Feedback going up: your blind spots

A blind spot is an effect of your own behavior that others see and you don't. Every lead has some, and the more senior you become, the fewer people volunteer to point them out. The real question isn't whether you have blind spots, but who tells you about them.

Iryna gave you a clear example at Klasna. On Thursday's call, when Taras asked about a refund, you answered in twenty seconds, before anyone else spoke. From your side it was helpful: the queue was long and you knew the answer. From the team's side it sent a message: refund questions go to the lead. Since then they skipped Olha's rotation. You had no idea.

Notice the shape of a blind spot: good intentions, a real reason, and an effect you can't see from where you sit. That's why it can't be found by reflection alone. It takes someone who sees the effect and feels safe enough to say so, which is what upward feedback is for.

Many new leads end every one-to-one with “Any feedback for me?” and hear “No, all good” every time. It's tempting to conclude there's nothing to fix. There are two better explanations.

The question is too wide. To answer, people have to search their memory on the spot for something critical about you. Most can't, or won't in thirty seconds, so “all good” is the easy answer.

The power gap is real. You now decide their shifts and you'll write their review. Criticizing that person feels risky, even for a former peer who used to tell you everything over coffee.

So silence tells you nothing. It isn't proof that the team has nothing to say, or that they don't care how they're led, and a longer meeting doesn't fix either cause.

Upward feedback, feedback from your reports about how you lead, has to be asked for on purpose, made specific and made safe. The next cards show how.

The fix for silence is to make the question small and the habit regular.

Narrow: ask about one specific thing you did, so the person can answer from a real memory. “When I reply in the team chat within a minute, does it help you or get in the way?” “What's one thing I did in the rotation this week that slowed you down?” “In Monday's call, did I leave enough room for others to answer first?” Each points at one behavior, the same principle as giving feedback on behavior, now turned around.

Often: one such question per one-to-one, every week. When it's routine, it stops feeling like a test of loyalty, and small criticizms become normal rather than dramatic.

Written option: some people find it easier to type than to say to your face, especially across a video call. Let them answer in the shared doc after the meeting.

Avoid wide questions (“How am I doing as a lead?”) and requests for reassurance (“You'd tell me if something was wrong, right?”): both invite “fine”.

An anonymous survey sounds like the safest way to hear the truth. In a team of four it's weaker than it looks.

“Anonymous” is thin. In a small team people guess who wrote what, and so will you. Knowing that, the writers soften their answers or skip the hard ones.

Four answers swing. With so few people, one person's bad week moves the whole result. A score that drops from one survey to the next may mean nothing more than a rough Monday.

That doesn't make surveys useless. Use the result as a hint to ask about, not a verdict: if “the lead takes decisions back” appears, raise it as a neutral question to the whole team. And never try to work out who said it. The moment people suspect you're decoding handwriting, honest answers stop, in the survey and in one-to-ones too.

Narrow questions in regular one-to-ones remain the main channel; a survey is a supplement.

The moment someone actually answers your narrow question is the moment that decides whether they'll answer again. The instinct is to explain. Resist it.

At Klasna you ask Olha: “When I make changes to the rotation, how does that land for you?” She answers: you said she decides how the rotation works, then on Wednesday you changed the order in the chat yourself, without a word. “So do I decide, or not?” The urge is to say Taras was offline that day. The reason may be real, but explaining now sounds like “you're wrong to tell me”.

A good response has five steps: listen to the end without explaining; thank her for saying it; ask for anything else or another example; say what you'll change and by when; and a few weeks later, show her what changed.

And notice she's right about the gap: decision rights you take back on a busy day weren't really handed over. That's the blind spot she just showed you.

People decide whether to be honest again by watching what happens next. A lead rarely means to punish feedback, but small reactions do it anyway.

The common ones: arguing back on the spot. “Well, you also…”, answering criticizm with criticizm. A colder tone that week: shorter messages, fewer thank-yous. Telling others who said it, even as a joke. Any one of these can end upward feedback for months, and not only from that person: the team watches how the first honest comment is treated.

What builds the opposite is visible safety: thanks in the moment, no change in tone, the comment kept private, and later a real change the person can see.

In a team of four, word travels in a day. One punished comment teaches everyone that “no, all good” is the only safe answer, and you're back to seeing nothing.

Sometimes the feedback is unfair, or misses context you have. The instinct is to correct the record on the spot, so the misunderstanding doesn't spread. Don't. Correcting immediately sounds like “you're wrong to tell me”, and people stop telling.

Instead: thank them anyway, and take a day. Then add context later, once, calmly, together with what you will change: “I thought about it. Here's what I'll change, and here's one thing I'll keep, and why.” Context after a pause sounds like reasoning; context in the moment sounds like defence.

Then close the loop. A few weeks later, go back to the person: “You told me the rotation order kept changing. Since then, only you change it.” That sentence is the strongest signal a lead can send. It shows the comment was heard, acted on and remembered, and it earns the next honest answer.

A lead who never closes the loop collects feedback that slowly dries up.

You don't have to act on every comment. Sort them into three piles, each an honest answer:

• Change now: “The rotation order keeps changing without Olha” → only Olha changes it; on a day someone is offline, she decides the swap.
• Explain why not: “Can we stop writing the three-line note on hand-ups?” → no, because without it the next person asks the family the same questions again.
• Think and come back by a date: for what needs checking first, with a day named.

The one dishonest answer is silence.

Watch for repeats. Two people independently naming the same habit is a pattern worth acting on openly. One comment once is a data point: ask the team a neutral question about it, without naming who raised it.

And tell your manager what you changed. Iryna says nothing builds a new lead's credibility faster than “they told me, and I fixed it”.

How it fits together: blind spots need upward feedback; narrow, regular, safe questions get it; listening, thanks and closed loops keep it coming. The next lesson is the month's case: Taras is fast and cuts corners.