Daniel sent us this one — he's been looking at our analytics dashboard, hundreds of thousands of listens, and he's realizing something that hits a lot of producers at exactly this stage. He knows people are listening. His wife listens while they cook dinner. He sends episodes to friends. They tell him they're listening. But beyond country-level analytics, he has no idea who any of these people are. They don't know each other. They can't talk to each other. And he's asking what we can actually do about that — what platforms, what strategies, what framework — to turn a hidden audience into an identifiable community. He specifically asked whether a Facebook Group is a good idea, or if that's too old school.
That tension he's feeling — between knowing the numbers are real and knowing almost nothing about the humans behind them — that's not a Daniel problem. That's the structural reality of podcasting. You put something out into the world, you see download counts tick up, and you're essentially staring at a crowd through frosted glass. You can make out shapes. You can't see faces.
Which is weird when you think about it, because podcasting is the most intimate medium we have. Someone's voice in your ears while you're doing dishes or driving. It's closer to a conversation than anything else in media. And yet the relationship is completely one-directional.
And here's why this matters right now — not as a nice-to-have, but as a survival question. As podcasting matures, the shows that stick around aren't the ones with the biggest download numbers. They're the ones with the most dedicated communities. Downloads measure reach. They don't measure relationship. And in a world where anyone can launch a podcast in an afternoon, relationship is the only moat that actually holds.
Daniel framed this as "I need to put the framework in place." I think that's exactly right. You don't stumble into community. It has to be designed.
The design question breaks into three layers. Where do you gather people? How do you actually get them to show up? And what do you do once they're there to keep them coming back? If you skip any one of those, you end up with what most podcast communities become — a launch spike, three weeks of enthusiasm, and then a ghost town where the last post was someone asking if the show is still active.
The graveyard of good intentions.
So let's walk through this properly. Layer one is where to gather. And Daniel asked about Facebook Groups specifically — is that too old school? I want to take that question seriously, because I think the knee-jerk reaction from a lot of producers is "Facebook is for grandparents, I'll use Discord." But the data tells a more complicated story.
What's the data say?
Facebook Groups still have one point eight billion monthly active users as of early twenty twenty-six. The per-member engagement rates — when the group has a clear purpose beyond "talk about the show" — still outperform most alternatives. The platform's not dead. What's dead is the idea that you can just create a group, drop a link, and watch community happen.
The objection to Facebook Groups isn't that they don't work. It's that most people use them badly.
The Buzzsprout research on community building emphasizes something that sounds obvious but almost nobody does — you need to choose your platform based on your audience's listening behavior, not your personal preference. If your listeners are cooking dinner, commuting, walking the dog — they're not sitting at a keyboard ready to jump into real-time chat. Discord is a fantastic platform, but it's synchronous by design. The conversation moves while you're away. If your audience listens asynchronously, your community platform should probably be asynchronous too.
That's the Transistor research too, right? Async-first platforms — forums, email lists — outperform real-time chat for audiences that listen during chores or commutes.
And that's a genuinely non-obvious finding, because the industry trend for the last five years has been "put everything on Discord." Discord is great for gaming, live events, anything where people are already at their computers. But for a podcast audience that listens in pockets of downtime, asking them to keep up with a Discord server is like asking someone who reads a book before bed to also join a book club that meets at two in the afternoon on a Tuesday.
The medium has to match the rhythm of the listener.
And that's where the "old school" objection to Facebook Groups gets interesting. Yes, Facebook has demographic decline among under-thirty-fives. Yes, organic reach has been algorithm-suppressed. But if your audience skews older, or if they're already on Facebook for other reasons, the barrier to joining your group is essentially zero. They're already there. They don't need to create a new account, learn a new interface, download a new app.
Whereas Discord is a higher-friction join. New account, new app, new mental model.
