There’s a particular kind of feedback that’s easy to keep deferring — and the deferral is so reasonable-looking that you can go a long time without naming it for what it is. It’s not feedback to struggling employees — those conversations, while hard, have an urgency to them that makes them easier to actually have. It’s feedback to strong performers. The people who are delivering, who you like working with, who you’d fight to keep on your team.
At Capital One, I had a direct report — someone I genuinely respected and relied on — who had developed a habit that was quietly causing problems. In team discussions, she would sometimes signal disagreement not by raising it directly but by going quiet and then raising concerns privately afterward, either to me or to peers. The work itself was excellent. Her judgment was sound. But this pattern was creating friction in the team that she couldn’t fully see, because the people on the receiving end weren’t telling her — they were telling me.
For a while, I said nothing. The timing never felt right, or she was in the middle of a difficult project, or I’d convince myself it wasn’t that big a deal. The truth, if I’m honest, is that I didn’t want to risk the relationship. She was high-performing and I was worried that raising something difficult might land badly and put distance between us. So I let it go. Once, twice, a few times.
This is a trap that leadership books talk about but don’t quite convey the texture of. The instinct isn’t malicious — it comes from a genuine place of wanting to protect someone you value. But what it actually does is deny them information they need. She was operating with an incomplete picture of how her behaviour was landing, and I was the one who had that picture. By staying quiet, I wasn’t protecting her. I was just making myself more comfortable.
When I finally raised it — a few months later than I should have, during a 1:1 that I’d intentionally kept low-stakes — her reaction was not what I’d feared. She was quiet for a moment, then said: “I suspected this was happening. I didn’t realise it was as visible as you’re describing” — and what followed was one of the better conversations I can remember having. She knew exactly what I was talking about once I named it, and she already had some insight into where it was coming from.
What I carried away from that experience is something I think about a lot: the longer you wait to give feedback, the more you’re implicitly telling someone that their behaviour is fine. Every week you stay quiet is a week they spend reinforcing the pattern, thinking they have your tacit approval. And when the feedback finally comes, it carries the additional weight of: why didn’t you tell me sooner?
There’s also a more subtle cost. When you develop a habit of comfortable silence with strong performers, you stop actually managing them — you just manage around them. You work the periphery instead of having the real conversation. And they can feel that, even if they can’t name it. The relationship becomes more careful, not closer.
The test I try to use now: if I’m holding back something I believe would genuinely help someone grow, the discomfort I’m feeling is not a reason to stay quiet — it’s the feedback telling me this conversation matters. The discomfort is roughly proportional to how much the person needs to hear it.
I also try to give feedback much earlier now, not because early feedback is easier, but because it’s less loaded. If I raise something at week three, it’s information. If I raise the same thing at month eight, it comes with eight months of implied history that makes both of us defensive.
The feedback I almost didn’t give ended up being something she said was useful — I’m glad I gave it. What changed is that I stopped treating discomfort as a signal to wait, and started treating it as a signal to go.