She never missed a one-on-one. That was the thing she’d have told you if you’d asked whether she was a good manager, and it was true.
Eight direct reports, every other week, thirty minutes each, on the calendar in ink. In two years she’d rescheduled maybe four times.
Then her strongest engineer resigned, and the exit interview said she’d been thinking about it for seven months.
(A composite. The pattern is real; the company isn’t.)
Seven months. Fourteen one-on-ones.
What was happening in those thirty minutes
They were good meetings, in a sense. They started on time. They had a rhythm.
The rhythm was: what are you working on, where is it, what’s blocked, what’s next. Efficient, useful, and completely replaceable by a written update — which is the test.
If the meeting could be replaced by a message, it already has been. You’re both just still showing up.
In fourteen consecutive meetings, this manager had never once been surprised. Not once had someone said something she didn’t already know or couldn’t have predicted. She’d have told you the meetings were going well. That was the evidence they weren’t.
The signal she missed
The engineer hadn’t been unhappy in an obvious way. Output never dropped — if anything it rose, which is common, because people who’ve decided to leave often become easier to manage, not harder.
What changed was that she stopped raising things. Around month three she went from proposing ideas in team meetings to executing what was decided. Nobody noticed, because nothing was wrong. There was no complaint to respond to, no conflict, no missed deadline.
People rarely go quiet because things are good. They go quiet because they’ve concluded that speaking up isn’t worth the cost, and that conclusion is usually made once and then never revisited.
Output is a lagging indicator. By the time it moves, the decision is months old.
What a real one-on-one requires
One question the manager genuinely doesn’t know the answer to, and then silence long enough to be uncomfortable.
That’s most of it. The status portion can be a message. The half of the meeting that can’t be automated is the half where you ask something open and then don’t rescue the pause.
Leaders avoid this for two reasons: it feels like a waste of the thirty minutes, and it risks surfacing something you’d then have to deal with. Both are smaller obstacles than they feel in the moment. The alternative is finding out in an exit interview.
Where to start this week
Move status to writing. Ask for the project update in a message beforehand. It frees the whole meeting and costs nobody anything.
Ask one question you can’t predict. “What’s something we’re doing that you’d change if it were up to you?” Then stop talking. The silence does the work, not the question.
Audit for surprise. Look back at your last five one-on-ones with someone. If you weren’t surprised once, you haven’t been running a conversation — you’ve been attending a report.
More on this is in Chapter 6 — Effective Individual Communication of Firm Foundations.
Working on this with a real team and want a second set of eyes? Start a conversation, or book a 30-minute call. No pitch — just a look at what you are actually dealing with.
Leave a Reply