Why Thin Content Fails Both Readers and Rankings

The word “thin” causes a lot of confusion, because people hear it as a synonym for “short,” and it isn’t. A two-sentence page that tells you a store’s hours and address is not thin. It answers the question completely. A two-thousand-word page that circles its topic without ever landing — definitions of things you already knew, restatements of the question, generic advice that would fit any industry — is thin at any length. Thinness is not a word count. It is the amount of information a reader has after the page, minus the amount they had before. On a thin page, that difference is approximately zero.

It’s worth asking why thin content exists at all, because nobody sets out to write it. It comes from a particular theory of SEO: that search engines reward pages, so more pages means more rewards. Under this theory, content is a quota. The plumbing company generates a page for every town in the county, identical except for the town name. The software company publishes a weekly post because the calendar says to, whether or not anyone had anything to say. The consultant asks a language model for eight hundred words on a topic and posts whatever comes back. In every case the page exists to exist. Its audience is a crawler. And you can feel this within seconds of landing on one, which is precisely the problem.

The theory behind the quota was once roughly true. There was an era when a page that merely mentioned a phrase could rank for it, and mass-producing mentions was a viable strategy. That era ended for a simple reason: search engines can watch what happens after the click. When searchers consistently land on a page and bounce straight back to the results, that behavior is data, and no amount of on-page optimization outweighs it. The machine doesn’t need taste to detect emptiness. It just needs to notice that humans, who do have taste, keep leaving. Modern ranking systems also evaluate sites as wholes, which produces the effect site owners find most unfair: fifty hollow pages don’t merely fail to rank, they drag down the ten good ones, because the site’s overall record is what earns trust. Thin content is not neutral filler. It is dilution.

I got to see this pattern at scale while building GazeSite. When you scan enough small-business sites, you find that the ones struggling hardest in search are rarely the ones that wrote too little. They are the ones that wrote too much of nothing — blogs stuffed with interchangeable posts, service pages that describe no actual service, location pages stamped out by template. The owners usually know, too. When I ask which pages they’re proud of, they name two or three. The other forty were homework. Nobody defends them; they just never got around to deleting them. A site is the sum of its pages, and most sites are mostly pages their own authors wouldn’t vouch for.

The way out is to invert the quota. Instead of asking “how many pages can we produce,” ask “what do we know that isn’t already on the internet.” Every real business knows things. The plumber knows which pipe failures actually happen in local houses and what they cost to fix. The consultant knows the mistake every client makes in month one. This knowledge is specific, earned, and — crucially — unavailable elsewhere, which means a page built on it adds information to the web rather than rearranging it. Such pages are harder to write, which is exactly why they work. Anything easy to produce is abundant, and search is an economy of scarcity: it exists to surface what’s rare. The generic post has ten thousand competitors before it’s published. The specific one may have none.

This suggests an uncomfortable piece of maintenance: deletion. If a page adds nothing — no traffic, no reader value, no purpose you can articulate — the best edit is usually removal, or merging several weak pages into one strong one. People resist this viscerally. The pages represent hours of work, and deleting them feels like admitting the hours were wasted. But the hours are gone either way; the only question is whether their output goes on quietly costing you. A hundred-page site with twenty excellent pages is weaker than the twenty-page site alone, both to the search engine averaging your quality and to the human visitor sampling it. Pruning is not destroying value. It is raising the average, and the average is what everyone judges.

The deepest problem with thin content, though, isn’t algorithmic. Every page is an argument for your competence. A visitor who lands on a hollow page learns something true: this organization publishes things it doesn’t mean. That lesson transfers. If the blog is filler, maybe the service is too. The reverse also transfers — a single page that teaches something real earns a trust that spills onto everything else you say. So the rule I’d offer is the one that would have prevented the whole phenomenon: never publish a page you wouldn’t send, personally, to one specific person with the problem it claims to solve. If you’d hesitate to send it, the search engine is right to hesitate to rank it. Readers and rankings fail thin content for the same reason. It failed them first.