Latest / New Manager Conversations with Fexingo: First 90 Days, Managing Former Peers, and Leadership Basics / How New Managers Can Run a Post-Mortem That Actually Prevents Rework
Transcript
- Lucas: One of the hardest things a new manager faces is the first time a project goes sideways on their watch. Not just a missed deadline — a real, 'we shipped it and it broke' kind of failure. Luna: Right, and the instinct is often to just move on. Fix the bug, apologize, promise it won't happen again. But that's a recipe for the exact same thing happening three months later. Lucas: Exactly. So today I want to talk about post-mortems — specifically, how a new manager runs a post-mortem that actually prevents rework. And I've got a concrete example from a company I've been following. Luna: Let's hear it. Lucas: Mid-market SaaS company, about 80 engineers. A new manager named Priya took over a team that had just launched a feature that was three weeks late and had a critical bug in production on day one. Luna: Oof. Tough spot for a first month. Lucas: Right. The senior director told her to 'run a post-mortem,' but didn't give her any structure. So she had to figure it out. And the way she approached it turned that failure into a process that cut rework by 40% over the next two sprints. Luna: Forty percent — that's a real number. How did she do it? Lucas: She used a three-question framework. And the first question is the most important: 'What actually happened, in timeline form?' Luna: So not 'whose fault was it' — just the sequence of events. Lucas: Exactly. She pulled everyone into a room — remote, actually, but same idea — and had them write out what they remembered on sticky notes on a Miro board. Then they aligned on a single timeline. No blame, just facts. Luna: I've seen this go wrong when the manager starts with 'okay, what went wrong?' People get defensive immediately. Lucas: Right. The timeline defuses that. Because everyone can agree on 'we had a code freeze on Tuesday, the last commit was Wednesday at 3pm' — that's not arguable. Once you have that shared reality, then you ask question two. Luna: Which is? Lucas: 'At what point could we have known this was going off track, and why didn't we?' Luna: That's the systems-level question. It's not about who — it's about the process. Lucas: Exactly. In Priya's case, they identified that the QA sign-off was happening too late — the team was testing after the feature was fully built, instead of incrementally. That one insight led to a change: they moved to a 'test as you go' approach per sprint. Luna: And that's the kind of change that prevents rework because it catches issues earlier. Lucas: Exactly. The third question is the one most post-mortems skip: 'What exactly are we going to do differently, who owns it, and when is it due?' Luna: So it's not just 'we should do better next time' — it's a concrete action. Lucas: Right. Priya made sure every action had a single owner and a deadline. No more than three actions per post-mortem. The team committed to: one, moving QA to start mid-sprint. Two, adding a 'pre-flight checklist' before any production release. Three, a five-minute stand-up every morning during the final week of a sprint. Luna: And she stuck to those three. That's the discipline part. Lucas: That's the discipline. And she followed up two weeks later in a team meeting — 'how are these going?' That follow-through is what makes the post-mortem actually stick. Luna: So what's the one phrase new managers should never say in a post-mortem? Lucas: 'Next time, let's just be more careful.' That's not an action. It's a wish. And it basically tells the team nothing will change. Luna: Yeah. It's like saying 'let's try harder.' That's not a system. Lucas: Exactly. The whole point is to make it so the next person doesn't have to rely on heroics or luck. The system should catch the mistake before it becomes a crisis. Luna: And that builds credibility for the new manager, too. When the team sees that their time isn't wasted in a blame session — they actually get better processes out of it — they'll trust you more. Lucas: A hundred percent. Priya told me that after two post-mortems, the team started asking to schedule them proactively after any medium-sized project, not just the failures. They saw value in it. Luna: That's the dream outcome, right? Post-mortem becomes a habit, not a punishment. Lucas: Yeah. And that's the mark of a mature team. So if you're a new manager, don't wait for a disaster. Start with a small project that went fine — run a 'what could we improve' session. Practice the framework when the stakes are low. Luna: And you've mentioned Priya a few times — is there a resource or a template you'd recommend for people who want to try this? Lucas: Yeah, there's a good one from Etsy's engineering blog — they call it a 'blameless post-mortem' template. It's basically the three questions we just walked through. I'll link it in the show notes. Luna: That's great. And speaking of resources that people actually use — you know, we keep this show ad-free because of listeners who find value in it and want to support it directly. Lucas: Yeah, that's right. If today's episode was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. It's the smallest thing, and it genuinely keeps us going. Luna: Absolutely. No pressure, just if it's useful. And back to the post-mortem — one thing I want to add: don't forget to celebrate what went right, too. Lucas: Oh, that's a great point. The three-question framework actually has a hidden fourth question: 'What did we do well that we should keep doing?' Luna: Because if you only focus on the failures, the team starts to dread these meetings. Lucas: Exactly. Priya made sure to start every post-mortem with a round of 'one thing that went better than expected.' That sets a constructive tone. And it also surfaces things you might accidentally change. Luna: So the overall framework is: timeline, detection point, action items, plus a positive round. Lucas: That's the whole thing. And the beauty is, it doesn't require any authority or seniority. Any new manager can run this with their team tomorrow. Luna: And in your first 90 days, that kind of structured, calm leadership is exactly what builds trust. Lucas: It really is. So if you're a new manager facing your first project failure — or even just a project that could have gone better — try the three-question post-mortem. Your team will thank you.