Latest / Community Marketing with Fexingo: Discord, Slack Groups, and Brand Communities / Why Brand Communities Need Exit Interviews Too
Transcript
- Lucas: So there's this moment in every community manager's life that nobody warns you about. You get the notification — member left. And if you're like most managers, you shrug and move on. New people join, old people leave, that's the game. Right? Luna: Yeah, I've definitely had that reflex. It's easy to focus on the growth number and ignore the leak. Lucas: Exactly. But a few months ago I came across a case that flipped that logic. A B2B SaaS company — they run a Slack community of about 50 product managers — decided to actually talk to every single person who left. Not a survey. A real conversation. And what they learned cut their churn rate by 30 percent in one quarter. Luna: Thirty percent. From exit interviews? That's a massive shift for a community that size. Lucas: It is. And the thing is, the community wasn't dying. It had steady growth, decent engagement metrics. But the manager told me she kept seeing the same names vanish after about six months. So she started a practice she called 'the exit conversation.' No automated email, no form. Just a 15-minute Zoom call with anyone who left the workspace. Luna: Did people actually agree to that? I feel like most folks who leave a Slack group just ghost. Lucas: That's what she expected too. But she sent a very short, personal message — something like, 'Hey, I noticed you left. No pressure, but I'd love to hear what we could do better. Coffee's on me.' And about 40 percent of people took her up on it. That's 40 percent of departing members willing to tell you exactly why they left, if you just ask the right way. Luna: Okay, that's a better response rate than most NPS surveys I've seen. What did she hear? Lucas: The big surprise — and this is the core of the episode — was that the reason most people gave wasn't about content, or spam, or pricing. It was about timing. They said the community had been useful early on, but after a few months, the conversations shifted to topics they'd already mastered. So they weren't learning anything new. They felt like the room had moved past them. Luna: Right — so it wasn't that the community was bad, it was that the community's center of gravity had moved to a more advanced level, and the early members got left behind. Lucas: Exactly. And that's a structural problem, not a content problem. Most communities have a single channel, a single feed, a single cadence. They treat all members the same. But the member who joined three months ago has completely different needs from the one who joined two days ago. And if you don't segment your community experience by tenure, you lose the people who feel like they've aged out. Luna: I love that phrase — 'aged out.' So what did she actually do about it? She can't exactly make the veterans dumb down their conversations. Lucas: She created what she called 'cohort-based channels.' When a new batch of members joined — she onboarded in cohorts of 10 every month — they got a private channel that only they could see for the first 90 days. That channel had a structured curriculum: week one is fundamentals, week two is case studies, and so on. After 90 days, the channel opened up into the main group. But by then, those members had built relationships with each other and had enough context to engage with the veterans. Luna: So it's like a graduated onboarding. The exit interviews told her that the problem wasn't the community — it was the drop-off between the newbie experience and the experienced experience. Lucas: Bingo. And she told me the single most actionable question in those exit conversations was this: 'At what point did you feel like this community was no longer for people like you?' That question surfaced a specific month — month four. Month one through three was great. Month four, people started feeling the gap. So she designed the cohort channel to last exactly 90 days, and then built a 'bridge' week where the cohort and the main group did a joint project together. That transition moment was the key. Luna: I think this is one of those ideas that sounds obvious in hindsight but almost nobody does. I've been in dozens of communities and I've never once been asked to give an exit interview. The closest I've seen is a generic 'why are you leaving?' dropdown with options like 'too many emails' or 'not relevant.' Lucas: Right. And those dropdowns give you shallow data. They tell you what, but not why. The real insight lives in the conversation. And honestly, if today's episode sparked something you've actually used in your own community — maybe you're thinking about setting up exit conversations yourself — that's the kind of thing that makes this show worth doing. If it was worth a coffee to you, you know where to find us: buy me a coffee dot com slash fexingo. Luna: Yeah, it's a small way to keep these conversations ad-free. And we love hearing that the ideas land in the real world. Lucas: So back to that cohort channel idea. The manager also found something unexpected in those exit calls. Several people said they left not because they stopped caring, but because they felt like they couldn't keep up with the volume. The main Slack channel had something like 200 messages a day. So even if the content was relevant, the firehose was overwhelming. Luna: That's a common problem, especially for working professionals who check in once or twice a day. The channel moves too fast. Lucas: Exactly. So she added a weekly digest. Every Friday, a bot compiled the top five threads from the week based on reactions and replies. That digest went into a read-only channel and also got emailed out. She said that single change reduced the 'information anxiety' complaints by almost half in the next quarter. Luna: And I'm guessing the digest also helped with the aging-out problem, because people who were behind could catch up on what mattered without having to scroll through everything. Lucas: Yeah, it's a low-effort retention tool. But here's the thing — she only knew to build it because she asked the departing members. The exit conversations gave her a list of specific pain points that she could prioritize. Before that, she was guessing. She was adding more content, more channels, more events. But the exit interviews told her: stop adding, start curating. Luna: I want to talk about the actual script she used. What were the questions? Because I think a lot of community managers would love a template. Lucas: Sure. She had five questions. One: 'What was the moment you first felt this community wasn't for you?' Two: 'Was there a specific conversation or interaction that made you want to leave?' Three: 'What was the most valuable thing you got from the community?' She always asked that one first, to start on a positive note. Four: 'If you could change one thing about this community, what would it be?' And five: 'Is there anything we could have done to keep you?' Luna: That fifth question is interesting because it forces the leaver to imagine a counterfactual. It's not just 'why did you leave?' — it's 'what would have kept you?' Lucas: Right. And she said the answers to that question were often surprisingly small. Things like 'send a weekly summary' or 'have a mentor for new members' or 'offer a beginner track.' None of those were expensive or time-consuming to implement. Luna: So the exit interviews basically became a product roadmap for the community. She was using churn as a signal for where to invest. Lucas: Exactly. And she measured the impact. After she implemented the cohort channels and the weekly digest, she tracked the retention of the next three cohorts. Churn at month four dropped from about 20 percent to under 10 percent. And overall monthly churn went from 8 percent to about 5.5 percent. For a 50-person community, that's a huge difference in the feeling of momentum. Luna: I wonder if this applies to larger communities too. Like, could a Discord server with 10,000 members do exit interviews? Probably not one-on-one. Lucas: You'd need a different method at scale. But the principle scales. You could sample a percentage of leavers, or do a short structured call with a random subset. The key is the conversation format — not a survey. Surveys give you correlations; conversations give you mechanisms. And mechanisms are what you can fix. Luna: So for a community manager listening right now, what's the one thing to do this week? Lucas: Pick three people who left your community in the last month. Send each of them a personal note — no templates — and ask for 10 minutes. Use those five questions. Take notes. I guarantee you'll learn something that changes how you think about your community. And if you do, let us know. That's the kind of feedback that makes this show better. Luna: I love that. It's a low-risk experiment. Worst case, you get ignored. Best case, you uncover a hidden churn driver. Lucas: And remember, the goal isn't to win everyone back. Some departures are healthy. But if you're losing people who genuinely valued the community, it's worth understanding why. That's the exit interview mindset.