Guides That Cite Pages You Can Open

Occupation landing pages are easy to write badly. The tempting version restates the occupation in slightly different words, adds a paragraph of encouragement, and links nowhere useful. It reads fine and it helps nobody, and within six months it disagrees with the real profile because the two are maintained separately.

The guides here take the opposite approach. They are short, and every substantive line points at a page that already exists on this site: the live occupation profile, the aptitude suites that profile already lists, and the skill lessons behind those constructs.

A guide that copies the profile has to be updated when the profile changes, and nothing fails when it is not. A guide that cites the profile is correct by construction: if the duties change, the guide points at the changed duties.

This is the same discipline the rest of the product applies to sources. The reason it shows up here is that a landing page is exactly the kind of content that quietly rots, because nobody re-reads it after it starts ranking.

It has a second effect that was not the original motive and turns out to matter more. Because a guide cannot say anything the underlying pages do not support, writing one surfaces gaps: an occupation with no aptitude suites mapped to it, or a construct with a thin lesson behind it, becomes obvious the moment you try to cite it. Several improvements to the career material came out of trying to write a guide and finding there was nothing to point at.

• A plain description of the work, drawn from the published profile.
• The aptitude suites mapped to it, with a direct route into practice.
• The skill lessons behind those suites, for the parts you want to learn first.
• A link into the occupation catalogue, if the guide turns out to be the wrong one.

Use a guide when you already have an occupation in mind and want to know what preparing for it involves. That is a genuinely different question from the one the matcher answers, and sending somebody with a specific job title through a sixty-prompt interest questionnaire is a poor trade of their time.

Use the matcher when you do not have a title in mind, and want the exploration to start from what you enjoy rather than from a job advert you read last week.

A guide only exists where there is a published profile behind it and real practice material to point at. Generating one for every occupation in the catalogue would take an afternoon and would produce pages that cite nothing, which is the failure mode this design exists to avoid.

Publishing fewer pages that each earn their links is the slower route and the only one that does not accumulate a maintenance debt nobody ever pays.

It is also the version that is useful to a reader arriving from a search engine. A page that exists because somebody might type its title, and which cites nothing they can open, is a landing page in the pejorative sense: it captures a visit and returns nothing for it. The test of a guide is whether the person who lands on it leaves with a page worth reading, and that is a much harder thing to fake than a ranking.

A citing page has one failure mode a copying page does not: the cited page can disappear. A guide pointing at a withdrawn occupation profile is a dead link in the most prominent place on the page, which is worse than a stale paragraph because it is visibly broken.

So the links are resolved against the registries rather than trusted. Every internal link on a guide has to resolve to a record that exists, and the check runs over the whole set rather than over a sample. That is the same discipline applied to the help centre, the tutorials and the blog, and it is the reason this expansion could add content confidently rather than carefully.

The blog explains how things work and why boundaries sit where they do. The help centre documents behaviour. The guides are neither: they are entry points for somebody arriving from outside with a specific question, and their job is to hand that person to the right existing page as quickly as possible.