When someone brings me a problem, every instinct I have says solve it. I can see the answer, or think I can, and handing it over would take thirty seconds and feel great. I have learned to sit on my hands instead, because the fast answer is usually the worst thing I can offer. It solves today’s problem and teaches the person nothing except to come back to me next time.
Mentoring is not consulting. A consultant is paid to deliver answers. A mentor is trying to build someone who generates their own answers, and you cannot build that capacity by doing the thinking for them. Every time I solve it for someone, I make myself a little more necessary and them a little more dependent, which is the reverse of the job. The goal is to become unnecessary, and you get there by talking less than feels natural.
So I ask instead of tell. Not in a coy, withholding way, where I clearly know the answer and am making them guess at it like a quiz. People hate that, and rightly. The questions I mean are real ones. What have you already tried. What are you afraid will happen if you pick the obvious option. What would you tell a friend in this spot. Half the time the person talks their own way to a better answer than I would have given, and now it is theirs, which means they can do it again without me.
The questions also surface the actual problem, which is usually not the one presented. Someone comes in asking how to handle a difficult colleague and twenty minutes of decent questions reveals the real issue is that they are afraid of conflict in general, and this colleague is just where it showed up. Solve the presented problem and they are back next month with a new colleague. Help them see the pattern and you have given them something that travels.
I try hard not to project my own path onto them. What worked for me worked partly because of who I am, and the person across from me is not me. Their strengths are different, their constraints are different, their tolerance for risk is different. Advice that amounts to do what I did is often quietly useless, and sometimes harmful, because it pushes them toward a version of success built for someone else’s temperament.
The hardest discipline is letting them make a mistake I can see coming, when the mistake is survivable. I want to grab the wheel. But a small failure they choose and then recover from teaches more than a success I hand them, and they remember the lesson because they earned it. I save the wheel-grabbing for the genuinely dangerous calls, which are rarer than my anxiety suggests.
The whole thing comes down to a kind of restraint that does not come naturally to people who are good at solving things. You hold back the answer you have so the other person can find one that is actually theirs. It is slower. It works. And the best sign you did it right is that, after a while, they stop needing to ask.