You can't start a community

Hey, I’m Isaac 👋 I founded Pistachio, a growth agency working with B2B brands like Atono and Clay to build trust, relationships and loyalty with their current and future customers.

If you want to chat about your own growth strategy, book a free call with me here.

Read time: 4 mins 26 secs

Interested in sponsoring these emails? See our partnership options here.

In July 2006, Walmart launched a social network for teenagers.

It was called The Hub, and it existed because Target was winning teenagers and Walmart wanted them back. Walmart had every advantage available at the time, including more than 100 million weekly shoppers, and a budget with no meaningful ceiling.

They shut the whole project down about ten weeks later.

Walmart screened every post before it went live, and members couldn't contact each other. It was a teen hangout where every conversation went through corporate head office.

Even if they fixed all of that, The Hub still would've died. Walmart had customers, but a community needs an audience.

Walmart had the right instinct in 2006. Today it's a consensus.

In beehiiv's State of Newsletters 2026, Austin Rief, co-founder of Morning Brew, said "Community will be for the next 10 years what content and newsletters were for the last 10".

He's describing a continuation. Community inherits the job newsletters have been doing, which is a direct relationship with an audience you own. Same job, one layer deeper.

The instinct is right. The sequence is what killed it for Walmart.

"Community is the next differentiator" gets interpreted as "go and launch a community". The mandate arrives with a start date and a member target. Somebody picks Slack or Discord, sends the invites, and watches four hundred people post three times a month.

A community is a phase of audience building. You notice one and you house it.

So why do these launches fail so reliably?

Andrew Chen, a general partner at Andreessen Horowitz, ran rider growth at Uber and wrote The Cold Start Problem, on why new networks die. At inception, he found, network effects run backwards. New users leave because not enough other users are there yet. He calls it anti-network effects, and the emptiness itself does the damage.

Member one arrives in an empty room and gets zero value. Member two arrives, watches member one get nothing, and leaves.

If that sounds like a 2006 problem, look at Google.

In October 2018, Google had 1.5 billion Gmail users. That same month, Google admitted 90% of sessions on their Google+ network lasted less than five seconds. All that distribution and nobody saw value in staying in the room.

Marketers already accept this logic. Ask anyone why a two-sided marketplace is hard to start and they'll explain the chicken-and-egg problem without pausing. But almost no one applies it to the context of a community.

Chen's answer to the cold start problem is the atomic network, "the smallest possible network that is stable and can grow on its own".

Facebook's was famously seven friends in ten days. Slack's was ten people inside a single company. You need the smallest group who care about the same thing, and you can find them among people who are already engaging with your content.

Dr Jen Ashton is the clearest example I've seen recently. She was Chief Medical Correspondent at ABC News. She started a free wellness newsletter on beehiiv and spent six months publishing it. It was just her own health journey, documented weekly, to 181,000 subscribers.

Then she opened a community. Five thousand people joined the first cohort, and she hit $750,000 in 90 days with zero ad spend.

All because she'd spent six months earning that audience before she asked for anything.

The Nerve is a UK investigative publication, also on beehiiv, who hit 1,000 paying members in the first week, converting roughly 12% of their audience. beehiiv's report explains the result as being "driven largely by Co-Founder Carole Cadwalladr's established online audience".

The same thing shows up outside media publishing. In 2015 there was a study in Marketing Science on a multi-channel retailer that launched a customer community. Roughly 19% of post-launch revenue from community members was attributable to their having joined. It was an existing customer base, converting into a new offer.

The audience came first, and it always does.

That's why newsletters matter more than the community conversation suggests. A newsletter muscles its way into an already-established habit space, the inbox, sitting between messages from your boss and your mum. After a few weeks of consistency the mere exposure effect kicks in. That's the highest-trust owned audience most brands will ever build.

When Ashton opened her community, beehiiv had no way for her to build one of her own. Her newsletter was on the platform and the room had to live somewhere else. Last week, that changed, and beehiiv launched Community directly inside the platform. Existing subscriptions become access, and everything you publish syncs in automatically.

I've argued that AI commodified marketing execution, and that being distinctively human is the only brand advantage left. Community is where that stops being a slogan. AI can write a passable newsletter, but it can't manufacture the fact that two of your members know each other.

Those connections are the part you don't own yet. Every week I hit send, and people write back. Those replies are hands-down my favourite part of doing this. But every one of those replies comes to me and nobody else. It's a private channel, invisible to every other reader.

I have never once introduced two Brand Chemistry readers to each other. The format gives me no way to do it. Every relationship runs from me to you, then sometimes from you back to me. It's one to many, or one to one.

Meanwhile, those reader-to-reader relationships form anyway. They just form somewhere else, in LinkedIn comments and other people's Slack groups. That fragmentation is the gap a community closes. beehiiv found the same thing in their paid-newsletter research. A newsletter is content, a community is belonging, and bundling the two builds a switching cost that content alone never will. Churning a subscription is easy. Cancelling somewhere you belong feels like leaving. Or as Ashton puts it, "information is everywhere, but belonging somewhere is rare".

So how do you know your audience is ready? Your readers could get value from each other. They're working on similar things. They could answer each other's questions better than you could. There's demand for more experiences rather than more content.

The strongest signal is the simplest. People are already talking to each other in your comments. That's the community, already happening. It just needs a home.

Some communities do start cold. Early Reddit did. So did a thousand Discord servers. But a brand has no reason to be the facilitator unless people already gather around its ideas.

Walmart had 100 million customers and couldn't fill their room.

Earn the audience first, and the room won't stay empty.

P.S. beehiiv just launched Community, which lets you build the room inside the same platform your audience already lives in. Start your community on beehiiv and use code ISAAC30 for 30% off your first three months.

If you enjoyed this post or know someone who may find it useful, please share it with them and encourage them to subscribe: brandchemistry.co/p/beehiiv-community