Good Ideas, Empty Rooms: The Quiet Collapse of Content Discovery

There's a specific kind of frustration that's become weirdly common among people who spend real time online. You make something careful. Something that took thought, research, maybe weeks of actual effort. You put it out there. And then... a guy who filmed himself reacting to a thumbnail gets 400,000 views in a day.

This isn't a rant about fairness. Fairness isn't really the internet's thing. But what's happening right now goes beyond the usual "algorithm is broken" complaint. Something more structural is going on, and if you're a creator, a reader, or just someone who still believes the web can surface things worth finding — you should probably understand it.

The Feed Is Full. The Signal Isn't.

Here's the basic problem: the platforms that control distribution were built to maximize engagement, not quality. That's not a secret. But the side effect nobody talks about enough is that as more content floods in, the optimization pressure intensifies. Algorithms don't get better at finding good stuff when there's more stuff — they get better at finding engaging stuff, which is a very different category.

Engagement, in platform terms, means clicks, watch time, saves, shares, comments. These metrics can absolutely correlate with quality. But they also correlate heavily with familiarity, emotional provocation, and format legibility. A video that follows a proven structure — hook in three seconds, conflict by ten, resolution at the end — will almost always outperform a video that's doing something genuinely new, even if the new thing is better.

So the system isn't broken. It's working exactly as designed. The problem is that what it's designed to do no longer matches what most thoughtful creators are trying to accomplish.

Optimization as a Tax on Originality

Think of it this way: every platform has an implicit tax on originality. The more your content diverges from established patterns, the more distribution you lose. The more you conform, the more reach you gain — but the more you sound like everything else.

For a while, creators could split the difference. You could be a little weird and still get found. But as the volume of optimized content has exploded — partly because the playbooks became public, partly because AI tools made template-following nearly frictionless — the middle ground has collapsed. The cost of not optimizing has gone up dramatically.

What this creates is a kind of attention recession. Not a shortage of attention overall — people are still spending enormous amounts of time online — but a recession in the distribution of that attention to anything that doesn't fit the mold. The rich (in engagement signals) get richer. The original gets quieter.

Why Mediocre-but-Optimized Beats Thoughtful-but-Unfamiliar

This isn't really about mediocrity winning. It's about legibility. Platforms — and to some extent audiences trained by platforms — have learned to recognize certain shapes of content and respond to them automatically. A listicle with a number in the headline. A video with a talking head and B-roll cutaways. A tweet thread that opens with a counterintuitive claim.

These formats work because they're familiar. Familiarity reduces friction. Reduced friction means more people engage before they bounce. More engagement means the algorithm pushes it further. It's a self-reinforcing loop, and the entry point is format recognition, not actual value.

Meanwhile, content that's doing something genuinely different — structurally, tonally, conceptually — requires the audience to do a small amount of interpretive work upfront. In a low-friction environment, that tiny bit of extra effort is enough to tank your early engagement numbers. And if your early numbers tank, the algorithm decides you're not worth distributing. The content dies before it finds its audience.

The Discovery Problem Is Getting Harder to Route Around

For years, the standard advice was: build community, go niche, find your people. And that still works, to a point. But even niche communities now exist inside algorithmic platforms, and those platforms have gotten much better at controlling what circulates even within them.

Subreddits, Discord servers, Substacks, TikTok For You pages — they all have their own internal distribution logic now, and that logic increasingly favors the same things the big feeds do. Even spaces that were built as refuges from mainstream optimization are starting to feel the pressure.

The old routing-around strategies — word of mouth, link-sharing, curated newsletters — still exist, but they're operating at a fraction of the scale they used to. Not because people stopped caring about recommendations, but because the sheer volume of content makes it harder for any single signal to cut through, even among people who are actively looking.

What Thoughtful Creators Are Actually Doing About It

The honest answer is that most of them are making trade-offs they don't love. Some are learning the optimization game well enough to get distribution, then using that reach to do the work they actually care about. Some are abandoning reach entirely and building small, direct relationships with audiences — email lists, paid communities, one-on-one stuff — where the algorithm isn't the gatekeeper.

Some are just making peace with smaller numbers. Which sounds like giving up, but actually isn't. If your work finds the right 500 people instead of the wrong 50,000, that's not a failure. It's a different kind of success that the current metrics infrastructure just isn't built to recognize.

And a few are doing something more interesting: they're treating the attention recession as a filter. If the algorithm won't surface your work, then the people who find it anyway are self-selecting for exactly the kind of audience you want. The friction becomes a feature.

The Bigger Picture

None of this is sustainable at the cultural level, though. When the systems we rely on for discovery consistently favor the familiar over the original, we don't just get worse feeds — we get a narrower culture. The ideas that don't fit the format don't just get less distribution. They get made less often, because people eventually stop making things that disappear.

That's the real cost of the attention recession. Not that your great video didn't go viral. But that over time, the pressure to optimize reshapes what gets made in the first place. And the stuff that might have changed how you think about something — the weird essay, the experimental format, the slow-burn argument — never gets made at all.

The feeds stay full. The rooms stay empty. And most people don't notice, because they never knew what they were missing.