Skip to main content
NewNMC’s advertising guidelines for doctors are in force from 6 Oct 2026. See what to change, and get a free compliance auditNew NMC advertising rules for doctors: what to change
AI Search

Wikipedia and Wikidata for Hospitals: Notability, COI and What Is Allowed

What it takes for a hospital to qualify for a Wikipedia article, how conflict-of-interest and paid-editing rules apply to marketing teams, and what to do instead of commissioning a page.

Wikipedia and Wikidata for Hospitals: Notability, COI and What Is Allowed

This article is for hospital marketing, communications and legal teams who have been told that a Wikipedia page would help them show up in AI answers, or who have received an email from a freelancer offering to create one. It explains what Wikipedia expects, what the conflict-of-interest and paid-editing rules require, and what a hospital can legitimately do.

The pressure to act is understandable. Peec AI's analysis of about 30 million cited sources, published in March 2026, ranked Reddit, YouTube, LinkedIn and Wikipedia as the most-cited domains across the AI search products it studied, with the caveat that rankings differ by industry and platform. Wikipedia matters to AI systems and to search. That does not mean every hospital can or should have an article, and the routes people use to get one create real risk.

Notability: the test comes first

Wikipedia articles are about subjects that are notable in Wikipedia's sense. The guideline for organisations and companies) says an organisation is presumed notable if it has been the subject of significant coverage in multiple reliable secondary sources that are independent of it. Each source is judged against four criteria: it must be significant (addresses the organisation directly and in depth), independent (not produced by people with a stake in it), reliable (with editorial oversight) and secondary (analysing rather than being primary material).

The guideline lists what does not count: press releases and coverage based on them, paid or sponsored material, routine business news such as expansions, hires or product launches, brief mentions and directory listings, minor awards, and primary documents such as annual reports. It also expects coverage beyond local media for most organisations. Hospitals have no special criteria; they meet the general test or they do not.

Apply this honestly to your hospital:

  1. List every independent article about the hospital that is not a press release, an advertorial or a routine announcement.
  2. For each, ask whether it discusses the hospital in depth: its history, a notable clinical programme, an independent investigation, a regulatory action, a notable event.
  3. Count how many come from national or international outlets or from academic or government sources.

If the list is short, the answer is that the hospital is not currently notable by Wikipedia's definition. Large, long-established institutions with sustained independent coverage often are. A well-run 150-bed hospital with good local press often is not. That is not a judgement about quality of care.

Conflict of interest: what the rules say

Wikipedia's conflict of interest guideline says COI editing is strongly discouraged and that editors with a COI are strongly discouraged from editing affected articles directly, with an exception for small, uncontroversial fixes such as removing spam or correcting typos. Anyone editing for pay must disclose who is paying them, who the client is and any other relevant affiliation. The guideline describes where to disclose, how paid editors should keep a visible list of their paid contributions and how COI editors can instead propose changes on the article's talk page using a request template so volunteers can review them. New articles should go through the Articles for Creation process, and COI editors should not review articles they are connected to.

For a hospital, "conflict of interest" covers more than the CEO. It includes the marketing team, the communications agency, the PR vendor, the web team and, in many readings, the doctors employed at the hospital when they edit about their employer.

The Wikimedia Foundation's Terms of Use add a legal layer. The paid-contribution disclosure page says the Terms of Use require editors to disclose their employer, client and affiliation for any contribution for which they receive or expect compensation, in at least one of three places: the user page, the talk page of the paid contributions or the edit summary.

Why paid editing is a risk

Four risks matter to a hospital.

1. Reputational. Undisclosed paid editing is a recognised offence on Wikipedia. If found, the article may be deleted, accounts blocked and the episode documented, sometimes on a public page about the editor or company. A search for your hospital's name could then return the story about the edit above the story you wanted.

2. Content. Promotional language, uncited awards and unsourced claims are removed. Articles built from your own materials are likely to be rewritten or deleted.

3. Legal and regulatory. In India, hospital and doctor communications are subject to advertising rules. A page you commission and control in substance can be regarded as your publicity. In other markets, such as the UAE, healthcare advertising may require permits. Any content you cause to be published about your services should be reviewed against those rules.

4. Vendor risk. Freelancers who promise a page often run networks of accounts or use sockpuppets. Your name is attached to what they do.

The point is not that a hospital must never touch Wikipedia. It is that the permitted routes are slow, public and subject to volunteer judgement.

What is allowed

These actions are consistent with the guidelines summarised above. Always read the current pages before acting.

