At some point you will be mentoring a person you also have to assess. You are their lead, or you write a paragraph about them once a quarter, or your manager stops you in a corridor and asks whether they are ready for the next level. You are now holding two jobs that pull in opposite directions, and pretending you are not is how decent people end up hurt.
The mentoring half only works if they bring you what they are bad at. The half-understood concept. The week they achieved nothing. The task they have been quietly avoiding for a fortnight. The assessing half means everything they tell you is, in some small way, evidence. If they work that out before you have said it out loud, the honest half of the relationship closes and you will not notice for months.
So say it first, plainly, in the first session. I am going to be asked for my view on your work. Anything you tell me here that I think belongs in that conversation, I will tell you it belongs there, and I will give you the chance to say it yourself before I do. That sentence costs you nothing and it buys you years of straight answers.
Then make the assessment boring. No review you write should contain anything they are hearing for the first time. If something would count against them, it gets said in the mentoring seat, early, in ordinary words, while there is still time to do something about it. A surprise in a written review is a failure of your nerve, not of their performance.
Keep some separation in the practical sense too. Do not let the mentoring conversation become the status meeting. Do not open it by asking how the ticket is going. Different room if you can, different day if you cannot.
And be honest about the parts you cannot hold. Whether to leave. Whether to apply elsewhere. Whether your own team is the problem. You are the wrong person for those, and you should say so and help them find a second mentor outside your reporting line who is not compromised.
Two roles, both real. The trick is that they always know which one you are in.
– Asael Shinder
