Word Count Is Not a Link Magnet

Nobody has ever added a link because a page was long. People link because a page settled something they needed settled. Length is at best a side effect of doing that thoroughly — and at worst it’s the thing that stopped the reader finding the answer, which is the opposite of what you were trying to buy.

Where the belief came from

The length habit rests on an old and endlessly repeated observation: pages with more words tend to have more links pointing at them. You’ll see it asserted as though it were a mechanism. It isn’t; it’s a correlation, and there’s an obvious third factor sitting behind both halves of it.

Thorough treatment of a subject that’s worth covering produces two things at once: a lot of words, and a page other people want to cite. The words didn’t cause the links. The subject and the thoroughness caused both. Run the counterfactual and it falls apart immediately — the page nobody would have cited at eight hundred words does not become citable at four thousand. It becomes a longer page nobody cites.

Distrust the framing whenever you see it. And note that this site isn’t going to hand you a competing figure: search engines publish almost nothing about how they weight links, correlation studies of this kind are cheap to produce and hard to interpret, and we don’t publish invented numbers in either direction. The honest position is that the evidence for length-as-cause is weak and the mechanism has never been described by anyone in a position to know.

What “beat the top result by being more comprehensive” optimises for

The advice sounds harmless. In practice it’s an instruction to write with one eye on a competitor’s outline, and it reliably produces three things:

Sections nobody asked for. You add a “history of” section, a glossary, a “common mistakes” list, because the pages you’re measuring yourself against have them. None of it came from a question a reader had. It came from a spreadsheet of headings.

Restatements of your competitors’ points. If your process is “cover everything they cover, plus more”, most of your page is a paraphrase of theirs. That’s not a page anyone has a reason to cite — the original already exists, and citing the original is what a careful writer does.

Your own claim buried. The thing you actually know, the sentence that only you could have written, ends up in the second half, under a wall of context. And a claim nobody reaches is a claim nobody quotes.

That last point is the fatal one, because citation requires quotability. Somebody has to be able to lift a sentence from your page, put it in their paragraph, and have it stand up on its own. Padding makes pages hard to quote: the assertions get hedged into mush, the good line gets split across three qualifying paragraphs, and there’s no single place to point at. The whole craft of writing claims people quote is the opposite of the comprehensiveness reflex.

The specific costs of padding

Length isn’t neutral. Each extra thousand words you didn’t need buys you a specific problem:

• The answer moves down the page. A reader who has to scroll to find out whether you’re going to answer their question often doesn’t. Neither does the person deciding whether to cite you, and they’re making that decision in seconds.
• The page stops being about one thing. A page covering six loosely related subjects can’t be linked precisely — someone who wants to reference your point about the third subject has nowhere exact to send people. This is most of why a narrow page earns links that a sprawling one doesn’t.
• It becomes expensive to keep accurate. Every fact on a page is a maintenance liability. A long page has more of them, so it goes stale faster and gets updated less, because nobody wants to re-read four thousand words to check whether any of it has changed. Then it rots — while still collecting links, which is worse than not collecting them.
• It hides your unique contribution. If a reader can’t tell within a screen what you have that nobody else does, they’ll assume you have nothing.

What earns the link at any length

Set length aside and look at what actually causes somebody to add a URL to their page. In practice it’s a short list:

• Resolving a disagreement. Two camps argue about something; you demonstrate which is right, or show that the question was badly posed. Everyone writing about the argument afterwards needs to point at your resolution.
• Showing the work. The whole calculation, the full method, the reasoning behind a judgement — so nobody has to redo it.
• Having the thing nobody else has. Your own measurement, your own diagram, your own artefact.
• Being right about an edge case. The situation where the standard advice fails. This gets cited hard, because it’s the only page that addresses it.

Every one of those can be done in a few hundred words. Some of the most-linked pages on the web are short — a definition, a single table, a one-screen explanation of a thing that used to be confusing. They earn links because they’re complete for their question, which is a different property from being long.

A practical test

Take a page you’re proud of the size of, and go through it section by section with one question: could I imagine somebody linking to this section on its own?

If yes, keep it, and then ask a follow-up — should it be its own page? Sections that are independently citable usually are, and splitting them gives each one a URL people can point at instead of forcing every citation through one crowded address.

If no, cut it. Not shorten it: cut it. A section nobody would cite is a section nobody wanted, and it’s making the rest of the page harder to use.

The honest version of “comprehensive”

Comprehensive is a real virtue when it means complete on a narrow question: nothing missing that a reader of that specific question needs, including the awkward parts. It’s a vice when it means “covers more ground than the competitor”, because ground isn’t what gets cited.

So write to finish the question, then stop. If new material arrives that doesn’t belong to that question, it’s the next page, not more of this one. That’s the discipline underneath every asset that earns links without being asked for — and it’s cheaper than padding, which is the rare case where the honest route is also the lazy one.