For a long time I thought being a good manager meant being useful fast. Someone brings you a question, you know the answer, you give it. Thirty seconds, problem solved, everyone moves on.
What I did not notice for a while was the pattern that builds underneath. The questions kept coming, and they kept getting smaller.
Speed teaches people not to decide
If the answer is always available in thirty seconds, working it out yourself is irrational. Why spend twenty minutes reasoning through an edge case when the person who already knows is one message away? So people stop reasoning through edge cases, and then a queue forms in front of you made entirely of decisions other people are capable of making.
In a written, distributed team this is worse, not better. Answering is easy and asynchronous. Nobody sees your face when you sigh.
What I changed
Two questions before any answer. What have you already tried? And: what do you think we should do? Not as a test — I say plainly that I want their read first, and that I will give mine after.
The first time, it costs more than thirty seconds and feels obstructive. By the fourth or fifth time, the message arrives already containing the analysis: here is the case, here is what I think, here is what I am unsure about. That message takes me fifteen seconds to answer, and the person who wrote it did not need me at all — they needed confirmation, which is a different and much cheaper thing.
Where it does not apply
Live incidents. Anything with a deadline measured in minutes. A distressed user waiting. In those moments coaching is self-indulgent and the right move is to answer, act, and talk about it afterwards.
The test I use now is simple: is this urgent, or does it just feel urgent because someone is waiting on me? Most of the time it is the second one, and the second one can afford a question.