That friction matters more than people think. Every extra step between "I'm interested" and "I'm inside" costs you a chunk of your potential community. A forty percent join rate on a Facebook Group versus a fifteen percent join rate on a Discord server — that's not about which platform is better in the abstract. That's about which platform matches where your listeners already are.
How do you actually make that call? If you're a producer looking at this, what's the decision framework?
I'd start with three questions. One — where does your audience already spend time online? If you have any analytics on referral sources or if you've ever run a listener survey, that's gold. Two — what's the listening context? Commute and chores suggest async. Active, desk-based listening suggests sync might work. Three — what kind of engagement do you actually want? If you want listeners submitting corrections and episode ideas, a forum or email list handles that beautifully. If you want live watch-parties or real-time discussion during episode drops, Discord or even Telegram might be the move.
The mistake is trying to do all of them at once.
The classic mistake. "We'll have a Facebook Group and a Discord and a Substack and an email list and a forum." Congratulations, you now have five empty rooms instead of one full one. The Transistor research is really clear on this — pick one platform and commit to it for at least six months. Dilution is the silent killer of podcast communities.
Let's talk about what a well-structured community actually looks like, because I think that's where Daniel's question about Facebook being "old school" gets at something real. The platform matters less than the structure. A Facebook Group with a clear purpose — a weekly challenge, a feedback thread, a resource swap — will outperform a Discord server that's just "talk about the show.
There's a case study that illustrates this perfectly. Jack Rhysider's Darknet Diaries Discord. He built the community around episode discussion threads that post automatically when a new episode drops. There's a predictable engagement cadence. Listeners know exactly when to show up and exactly what to do when they get there. That predictability is what keeps people coming back — not the platform, not the branding, not the emoji reactions.
The counter-example is the tech podcast that launched a Facebook Group and got forty percent of listeners to join — impressive — but only three percent were active weekly. The group had no structured reason to return. No weekly thread, no prompt, no ritual. It was just a room with a sign on the door.
That's the thing — community is not a destination. It's a loop. You can't just open the doors and expect people to figure out what to do. You have to design the thing they're supposed to do, and you have to make doing it feel consequential. If a listener posts a correction and it never gets acknowledged on air, they're not going to post a second one. If they submit an episode idea and it disappears into the void, they're not going to submit another.
The platform choice is really downstream of the engagement design. You figure out what kind of participation you want, then you pick the tool that enables it.
And I think that's where we should go next — how to actually design those invitations, those friction points that convert a passive listener into someone who raises their hand and says "I'm here, I want to participate." Because getting the platform right is step one, but getting people to walk through the door is a different skill entirely.
Daniel's question about "putting the framework in place" — that's really what we're talking about. The framework isn't just the software. It's the invitation, the ritual, the feedback loop. The whole machinery of turning listeners into members.
Which sounds like a lot, but the mechanics are actually pretty straightforward once you see them. And the payoff — having an identifiable community instead of an analytics dashboard full of shadows — is the difference between broadcasting and actually building something.
I think that's the thing to name before we go any further — this phrase "hidden audience" gets thrown around a lot, but operationally it means something very specific. It's the gap between a download event and a human relationship. Your analytics dashboard tells you someone in Melbourne listened for thirty-seven minutes on a Tuesday. It doesn't tell you they're a paramedic who listens on night shifts, or that they disagree with your take on episode two hundred and twelve, or that they have an expertise in something you've never even thought to cover.
The dashboard gives you a census. It doesn't give you a community. A census tells you how many people exist in a place. A community tells you what they care about, what they'll show up for, and whether they'll bring someone else next time.
Here's where the distinction gets sharp. A download count is a lagging indicator of reach. It tells you what already happened. A community is a leading indicator of resilience. If your downloads dip — algorithm change, platform shift, whatever — a community notices and asks what's going on. A download count just goes down silently.
The "hidden" part isn't just that you can't see them. It's that the relationship is latent. It's there, it's real, but it hasn't been activated. And the framework Daniel's talking about — that's the activation layer.
