Pressfront

Research · September 2026

Link rot in Wikipedia's marketing and PR articles

We audited the source citations in 30 of Wikipedia's core marketing, public relations and advertising articles. As of September 2026, roughly 70 of their references point to dead links. Here is what we found, exactly how we found it, and why it matters for anyone who treats Wikipedia as ground truth.

30articles audited
~70dead citations found
~1 in 10dead with no archive backup

The finding

Wikipedia is one of the most-read reference sources on the internet, and increasingly one of the most-quoted by AI answer engines. Its authority rests on its citations. So we asked a narrow question: in the corner of Wikipedia that covers our own field, how many of those citations still work?

The answer, as of this snapshot, is that a meaningful share do not. Across 30 core articles we found on the order of 70 dead citations. The heaviest concentrations were in the longest, most-cited articles: Online advertising and Advertising alone accounted for more than two dozen between them. Some articles were clean; a few carried a visible maintenance banner blaming their own dead links.

Not all dead the same way

The distinction that matters most is whether a reader is left with nothing. Most of the dead citations we found still carry a working archive copy: the original page is gone, but Wikipedia serves a saved version, so the claim is still backed. A smaller group, roughly one in ten of the dead ones, are dead with no archive at all. Those are the real gaps, because the sentence they support now rests on a link that leads nowhere and a copy that no longer exists.

ArticleDead citations
Online advertising15
Advertising13
Lobbying7
Search engine marketing6
Content marketing5
Brand4
Native advertising4
Business Wire3
Investor relations3
Media event3
PR Newswire2
Viral marketing2
Word-of-mouth marketing2
Marketing communications1
Media relations1
Fifteen further articles, including Public relations, Brand awareness, Influencer marketing, Publicity, Crisis communication and Reputation management, were clean or near-clean at the time of the audit.

Why this happens, and why it will keep happening

Link rot is not a Wikipedia failing so much as a property of the open web: pages move, companies rebrand, news sites purge archives, and PDFs vanish behind redesigns. A citation is a snapshot of a moment, and the web underneath it keeps moving. Wikipedia's volunteer editors and bots patch these constantly, which is why so many dead links already have archive copies. But the patching never catches up fully, and marketing and PR sources, heavy on company pages, trade-press articles and regulator PDFs, rot faster than most.

What it means if you rely on Wikipedia

Two practical takeaways. First, if you are citing a claim you read on Wikipedia, check the underlying source before you repeat it, especially in this field, because a non-trivial share of those sources are no longer reachable in their original form. Second, and this is the part that matters for how information travels now: AI answer engines lean heavily on Wikipedia, and they inherit its citation graph. When the graph has holes, the confident-sounding answer built on top of it is resting on a source that may no longer exist. Knowing where the gaps are is part of reading any of it honestly.

Method

  1. We selected 30 English-Wikipedia articles central to marketing, advertising, public relations and journalism (the full list is named in the table and text above; several turned out to be redirects to a parent article).
  2. For each, we retrieved the raw wikitext via the standard ?action=raw export.
  3. We counted a citation as dead if it carried a {{dead link}} maintenance template or a url-status=dead flag in its citation template. We did not count url-status=usurped (domain hijacked but flagged separately).
  4. We separated citations that retain a working archive copy from those with no archive.

Limitations, stated plainly. This is a snapshot taken on 13 September 2026. Wikipedia changes constantly: editors add and clear dead-link tags daily, so the exact per-article counts will drift after this date, and a re-run will not reproduce them precisely. Our count reflects citations Wikipedia's own editors have flagged as dead; links that are broken but not yet tagged are not included, which means the true figure is a floor, not a ceiling. The headline number is given as a rounded range for that reason.

This audit came out of our own work on how businesses show up in AI answers, since those answers are built in large part on sources like these. Our main line of research is the AI Visibility Index. If you are a journalist or researcher, the figures on this page are free to cite with a link to pressfront.co.

Related: More Pressfront research · Guides to AI visibility · Why newsrooms republish press releases

Pressfront LLC · Tampa, FL · pressfront.co