The channel that built your audience is getting worse on purpose. You did nothing wrong. That is what channels do once they fill up.
Every marketing channel runs the same arc: it works, the crowd arrives, the returns collapse. Andrew Chen has a name for it, the “Law of Shitty Clickthroughs.” As a channel matures, response drops and the cost of reaching anyone climbs.
Look at where the big ones sit today. SEO fights Reddit threads and AI listicles for a single slot on the page. Email opens have slipped under a third, paid acquisition costs more every quarter, and an influencer spike is gone inside a week. None of that is a bad month. It is where a channel lands once enough people crowd into it.
AI makes the old channels worse, faster
The hope is that AI reverses this. It accelerates it instead.
Give every team the same tools to write, target, and publish, and the tools stop being anyone’s edge. What they add is sheer volume, poured into channels that were already full, at almost no cost. The clickthrough curve that used to take years to bend now bends in months.
A handful of AI products went viral this year on novelty alone, with no real marketing behind them. That window is closing. Once the novelty wears off, saturation comes back, and it comes back faster than before.
So the useful question is not how to squeeze more out of a decaying channel. It is the one Chen states directly: what can you do that no one else is doing with the new technology?
The move AI actually unlocks
AI does not rescue the big channels. It makes the small ones affordable for the first time.
For years the most durable audience tactic was also the least scalable. You run a focused space for a specific group of people, you show up in it, and you build a real relationship over time. Newsrooms did this with their most loyal readers. Brands did it with their best customers. It worked, and almost nobody could staff it at scale, because curation, moderation, and daily attention ate too much human time.
That math has changed. A lean team can now curate a feed, draft the daily update, surface the conversation worth having, and keep a niche space alive without a twenty-person desk behind it. The exact work that once made owned communities unscalable is the work AI handles well. People keep control of judgment and voice. Agents carry the volume.
The tactic that was too manual to scale is now within reach of a small team.
Why niches, and why now
The case for owned niche communities is practical, and it starts with what social platforms have turned into.
The reach you built on someone else’s platform was always rented. Now the rent is rising and the product is degrading at the same time. Feeds fill with ads and recommended strangers. Organic reach for the accounts people deliberately chose to follow keeps sinking. The platform optimises for its own revenue, not for your relationship with your audience, and people can feel the difference. They trust the shared feed less every year.
That decline is not just your problem. It is your opening. When the open feed gets noisier, a smaller space with a clear purpose gets more valuable, not less. People are not asking for more content. They want somewhere that makes sense to them, where the same voices show up and yesterday’s conversation still means something today.
It is also a position a competitor cannot copy. Your ad creative can be cloned overnight. A community that trusts you and comes back on its own cannot.
What this looks like in practice
Owning the relationship means owning the place it happens. For most teams that is a branded space of their own: a mobile and web home where a specific audience finds curated content, talks to each other, and returns without an algorithm dragging them back.
The metric moves with it. Reach is cheap. Return is everything. A rented channel counts impressions. An owned community measures whether people come back on their own, recognise each other, and treat the space as theirs.
None of this survives a thin product. This is Chen’s caveat, and it holds: marketing amplifies something good, it cannot save something weak. An owned community with no reason to return is one more empty channel that happens to carry your logo. The content, the rhythm, and the reason to belong all have to be real.
The takeaway
Stop fighting for a channel that was built to decay. The routes everyone uses will keep getting more expensive and less effective, and AI pushes that along rather than pulling it back.
The opening is on the other side of that. Tactics that were once too small and too manual to run at scale are now within reach of a lean team with the right tools. Owned communities in specific niches were always the strongest form of audience you could have. Now they are a realistic one as well.
Belonging does not happen by accident. It has to be designed. The teams that start building their own space now will own the relationship while everyone else keeps bidding up the price of rented reach.