And the activation layer isn't about begging for reviews or tacking "join our Facebook Group" onto the end of every episode. That's the podcast equivalent of a restaurant that puts a tip jar next to the register and calls it a loyalty program. It's passive. It's generic. It asks nothing specific of anyone.
What's the alternative?
That's the phrase I keep coming back to. You design specific moments where the listener has to make a small decision — do I send that correction, do I vote on that topic, do I respond to that question — and in making that decision, they cross a line from audience to participant. The friction isn't a bug. It's the whole point.
Which means the framework has three jobs. You need a place for them to gather, a reason for them to show up, and a loop that makes showing up feel worth doing again. Platform, invitation, retention.
The order matters. Most producers start with platform — "let's set up a Discord" — and then wonder why nobody's in it. You start with the invitation design, then pick the platform that serves it, then build the retention mechanics on top.
Let's get concrete about the platforms themselves, because Daniel asked about Facebook Groups directly and I think the honest answer is more interesting than yes or no.
Is a Facebook Group actually a good idea in twenty twenty-six?
It depends entirely on what you're asking it to do. The raw numbers are still staggering — one point eight billion monthly active users. That's not a platform in decline, that's a platform that's become infrastructure. But here's the catch. Organic reach inside Groups has been algorithm-suppressed for years. If you post something, maybe fifteen to twenty percent of your members actually see it in their feed. The rest have to actively navigate to the group.
The low barrier to entry is real, but the visibility is throttled.
And the demographic shift is real too. Under-thirty-fives are leaving Facebook in measurable numbers. If your podcast audience skews younger, you're building on land that's slowly eroding. But if your listeners are thirty-five and up, they're probably already on Facebook for other reasons — neighborhood groups, hobby groups, school parent groups. Your community isn't a new destination, it's just another tab in a place they already inhabit.
Which is the whole argument for choosing based on audience behavior rather than personal preference.
And that's straight from the Buzzsprout research — match the platform to the listening context, not to what you personally enjoy using. I know producers who love Discord because they're on Discord all day. But their audience listens while folding laundry.
What's the engagement cost of choosing Discord over a forum?
Discord's friction is higher than people admit. New account, new app, server discovery is clunky, and the interface assumes you're going to be present. Threaded conversations help, but the fundamental rhythm is real-time. If you check in once a day, you've missed seventeen conversations that already concluded. That's exhausting for a casual participant. They stop checking.
Whereas a forum — or an email list — waits for you.
It waits for you. You open it when you're ready, you see what's new since last time, you respond at your own pace. The Transistor research found exactly this — async-first platforms consistently outperform real-time chat for audiences that listen during commutes or chores. The engagement isn't as intense in the moment, but it's more durable over time.
Which brings us to Substack and dedicated forums. The ownership argument.
This is where it gets strategic. With Facebook or Discord, you don't own the relationship. You're renting space. If the algorithm changes, if the platform pivots, if they decide to deprioritize Groups or servers in favor of whatever their next thing is — your community is collateral damage. A forum you host yourself, or an email list on Substack, or even a self-hosted mailing list — you own the data, you own the relationship, you can migrate it, you can reach people directly.
The tradeoff is you have to create more content to keep it alive.
That's the real cost. A Facebook Group or Discord can run partially on member-to-member interaction. A Substack or forum requires the host to be more present — writing posts, seeding discussions, being the gravitational center. If you're already stretched thin producing the show, adding a newsletter on top can tip you into burnout.
The platform decision is really a triangle. You're trading off between low barrier to entry, ownership of the relationship, and how much additional content you need to produce.
That's a good way to frame it. And I think the case study that brings this to life is Darknet Diaries. Jack Rhysider built his Discord around automated episode discussion threads. New episode drops, a thread appears, the conversation has a natural starting point. He's not in there constantly seeding discussion — the episode itself is the seed. That's a design choice that reduces the host's ongoing labor.
