Latest / New Manager Conversations with Fexingo: First 90 Days, Managing Former Peers, and Leadership Basics / How New Managers Can Give Feedback That Sticks
Transcript
- Lucas: So you've been in the manager role for maybe six weeks now. You've done the listening tours, you've got the one-on-ones on the calendar. But there's this one person on your team — let's call her Priya — who is smart, experienced, and she keeps delivering work that's ninety percent there. The last mile is always missing. And you know you need to say something, but every time you start, you either sugarcoat it so much she doesn't hear the critique, or you come across harsher than you intended and she shuts down. Luna: That gray zone — where the person is good but not great on a specific thing — is honestly harder than dealing with a clear underperformer. Lucas: Much harder. Because with underperformance, the gap is obvious. Everyone sees it. You have data, you have a Performance Improvement Plan template. But with the ninety-percenter, you're trying to get someone from good to excellent without making them feel like you think they're failing. Luna: And the stakes are high because if you don't address it, the behavior becomes the standard. The rest of the team watches you not correct it, and they assume that's the bar. Lucas: Exactly. So today I want to offer a specific framework for giving feedback that actually changes behavior. It's called the Feedback Sandbox. I first heard this from a leadership coach named Kim Scott — the same person who wrote 'Radical Candor' — but I've adapted it a bit for new managers. Luna: I love the sandbox metaphor. It implies a contained, safe space where you can try things and it's fine if you knock something over. Lucas: That's exactly it. The sandbox has four walls. Wall one: you describe the specific behavior you observed — not the person's character, not your interpretation, just the action. Wall two: you describe the impact that behavior had on you, the team, or the work. Wall three: you ask a genuine question about their perspective. Wall four: you co-create a next step. Luna: Can we run that with the Priya example? Lucas: Absolutely. So Wall one: 'Priya, in the last two client reports you sent me, I noticed each had a few data points that were outdated — the Q3 numbers instead of the Q4 ones.' That's it. No 'you're sloppy,' no 'you need to pay more attention.' Just the behavior. Luna: And that already lowers the temperature. It's factual. It's not an attack. Lucas: Right. Wall two, the impact: 'When I see those numbers, I have to stop and verify them before I send the report up. That adds about an hour of my time, and it also creates a risk that an outdated figure slips through.' Luna: Now you're explaining why it matters — not just that it's wrong, but the consequence. Lucas: Wall three: 'What's your process for updating those data tables? Is there something in the workflow that makes it easy to miss the refresh?' This is the critical step. You're not assuming you know the root cause. Maybe she's pulling from a shared drive where someone else controls the file. Maybe she's rushing because she's getting the data at 5pm. You don't know until you ask. Luna: And just by asking, you signal that you're on her side. You're troubleshooting together, not blaming. Lucas: Wall four: 'Okay, so it sounds like the data source isn't clearly labeled. How about we add a version date to the file name, and I'll also check in with you the day before a report is due to flag any data pulls you might need?' That's the co-created next step. She owns the labeling, you own the check-in. Luna: And all four walls are contained in one conversational sandbox. You don't pile on past mistakes. You don't bring up a different issue. You solve this one thing. Lucas: One thing I'd add: the research on feedback suggests that the ideal ratio of positive to corrective feedback is about five to one. That's from the work of psychologist Marcial Losada. So if you're having a conversation that's mostly corrective, you need a reservoir of specific praise you've given that person in recent weeks. Otherwise even a well-delivered sandbox can feel like an attack. Luna: That five-to-one number — I've seen it cited in a lot of team dynamics studies. It's not just a nice to have. Brains actually process negative input more intensely. It's called the negativity bias. Lucas: Right. So before you step into a sandbox with someone, make sure you've genuinely noticed and named the things they're doing well. And here's a trap I see new managers fall into: they think praise has to be big and formal. It doesn't. A simple 'Hey, I saw you handled that client question really well — you kept your cool and gave a clear answer' in the hallway can be a deposit in that emotional bank account. Luna: I also think timing matters. Never open a one-on-one with corrective feedback. Start with something positive or a check-in on their wellbeing. Then ease into