the mentee's earlier identity is the one everyone keeps talking to.
What does it look like in practice?
The question usually stays dormant until something outside the relationship forces it. A merger absorbs the firm and the name on the door changes. A planned buyout that would have let the next generation take over falls through. Until that kind of shock arrives, the pattern is easy to miss, and it is recognizable from outside the relationship more easily than inside it.
Decisions about the mentee's career still pass through the mentor by default, even when the mentee is fully capable of making them alone.
The mentee accepts compromises they would not accept from anyone else, framed internally as loyalty or gratitude rather than constraint, in a pattern that resembles bowing to the most powerful figure in the room.
The same arguments resurface at every strategic conversation. New decisions are made, but the underlying disagreement is never resolved.
Career progression gets discussed in terms of "when the mentor is ready" rather than "when the mentee is ready."
The mentee has a vague sense that some of their professional ambition is on pause, without being able to name exactly what is pausing it.
The firm's operating logic treats the senior partner's preferences as the default. Other voices are heard but rarely change outcomes.
Spencer Stuart's 2024 research on CEO succession found that 50% of organizations rely primarily on executive coaching as their development mechanism, while only 32% use structured internal rotations that put leaders in roles with real accountability (Spencer Stuart, 2024). That ratio matters here. Coaching deepens the existing relationship. Rotations create the structural break that lets the next generation actually take ground.
What does the redefinition shift require?
The shift the mentee needs is not to leave the relationship. Ending the structural arrangement is not the same as ending the connection. Mentor and mentee can remain colleagues, friends, or occasional advisors. What ends is the assumption that one person's authority defines the other's professional path.
Right now, the mentee sees the relationship as a bond, and the bond is real. With perspective, they would see something else. A structural arrangement that worked for both people for a long time has now stopped working for one of them. Loyalty does not require staying inside an arrangement that no longer fits. Ending a phase is not ending a connection.
The harder part of the shift is that it usually has to come from the mentee, because the senior person is often responding to incentives that point the other way. The redefinition stalls partly because the senior partner benefits from it stalling. His financial return and his retirement are often better served by the mentee staying on his matters, and when an outside event closes off the planned succession, resentment can settle underneath the refusal. None of that makes the senior partner a villain. It makes him a person reading a clear set of incentives. The mentee often cannot name the dynamic, because the language of mentorship makes ending it feel like betrayal, and the firm has no mechanism for the conversation. So nothing happens, until a structural change arrives from outside, and the choice becomes unavoidable.
3Peak Wisdom
In our work with founders and senior leaders, this pattern shows up almost everywhere closely-held firms exist. It is one of the deepest forms of authority by relationship rather than authority by structure. The senior person carries weight by virtue of who they are, not by virtue of where they sit. The next generation absorbs that weight by association. For a while, it is enough.
The relational system has limits. When the next generation matures or the firm grows, the absence of structure becomes the problem. Authority that lives in one relationship cannot scale. Identity formed inside one relationship cannot graduate from it without a structural break. This is why so many succession conversations stall, why senior partners find themselves with capable potential successors who somehow never seem ready, and why mentees in their forties and fifties still describe their professional life in terms of someone else's permission.
Which of your most formative professional relationships has stopped developing you, and what would change if you named that out loud?
Frequently Asked Questions
How do you know when a mentor-mentee relationship has reached its natural endpoint?
Look for the signals academic research has identified for the redefinition phase: the mentee is making compromises they would not accept from anyone else, career conversations keep stalling on the same arguments, and professional decisions still default to the mentor even when the mentee is fully capable of making them alone. Kram's four-phase model, foundational since the 1980s, treats redefinition into a peer relationship as the healthy ending. When that ending does not arrive, the relationship has overstayed.
Can a mentor-mentee relationship continue after the formal mentorship ends?
Yes. Ending the structural arrangement is not the same as ending the relationship. The mentor and mentee can remain colleagues, friends, or occasional advisors. What ends is the assumption that one person's authority defines the other's professional path. Most successful mentorships transition into peer relationships once the mentee has consolidated their own authority. The mentor's role becomes consultative rather than directive.
Why is ending a long mentorship so difficult in closely-held firms?
In closely-held firms, the senior partner's presence often functions as the operating system. People orient around it, decisions pass through it, and the firm has no mechanism for the conversation about transition. Ending a long mentorship here is more than a personal conversation. It is a structural change to how the firm operates, which is why so many of these transitions stall until an outside event forces them.