  • Fix obvious factual errors on the talk page. If the article exists and states the wrong number of beds, the wrong address or an incorrect date, post a request on the article's talk page with a reliable source, disclose your connection and let volunteers decide.
  • Remove clear vandalism or spam. The exception for uncontroversial edits is narrow. If in doubt, ask.
  • Disclose, always. Use your own account, state your role and employer on your user page and keep the list of any paid edits visible.
  • Propose a draft through Articles for Creation. If your notability list is strong, a disclosed draft can be submitted for review by independent volunteers. Expect questions, and expect refusal if the sources are weak.
  • Supply sources, not text. You can publish material that independent sources may later use: annual reports, accreditation records, clinical outcomes published in journals with the clinicians' consent, official statements. Volunteers can cite those if independent coverage exists.

Want this applied to your practice? Get a free ai search audit.

Book my free audit

What is not allowed or is unwise

  • Creating the page yourself, without disclosure, from the hospital's own materials.
  • Hiring a freelancer who will not disclose the arrangement on Wikipedia.
  • Editing competitor pages.
  • Removing criticism that is well-sourced.
  • Using multiple accounts.
  • Creating a Wikidata item for the hospital full of promotional descriptions or references to your own website only.

Wikidata is a different gate

Wikidata's notability criteria are lighter. An item is acceptable if it has a valid sitelink to a Wikimedia project, or refers to a clearly identifiable entity that can be described using serious, publicly available references, or fulfils a structural need. A hospital group with a documented presence can meet the second test without a Wikipedia article. That is why Wikidata is more practical for most hospitals than a Wikipedia page.

Still, keep to the same principles: avoid promotional text, give references, disclose your connection if you edit, and do not create items you cannot support. Our companion post on building the hospital's entity graph is a technical matter and sits alongside this one; the point here is the permission and risk boundary.

What to do instead

Wikipedia is one source among many that AI systems and search engines read. If the hospital is not notable by Wikipedia's rules, work on the sources that matter anyway.

  1. Earn independent coverage. A clinical programme with published data, a public-health collaboration, an investigation-worthy innovation or a genuine local-health story attracts press. Press releases will not count, but a journalist's article might.
  2. Publish expert content under named, registered clinicians. Reviewed by the provider, with sources. See our guide to E-E-A-T authority building.
  3. Keep your own entity facts consistent. Business Profiles, accreditation listings, LinkedIn and structured data on your site, as covered under medical entity SEO.
  4. Build a citation programme. Society pages, accreditation bodies, medical journals, government health portals. The LLM citation building service describes this work.
  5. Read the full approach to AI visibility in the AI search optimisation guide.

Test whether any of this moves answers

If you do secure a legitimate Wikipedia or Wikidata presence, measure carefully. Build 50 queries about the hospital, its departments and its doctors. Run them on the AI surfaces your patients use, on two days, before and after, and log which sources are cited and whether the facts are correct. Because answers vary between runs, resist drawing conclusions from small changes. Record the log. If nothing changes, your page may still be worth having for accuracy reasons, but you should not promise a board member an AI visibility lift.

Where to start

  • Write the notability list: independent, in-depth, non-routine, non-local sources only.
  • If fewer than a handful qualify, stop and spend the effort on coverage and entity consistency.
  • If an article already exists, review it, collect sources and post disclosed talk-page requests for errors.
  • Never accept an offer that will not disclose the paid relationship on Wikipedia.
  • Write a short internal policy that says who may edit what, and what must be disclosed.

If you want help structuring citations and publication workflows that support visibility across AI search, AiLLMz does that, and it should sit behind, not instead of, the earned coverage described above.

Filed underWikipedia page for hospitalWikipedia notability hospitalWikipedia conflict of interestpaid editing Wikipedia riskWikidata hospital

About the author

Co-Founder & CTO · Gurugram, India

Arush builds the systems the rest of the company runs on — the patient-acquisition pipelines, the AI chatbots, the analytics that tie a single click to a booked appointment. If a campaign reports a number, it's because something his team built is tracking it.

Free guide · 15 pages

The Patient Acquisition Blueprint

A 90-day plan from first search to booked appointment.

The Pioneers Brief — one essay, three benchmarks, one teardown, every Thursday. Subscribe free →

Free ai search audit

Your next chapter starts with a conversation.

Three quick questions. We come prepared with a plan for your specialty and city.

  • A 30-minute call with a healthcare specialist
  • Your top three growth opportunities, in writing
  • Honest advice — even if that means not hiring us
  • No obligation and no hard sell
You’ll speak with a senior strategist
Consultations are hosted by Arush Thapar’s team.
Step 1 of 3Takes 30 seconds

What do you want to improve first?

Details stay privateNo obligationReply within 1 working day

Client portfolio · Healthcare websites

Work relevant to wikipedia and wikidata for hospitals notability coi and what is.

Explore client work →