The deletion log for a recreated article often holds exactly one code: G4. An administrator compares the new page to the one a discussion already removed, judges them substantially the same, and deletes it again — sometimes within minutes of its reappearing, with no new discussion and no warning first.
Speedy deletion criterion G4 only reaches pages removed through an actual discussion: an articles-for-deletion debate, a deletion review, or another venue with a closing rationale on record. It does not cover a page that expired through proposed deletion, where nobody even objected within the seven-day window, and it does not cover a page that has genuinely changed since the version that was deleted. The test is substantial identity to what came down before, not identical wording. Reordering paragraphs, swapping a few adjectives, or tacking on a sentence of commentary does not move a page out from under its own deletion rationale.
Applies to sufficiently identical copies of a page deleted via its most recent deletion discussion. This excludes pages that are not substantially identical to the deleted version.
The mistake we see most often at this stage treats the deletion as a writing problem. A page came down because three of its four sources were press releases and the fourth was a staff biography, and nobody found independent coverage to hang notability on. The response is usually a rewrite — tighter sentences, a reorganized infobox, a warmer tone. None of that is what an administrator checks. What has to change is the evidence: a profile that did not exist when the first discussion ran, a review in a publication with its own editorial staff, coverage that treats the subject as the subject rather than mentioning it in passing. Without material like that, the second version is the first version in different clothes, and G4 reads it exactly that way.
Recreate a title often enough and administrators can protect it from creation altogether — 'salting' it, in the project's own shorthand — so that nothing below a set account threshold can start that page, deleted history or not. Lifting that protection takes more than patience. It usually takes a request at deletion review with the new sources attached, or a draft built in draft space and routed through Articles for Creation, where a second reviewer confirms the sourcing clears the bar before the title reopens.
Two paths exist for a page that came down in a discussion, and they answer different questions. Deletion review asks whether the original close got the sources that were actually presented wrong — a check on the process, not a second argument for notability on the same evidence. Draft space and Articles for Creation ask what the record shows today, including coverage that may not have existed at the time of deletion. Most legitimate recreations belong on that second path, and the draft should say plainly, in its own talk page note, exactly what is new since the deletion — because that is the first thing any reviewer will look for.