the sandbox. Lucas: Absolutely. And never say 'Can I give you some feedback?' That almost always spikes anxiety. The person's amygdala goes into threat detection mode and their ability to actually hear what you say drops by about thirty percent. Luna: So what's the alternative opener? Just start with the behavior, like you did? Lucas: I'd recommend a soft lead-in that signals context without asking permission. Something like: 'I want to walk through something I noticed in the last report, so we can tighten up the process. Sound good?' That's collaborative, not permission-seeking. They'll almost always say yes, and their guard is slightly lower because you've framed it as a process improvement, not a personal critique. Luna: And if they get defensive anyway? What then? Lucas: That happens. Especially with high performers. When someone gets defensive, don't push harder. Stay inside the sandbox. You can say: 'I hear that this feels like I'm criticizing you. That's not my intent. My goal is to make sure your work gets the recognition it deserves, and I don't want outdated numbers to undermine that.' Then pause. Let them respond. Often the defensiveness is fear that you think less of them. Reassure without backing down from the issue. Luna: I want to talk about another common scenario: giving feedback to someone who used to be your peer. That carries extra weight because the relationship history is different. Lucas: Huge. If you were peers three months ago, and now you're their manager, the power dynamic is fresh. They might be hypervigilant to any sign that you're 'acting like a boss.' So the sandbox is even more important, because it's transparent and collaborative. You're not pulling rank. You're saying 'Here's what I saw, here's the impact, what do you think?' Luna: And you might need to acknowledge the history explicitly. Like, 'I know we used to work side by side, and I want to be upfront that this feedback comes from a place of wanting the team to succeed, not from any change in how I see you.' Lucas: That's a good line. It's honest. It names the elephant. And then you go into the sandbox. The framework works because it's not about authority. It's about shared goals and observable facts. Luna: Let's talk about the opposite problem: what about giving positive feedback? Do we need a sandbox for that too? Lucas: Great point. Positive feedback is simpler but people often mess it up by being vague. 'Great job' doesn't tell the person what to repeat. Effective praise follows the same behavioral-specific model: 'When you caught that data error before the client saw it, you saved us a lot of embarrassment. Thank you.' That's wall one — behavior. Wall two — impact. You don't need the question or the next step for praise, but the specificity is what makes it stick. Luna: So the rule is: specific behavior, specific impact, every time. Lucas: Every time. And here's a practical habit I recommend to new managers. At the end of each week, jot down one specific piece of positive feedback and one specific piece of corrective feedback you want to deliver the next week. That way you're not scrambling in the moment. You're intentional. Luna: I think that's a great takeaway. And it ties into something we've talked about before — being deliberate about your leadership, not reactive. Lucas: Exactly. And on that note, I want to mention something for anyone listening who's found value in this show. We keep the podcast ad-free because of listeners who support us at buy me a coffee dot com slash fexingo. It's a simple way to say 'this matters to me,' and it helps us keep making episodes like this one. If you've ever used the feedback sandbox in a real conversation and it worked, that's the kind of impact we're trying to multiply. Luna: Yeah, and it's completely optional. We just appreciate that you're here, thinking about how to be a better manager. That already puts you ahead. Lucas: So back to the sandbox. One more tip: when you're the one receiving feedback — and as a new manager you absolutely should be — model the behavior you want to see. When someone gives you feedback, don't defend. Say 'Thank you, that's helpful. Can you tell me more about what you saw?' That shows your team that feedback is safe. Luna: And it builds your own feedback culture from the top down. Lucas: Alright, let's do a quick recap. The feedback sandbox has four walls: behavior, impact, question, next step. Use a five-to-one praise ratio. Never start with 'can I give you feedback?' Always be specific. And if you're managing former peers, name the dynamic. Those are the tools. Practice them in a low-stakes setting first — with a peer manager, not a direct report — until they feel natural. Luna: I think the hardest part is just starting. But once you do it a couple times, it becomes part of your rhythm. Lucas: It does. And the payoff is a team where people actually grow, because they know what to keep doing and what to change. That's the whole point.