The counter-example you mentioned earlier — the tech podcast with forty percent join rate and three percent weekly active users — that's the same platform, completely different design. No structured reason to return.
That's the thing about Facebook Groups specifically. When they work, they work because there's a weekly ritual. A feedback thread every Monday. A resource swap every Wednesday. A challenge every Friday. The platform provides the room, but the structure provides the reason to walk through the door. Without the structure, you've just built another empty room on the internet.
The answer to "is Facebook too old school" is — it's not about the school, it's about the curriculum.
We've mapped the platform landscape. But the harder question — and I think the one Daniel's really driving at — is how you actually get people to show up. The invitation design.
Because the default is "join our community, link in the show notes." Which is the conversational equivalent of handing someone a business card and walking away.
It fails for the same reason most calls-to-action fail. It's generic. It asks nothing specific of anyone. The listener hears it and thinks — okay, join and then what? If you can't answer that in the invitation itself, they're not going to click.
What's a good invitation look like?
The best ones are tied to the content of the episode the person just heard. Not "join our Facebook Group" — but "we talked about X today, here's a thread where you can tell us what we got wrong." Or "we covered Y, what did we miss, there's a link in the show notes where you can add your take." The invitation is specific to the moment, which makes it feel less like marketing and more like continuing the conversation.
The correction and clarification thread. That's a smart one, because it gives people permission to be critical in a structured way.
Critically, it lowers the bar for participation. You're not asking someone to introduce themselves, share their life story, be witty in a group chat. You're asking them to contribute one specific thing they already know. If they caught a factual error, they can just say that. It's the smallest possible ask that still creates real engagement.
The "what did we miss" prompt at the end of an episode works the same way. You're inviting expertise, not small talk.
And that's the Descript research finding in practice — successful podcast communities have a feedback-to-content pipeline. Listener input visibly shapes future episodes. When someone submits a correction and you acknowledge it on air two episodes later, you've just demonstrated that participation is consequential. It actually changes the thing they're listening to.
Which loops back to what you said earlier about community being a loop, not a destination. The feedback-to-content pipeline is the engine of that loop.
And the speed of the loop matters more than the depth. If a listener sends in a correction or an idea and it takes six months to get acknowledged — if it ever does — the loop is broken. They've already moved on. But if you close the loop within two episodes, you've created a direct line between "I participated" and "I heard my participation reflected back at me." That's addictive.
The invitation isn't just the ask. It's the implicit promise that the ask leads somewhere.
That brings us to the third layer — what you do once they actually show up. Because this is where most communities fail. The launch spike happens, people trickle in, and then... No structure, no ritual, no visible reason to return.
The graveyard of good intentions, as we said.
How do you prevent that? The mechanics are actually pretty straightforward. First, set up a dedicated email address — community at my weird prompts dot com, for example — with an auto-reply that includes the link to your chosen platform and a one-sentence description of what to do when you get there. That auto-reply is doing real work. It's setting expectations before the person even arrives.
They're not walking into an empty room confused. They're walking in with a mission.
Second, establish a recurring segment that explicitly draws from community input. A Listener Spotlight every five episodes where you read and respond to one detailed piece of feedback. Not just "thanks for the kind words" — actually engage with the substance of what they sent. Argue back if you disagree. Build on it if you don't.
The Memory Palace does something adjacent to this. Nate DiMeo's email list isn't a community platform in the traditional sense — it's just him sending short, personal notes alongside episode links. But the intimacy of the channel makes it feel one-to-one even though it's one-to-many. The listener feels like they're being written to, not broadcast at.
That's a design choice that works brilliantly for a certain kind of show. It doesn't scale to group discussion, but it scales beautifully to emotional connection. The question for any producer is — what kind of connection serves your show? If you want listener-to-listener interaction, you need a group platform. If you want listener-to-host intimacy, an email list might actually be better.
