Who Actually Links to Things, and What They Need

Almost every editorial link on the web — a link somebody added because they wanted to, not because they were paid or asked — is placed by one of a small number of people doing a job. A reporter filing a story. A blogger mid-argument. Someone answering a question in a forum. A technical writer documenting an integration.

Each of those people needs a genuinely different kind of page, and each has a different disqualifier. If you don’t know which of them you’re publishing for, you’re guessing. Here’s the taxonomy.

Reporters and trade journalists

They need facts they can attribute, and a human to credit. Reporting is largely the assembly of sourced claims, so a reporter’s link is a citation in the strict sense: it exists so a reader can check the claim. What they want from you is a number, a finding, a policy, or an expert opinion, tied to a named person or organisation, with enough method behind it that quoting it isn’t a risk.

What disqualifies you: unsourced assertions, no publication date, no author byline, and numbers that can’t be traced to how they were produced. A trade journalist working to deadline will drop a source that doesn’t say where its figures came from, because their editor will ask.

What they will never do: link your product page. A reporter links the evidence, not the vendor. If your only citable asset is a page describing what you sell, you are not in this game.

They need something to hand off mid-argument. They’re making a point, they reach a sub-point they don’t want to spend three paragraphs on, and they link out so they can keep moving. The link is a shortcut for them and a courtesy to the reader.

That means they want one page about one idea, with a title that names the idea. The best possible target is a page whose headline is the exact thing they just asserted. Long omnibus guides get linked far less often than their quality suggests, because pointing at a 6,000-word guide doesn’t discharge the handoff — the reader still has to go hunting.

What they will never do: read your whole page. They’ll scan it long enough to be confident it says what they think it says. If your first screen doesn’t confirm the claim, they move on.

People answering questions in forums, comments and communities

They need a page that answers the literal question, fast. This is the most underrated source of links on the web. Enormous numbers of links are placed by ordinary people helping a stranger — in a community thread, a comment, a chat channel, a subreddit-style board, a Q&A site. They are not marketers. They have no agenda. They just want to stop typing and point at something that finishes the explanation.

What they want: the answer visible immediately, in plain language, at a URL they can paste. No signup wall, no email gate, no interstitial, no PDF download, nothing that makes them look bad for recommending it. A page that answers one question exactly is worth more here than a definitive treatise.

What they will never do: read 3,000 words to find the answer, or recommend anything that asks the person they’re helping for an email address. Both feel like a bait-and-switch on someone they were trying to help.

Documentation and support writers

They need stable, precise, boring accuracy. Technical writers — including ones inside other companies, writing internal docs you’ll never see — link out to explain a concept, a format, a standard, or another product’s behaviour that isn’t theirs to document.

Their priorities are unusual: they care less about readability and more about being correct and unchanging. A page that will still be accurate in two years, at the same address, beats a livelier page that might not be. Version notes, change logs and “last updated” dates help. So does saying plainly what your page does not cover.

What they will never do: link something that reads as marketing, or something they suspect will move. A docs link that rots generates a support ticket, so they’re conservative by habit.

Teachers, course authors and curriculum builders

They need explanations pitched at a defined level, plus reusable pieces. Someone building a lesson, a training module, or a syllabus needs material they can assign. They’re matching to an audience: beginners, practitioners, or specialists — and a page that doesn’t declare its level is hard to place.

What earns their link: a clear statement of assumed knowledge, a diagram they can show on a slide, a worked example they can walk through, an exercise with an answer. Explicit permission to reuse a figure with credit removes their main hesitation.

What they will never do: assign something whose reading level they can’t predict, or use a diagram whose licensing is unclear.

Curators of resource lists and directories

They need to be able to describe you in one line. A curator is deciding whether you belong on a page of recommendations, and their real constraint is the annotation they have to write next to you. If summarising your page takes a sentence with two clauses and a caveat, you’re a hard fit.

What helps: a page that is unambiguously one thing, with a name that describes it. What hurts: a page that is sort of a guide, sort of a tool, and sort of a product tour. There’s more on this in landing on resource lists.

What they will never do: include something that duplicates an entry they already have. Being the second-best example of an existing category is worse than being the only example of a smaller one.

Wiki-style and reference editors

They need a primary source, not a summary. Reference editors work under sourcing rules — formal or cultural — that push them toward origins. If your page restates what somebody else established, an editor will cite the somebody else. If your page is where a fact, dataset, definition or measurement came from, you’re eligible.

What they will never do: cite a page that summarises another source when the other source is one click away. That’s not pickiness; it’s the whole point of their job. See primary sources get cited for what being primary actually requires.

What to do with this

Don’t write for all seven. They want incompatible things — the forum answerer wants brevity, the docs writer wants exhaustive precision, the reporter wants a named number, the teacher wants a diagram. Trying to satisfy everyone produces a page that satisfies nobody, which is one of the ways a page ends up good and uncited.

Pick one or two who plausibly write about your topic, and publish for them specifically. Then check your candidate assets against the four conditions a page has to meet before you build anything expensive.