Why Only the Community Manager Ends Up Posting

A Discord Where Only the Manager Posts — A Familiar Sight

“Should we start our own community too?” I hear this question constantly from startup founders and marketers. The vision is seductive: open a Discord channel, spin up a Slack workspace, and customers will gather to talk about the product, help each other out, maybe even contribute to the development roadmap. Honestly, it’s a genuinely appealing picture.

But among the OpenAI, Google, and Microsoft-related communities I’ve looked at, some had gone days without a single new post, questions left unanswered. Meanwhile, conversations on user-run communities like Reddit were often far more active. Slapping a big company’s name on a space doesn’t make conversation happen on its own.

Among the communities I’ve observed firsthand, a lot of them reached a point, just months after launch, where the community manager was the only one posting. They’d created a reason for people to join — but not a reason to keep coming back.

Why the Virtuous Cycle Companies Expect Rarely Materializes

Companies post a product update and expect customers to react. A question goes up, and they expect other customers to answer it. Ideally, people would also share tips and give feedback on the product roadmap.

But getting people to join and getting them to participate are two different things.

Jakob Nielsen’s “90-9-1 rule,” described back in 2006, addresses exactly this gap. The vast majority only read, a smaller group participates occasionally, and a tiny fraction accounts for most of the activity — expressed as 90%, 9%, and 1%. It doesn’t mean only 1% are allowed to post, nor is it a figure about how likely a community is to succeed. Nielsen’s original piece

Other distributions exist too. Higher Logic, a community software vendor, analyzed its own customer sites in 2020 and found that in communities with fewer than 5,000 members, 33% of users created or replied to posts. That figure describes this particular vendor’s customer sample — participation patterns clearly shift depending on a community’s purpose and composition. Higher Logic’s analysis

Why Communities Go Quiet Within a Few Months

People join communities because they need something. You join a fishing forum to get better at fishing; you join a travel cafe to get trip advice or find travel companions.

Brand communities need that same kind of reason to participate. If what customers expected was problem-solving or genuine exchange, but all that gets posted is sales announcements, the motivation to come back withers.

Operational format matters too. In spaces like Slack or Discord, where recent activity is right there in front of you, dead air is glaringly obvious. If the last message was posted a month ago, a newcomer might hesitate to post a question, unsure whether anyone will answer.

Picture a pollack-stew restaurant in Sokcho with a 30-minute wait, right next to an equally decent restaurant that’s sitting empty. The mere sight of a crowd shapes people’s choices. I think the same logic applies to communities: showing that conversations and answers are actually flowing helps draw in new participants.

If all that accumulates is a stream of announcements from a single manager, it’s worth revisiting what customers actually expected when they joined. If the goal really is just broadcasting announcements, that’s fine — but if what you wanted was conversation among customers, then your operating goals and actual activity have drifted apart.

Patterns Common to Communities That Actually Work

In the cases I’ve observed, three things mattered: consistent activity, something gained from participating, and a format the team could sustainably operate.

A good example of consistently giving people a reason to show up is Twelve Labs’ Multimodal Weekly. It’s a recurring online meetup hosted by a video AI company. Rather than just pitching their own product, they’ve invited outside researchers and developers from other companies to discuss multimodal AI research and use cases. Episode 40, for instance, featured Blackbird’s video editing use case. Even someone with no immediate plans to buy the product has a reason to show up and learn. I think highly of this format — a fixed topic, a fixed cadence, meeting again and again. Twelve Labs’ announcement for episode 40

The reward for participating doesn’t have to be money or merchandise. In streamer fan communities, having your story or your joke read and reacted to by a streamer you admire can itself be the reward. Chimchakman’s community is a good example of this kind of structure — what matters is that participation gets a response.

You should choose your format to match the cadence your team can actually sustain. For an early-stage community without much conversation yet, I think topic-based forums often beat real-time chat. On a forum, past questions and answers remain as a searchable archive — even if no new posts appear for a while, people can still find value in what’s already there. A newsletter-style approach, sending updates on a fixed schedule, is another way to sustain the relationship. You don’t necessarily need to open a live chatroom where every lull in activity is instantly visible.

Oswarld’s Lens

Honestly, I’ve seen far too many teams fail after assuming that simply building a community would be enough.

Working with startups on go-to-market strategy, I hear “let’s build a community” all the time. And I’ve watched the same failure pattern repeat itself over and over. They open a Discord, create a bunch of channels. They use gift vouchers to pull in sign-ups, but a few months in, engagement dries up. And then the community manager takes the blame.

What really gets to me is the position that community manager ends up in. If it works, that’s just them doing their job. If it doesn’t, it’s easy to pin the failure on them personally. Handing one person a performance target without ever defining the purpose of participation or the support they’d need makes it awfully hard for them to build a solid career.

So instead of signing an expensive platform contract upfront and buying 100s of sign-ups through ads, I recommend meeting 10 people offline first. Find out what would actually make them want to meet again — and only once that engagement is real, expand to 20, then 100.

Collecting contact info with a gift voucher and getting someone to stay actively engaged are two entirely different jobs. Leaders need to understand that difference before marketers and community managers on the ground can build a strategy that actually works.

Decide This Before You Start

Figure out first who’s coming for what kind of help, who will respond when someone participates, and what cadence your operating team can realistically sustain. Growing member count before you’ve answered those questions is a good way to repeat the failures I’ve described. Building a reason to join and a reason to come back next week — that has to come first.

References & Further Reading

• Jakob Nielsen, The 90-9-1 Rule for Participation Inequality in Online Communities, October 8, 2006.
• Higher Logic, Analysis of Engagement Data from Its Own Customer Communities, June 29, 2020.
• Twelve Labs, Announcement for Multimodal Weekly 40.

The author is Oswarld (Kwangseob Ahn). Current roles: Adjunct Professor at Sejong University, Strategy Consultant at INLEVEL9. Career, research, books, and recent work are kept current on the About page. Latest · July 2026: HEMA-2: A Consolidation-Aware Tri-Memory Architecture with Multi-Channel Scheduling for Lifelong Conversational AI.