Notability decides whether a subject gets an article. It says nothing about what belongs inside one — and conflating the two produces pages that survive their first deletion discussion, then get gutted over the months that follow.
Wikipedia's editors have a word for the result: a coatrack. The subject is real and the article's opening paragraph is sourced well enough to justify a page. What follows, though, hangs off that opening paragraph the way a coat hangs off a hook — a list of partnerships, a run of minor industry awards, a paragraph on a sustainability pledge nobody outside the company ever reported on, a client roster copied from the company's own site. None of that material would earn the subject an article by itself. Once independent sourcing has done its work at the top of the page, it becomes free real estate for everything that couldn't.
An article that is nominally about one subject, but is written to carry a different one — typically promotional, biased, or simply unrelated to what makes the nominal subject notable.
Two articles about companies in the same industry, both written the same year, show how differently the underlying cases can be. One carried two paragraphs a trade reporter would recognize — a founding date, a funding round, both covered independently — buried under six more about company culture, a wellness initiative, and executive bios sourced to the company's own newsroom. An editor tagged it, cut the six paragraphs in an afternoon, and what remained was short, accurate, and never nominated for anything. The sourced material had never needed the padding.
The other article lost the same categories of padding and had almost nothing left underneath — one funding announcement, rewritten by four outlets in a single week, and a hiring notice. Trimming it didn't produce a shorter article. It exposed a page that had never really been about anything independent sources had chosen to cover, and a deletion nomination followed inside a month.
The difference between those two outcomes was decided before either editor picked up scissors. Proportion is diagnostic: an article that shrinks to a stable stub once the unsourced material is removed was carrying a genuine case all along. One that collapses into nothing was never carrying a case — it was wearing one, briefly, at the top of the page.
We draft the other way around from how padding accumulates. Coverage gets found first, and the length of the finished article is whatever that coverage can responsibly support — not whatever a client can supply from their own archives. A page that stays three paragraphs because three paragraphs is what independent sources actually cover holds up better, in every discussion that follows, than one padded to look substantial and built to be caught.