The Reply All example is instructive too. Their Yes Yes No segment explicitly solicited listener-submitted internet mysteries. The community knew their input could literally become the next episode. That's the feedback-to-content pipeline at its most extreme — your weird Twitter find might be the thing a million people hear about next week.
You don't need a show the size of Reply All to make that work. A shared document — Google Doc, Notion page — where listeners can propose episode topics and upvote each other's ideas. Then you reference the upvote counts on air. "We saw the suggestion about deep-sea mining got forty-seven upvotes this week — we're researching it." Suddenly the community has agency. They're not just consuming the show, they're steering it.
The upvote count being mentioned on air is the key detail there. It closes the loop publicly.
It creates a virtuous cycle. Someone hears you mention the upvote count, they realize their vote actually matters, they go vote on the next thing, they tell a friend to vote too. The engagement isn't performative because it has a visible outcome. That's the difference between a community that feels alive and one that feels like a suggestion box nobody checks.
If I'm a producer listening to this, the playbook is — pick one platform that matches my audience's listening context, design specific content-tied invitations instead of generic CTAs, and build at least one feedback-to-content loop that closes within two episodes. That's the machinery.
Commit to it for six months. Not six weeks. The first two months will feel slow. That's normal. Community building is a compounding game — the early returns look flat, and then suddenly there's enough critical mass that conversations start happening without you. But you have to stay in the room long enough for that to happen.
Let's pull this into something a producer can actually act on this week.
First — pick one platform and lock in for six months. Not Facebook and Discord and Substack simultaneously. That's how you end up with three quiet rooms and a burned-out host. Start with the platform that matches your audience's listening context. Commuters and cooks get a forum or email list. Desk listeners might get Discord.
The Buzzsprout point — match the platform to the behavior, not your preference.
The first eight weeks will feel like talking to yourself. That's not failure, that's compounding. You're building the gravitational center.
Second move — design a feedback-to-content loop that closes within two episodes. If someone sends a correction or an idea, name it on air within two episodes. Not six months. Not "we'll get to it eventually." Speed of acknowledgment matters more than depth.
That's the Descript finding in practice. Participation has to feel consequential. The listener needs to hear their contribution reflected back at them while the moment is still warm. That's what converts a one-time contributor into a repeat participant.
Third — measure the right thing. Don't obsess over total member count. Track return rate. What percentage of your members engage more than once a month? A community of two hundred active members is worth more than two thousand people who joined and never came back.
That's the metric that tells you whether your structure is actually working. If people join but don't return, your invitation worked but your retention loop is broken. If they return, you've built something self-sustaining.
Which brings us to a question I've been turning over. As AI-generated content and synthetic voices become more common — and they're already flooding the feed readers and the podcast directories — what actually differentiates a human-made show? I think the answer, increasingly, is going to be community.
Because the AI can replicate the voice, the format, even the banter. But it can't replicate the relationship between a host and a listener who's been sending in corrections for two years.
An AI can generate a podcast that sounds like us. It can't generate a listener who cares enough to upvote a topic suggestion or send a voice memo about something we got wrong. Community becomes the proof of work. The thing you can't fake.
Which means the shows that start building this now — not next year, not when the AI landscape gets even weirder — they're going to have a twelve to eighteen month head start on everyone who's still treating community as a nice-to-have.
The next frontier isn't just community platforms. It's community-owned content. Imagine a podcast where listeners vote on episode topics, submit audio clips, and co-create the narrative. That's not science fiction — the tools already exist. The question is who builds the framework first.
The trajectory is pretty clear. First you identify your audience. Then you gather them. Then you give them agency. At some point, the line between producer and listener starts to blur — and that's not a bug.
If you found this useful, send us a voice memo at community at my weird prompts dot com. We might play it on the next episode. We'd love to hear what platform you'd pick, or what's been stopping you from building your own community.
This has been My Weird Prompts. I'm Corn.
I'm Herman Poppleberry. We'll catch you next time.
