Latest / Manager Mondays with Fexingo: People Management, Team Leadership, and First-Time Managers / Why Your Team Needs a Blameless Postmortem Culture
Transcript
- Lucas: So there's a moment every manager dreads — you get the Slack message that something broke, and it's bad. Maybe a customer-facing outage, maybe corrupted data, maybe a missed compliance deadline. Luna: And your first instinct is to ask: 'Who did this?' Lucas: Exactly. And that instinct is the problem. Because when you lead with blame, people go quiet. They hide mistakes, they cover their tracks, and you lose the very information you need to prevent the next one. Luna: So what do you do instead? I know the popular answer is 'blameless postmortems,' but I think a lot of managers hear that and think it means no accountability. Lucas: Right — and that's the biggest misconception. A blameless postmortem doesn't mean nobody's responsible. It means you separate individual intent from systemic causes. You assume good faith. You ask: 'What in our system — our processes, our tools, our communication — allowed this to happen?' Luna: I've seen teams where a postmortem is just a blame session dressed up with a nicer name. So how do you actually do it right? Lucas: Let me walk through a real example. A SaaS company — about 150 engineers — had an incident where a configuration change accidentally deleted four hours of customer data. The engineer who made the change was junior, it was their second week, and they were following a runbook that had an outdated step. Luna: Oh, that's painful. Classic systemic failure. Lucas: Exactly. The old culture would have been: fire the junior engineer, or at least put them on a performance improvement plan. But the engineering VP had been building a blameless culture. So they convened a postmortem within 48 hours — and they invited the junior engineer, but as an observer, not a target. Luna: Interesting. So how did they structure the meeting? Lucas: Five parts. First, timeline creation — everyone who was involved contributed a chronological log of what happened, in a shared doc, before the meeting. That way no one gets ambushed. Second, impact assessment — how many customers affected, how much data lost, duration of the outage. Luna: So you're not avoiding the severity. You're just not assigning personal blame. Lucas: Right. Third, root cause analysis — they dug into why the runbook was outdated. Turns out the team that maintained it had been understaffed for six months. Fourth, contributing factors — they listed everything from 'no automated rollback' to 'the junior engineer didn't have a peer review step.' Luna: And the fifth part? Lucas: Action items with owners and deadlines. Every factor got a concrete fix. The outdated runbook? Assign a dedicated owner to audit it quarterly. No automated rollback? Prioritize building one. No peer review? Add a mandatory approval step for config changes. Luna: So the outcome wasn't punishment — it was a roadmap for improvement. Lucas: Exactly. And here's the kicker: within six months, the team reduced production incidents by 83 percent. Because they addressed the system, not the person. Luna: Eighty-three percent — that's huge. But I want to push back on the 'blameless' framing. Some managers I've talked to say it lets people off the hook. If someone repeatedly makes the same careless mistake, isn't there a point where it becomes a personal accountability issue? Lucas: That's a fair point. Blameless doesn't mean consequence-free. If someone repeatedly violates a process they've been trained on, that's a performance conversation. But the blameless postmortem is about the first failure. You assume good intent until patterns prove otherwise. Luna: So you're saying: one mistake? System problem. Repeated mistakes? Individual problem. Lucas: That's a good shorthand. And even then, you still ask: 'Why did we let this pattern persist?' Maybe their workload is too high, maybe they need more support. But yes, at some point it becomes a performance issue. Luna: How do you introduce a blameless postmortem to a team that's used to a blame culture? I imagine you get pushback. Lucas: You start small. Pick an incident that's moderate — not catastrophic, but significant enough to matter. Say to the team: 'Let's try a different approach this time. We're going to focus on what in our system allowed this to happen. No names, no blame. Anyone can speak freely.' Luna: And if someone starts pointing fingers during the meeting? Lucas: You redirect gently. Say: 'That's a valid observation about the person's action. But let's ask what led them to take that action. What in the process guided them there?' The facilitator has to model the behavior. Luna: I've also heard of teams using a 'safety score' — asking everyone to rate how safe they feel to speak up, on a scale of 1 to 5. And if the average is below 4, you know you have a culture problem. Lucas: That's a great practice. Google's Project Aristotle found that psychological safety was the number one predictor of team effectiveness. A blameless postmortem is one of the most concrete ways to build it. Luna: Alright, but let's get tactical. For a manager who's never done one, what's the one thing they should focus on to get started? Lucas: The one thing: separate the 'what' from the 'who.' Write the timeline based on actions and system events, not people's names. Frame every question as: 'What in our process caused this step to fail?' Instead of: 'Why did you do that?' Luna: And the action items — how do you make sure they don't just collect dust? Lucas: Each action item needs a single owner and a deadline. No 'the team will handle it.' And then you track them like any other project. In the SaaS company example, they reviewed outstanding action items at every weekly engineering stand-up until they were closed. Luna: That's smart. Because the whole point is to prevent the next incident, not just to feel better about the last one. Lucas: Right. And here's something I've seen managers miss: after you close the action items, you should circle back to the person who made the initial mistake — if they know you made changes based on their incident, it reinforces that the system works, and they'll be more likely to report the next one. Luna: That's a nice feedback loop. It also builds trust. Lucas: Exactly. And trust is what makes a team resilient. When people know they won't be punished for surfacing problems, they surface them faster — and you catch small issues before they become big ones. Luna: Speaking of small things that become big — I want to mention something that keeps this show going. We don't run ads, and that's intentional. A lot of listeners have told us they appreciate that. If you've gotten value from these episodes, buying us a coffee — literally, at buy me a coffee dot com slash fexingo — helps us keep it that way. It's a small amount, but it genuinely makes a difference. Lucas: Yeah, we really mean that. Every contribution helps cover hosting costs and keeps the podcast independent. So if you've enjoyed today's episode, that's the best way to support us. Luna: Alright, back to blameless postmortems. One last thing — how do you handle a situation where the failure was caused by a clear violation of policy? Like, someone bypassed a security protocol deliberately. Lucas: That's a different category. If it's a deliberate violation — someone knowingly broke a rule for personal gain or out of malice — that's not a postmortem situation. That's a disciplinary process. But even then, you ask: 'Why was it possible for them to bypass the protocol?' Maybe the protocol was too hard to follow, or there were no automated guardrails. The blameless lens still gives you systemic insights. Luna: So it's not about being soft on people — it's about being rigorous about systems. Lucas: That's the core. Blameless doesn't mean consequences don't exist. It means you don't let your judgment be clouded by the search for a scapegoat. You fix the system first, then address individual behavior if the pattern persists. Luna: I think that's a good place to leave it. Next week, I want to talk about something I've been wrestling with: how do you manage a team where you're the youngest person in the room? Lucas: Oh, that's a great one. I have some thoughts — and some mistakes I've made. Let's do it.