Which Sources Wikipedia Accepts — and Which It Doesn't
If you've ever tried to publish a Wikipedia article — about your company, yourself, a product, or a project — you've probably run into the puzzling verdict: "no reliable sources." Meanwhile, the subject has a website, social accounts with hundreds of thousands of followers, a dozen publications, and even a segment on local TV. How can there be no sources?!
Take a breath. Let's break down how Wikipedia actually evaluates sources: which ones it accepts, which ones go straight to the bin, and why a single mention in a major newspaper outweighs your entire corporate blog. This is the most important piece we've published: the vast majority of rejections and deletions — whether the article is about a company, a person, or an event — come down to sources.
Take a breath. Let's break down how Wikipedia actually evaluates sources: which ones it accepts, which ones go straight to the bin, and why a single mention in a major newspaper outweighs your entire corporate blog. This is the most important piece we've published: the vast majority of rejections and deletions — whether the article is about a company, a person, or an event — come down to sources.
The core principle: "verifiability, not truth"
Wikipedia has a foundational rule that sounds almost philosophical: verifiability matters more than truth (WP:V). The encyclopedia doesn't care what "really" happened — it cares what independent, reliable sources have written about it.
The subject of your article can be a market leader, a renowned expert, or a landmark event — but if no one except you has written about it, that fact doesn't exist as far as Wikipedia is concerned. And the reverse: if Reuters, Forbes, and a respected trade journal have covered the subject in depth, the fact becomes "encyclopedic."
Everything else follows from this. Wikipedia runs every source through four questions (WP:RS, WP:GNG):
The green zone: what Wikipedia acceptsMajor national and business media. The New York Times, Reuters, Bloomberg, the Financial Times, The Wall Street Journal, Forbes — the gold standard. An analytical article that examines the subject of your article in substance is the strongest possible argument for notability.
Reputable trade publications. You don't have to be a front-page story. For tech that might be TechCrunch or The Verge (editorial pieces, not contributed posts!); for retail, the established trade journals; for pharma, the industry press. What matters is that the outlet has a real editorial team and a reputation in its niche.
Books and academic publications. If the subject appears in a monograph, a serious history of the field, or a peer-reviewed paper, that's a top-shelf source. Rare, but it carries a lot of weight.
Government and official records. Registries, court decisions, regulator data, statistics. Fine for confirming facts (incorporation date, licenses), but not for proving notability — more on why those are different things below.
Quality regional media. These count, with a caveat: proving notability with regional press alone is hard. A piece in a local paper will verify facts, but clearing the notability bar usually takes coverage a level higher.
The red zone: what never countsThis is where it starts to hurt. Brace yourself: this list probably includes almost everything you're proud of.
Your own website. Good only for trivial things: how the name is spelled, who runs the organization, basic biography. Notability — never. The logic is simple: the subject is an interested party.
Press releases. They don't count no matter where they're published — on your site or on a newswire. A press release is a reference letter you wrote for yourself. Roughly like a CV where the "weaknesses" field says "works too hard."
Press-release reprints (churnalism). The sneakiest trap of all. A news item on a decent-looking site, dressed up as journalism — but it's your own release with a light comb-over. Moderators spot these in seconds: identical wording across five outlets on the same day gives the source away instantly.
Any paid publications. Advertorials, "partner content," native ads, articles marked "sponsored" — all out. And an experienced reviewer will recognize a paid piece even without the label: by the tone, the absence of criticism, and ten brand mentions per paragraph.
Social media. Your YouTube channel with 500,000 subscribers impresses investors, but not Wikipedians. Facebook posts, LinkedIn updates, Instagram stories, X threads, TikTok videos — all of it is self-publishing with no editorial oversight or fact-checking. The same goes for other people's posts about you: an influencer is not a newsroom.
Directories, aggregators, listings. A profile on a review site, a company-directory card, a marketplace listing — these pages are generated automatically or created by the subject itself, and they prove nothing.
Wikipedia citing itself. You can't support a claim in one Wikipedia article with a link to another. And then there's the classic "citogenesis" trick: someone inserts a claim into Wikipedia, a careless outlet reprints it, and the reprint is then used to "verify" the original claim. Moderators know this move better than the people who invented it.
Wikipedia has a foundational rule that sounds almost philosophical: verifiability matters more than truth (WP:V). The encyclopedia doesn't care what "really" happened — it cares what independent, reliable sources have written about it.
The subject of your article can be a market leader, a renowned expert, or a landmark event — but if no one except you has written about it, that fact doesn't exist as far as Wikipedia is concerned. And the reverse: if Reuters, Forbes, and a respected trade journal have covered the subject in depth, the fact becomes "encyclopedic."
Everything else follows from this. Wikipedia runs every source through four questions (WP:RS, WP:GNG):
- Is it independent of you? Not written by you, not paid for by you, not commissioned by you.
- Is it reliable? There's an editorial team, a reputation, fact-checking, and accountability for errors.
- Is it secondary? The source analyzes and interprets rather than merely relays. A journalist's article about the subject is secondary; a registry entry, a raw interview transcript, or a database record is primary — usable for facts, useless for notability.
- Is the coverage significant? You're covered in depth, not mentioned in a single line of a list.
The green zone: what Wikipedia acceptsMajor national and business media. The New York Times, Reuters, Bloomberg, the Financial Times, The Wall Street Journal, Forbes — the gold standard. An analytical article that examines the subject of your article in substance is the strongest possible argument for notability.
Reputable trade publications. You don't have to be a front-page story. For tech that might be TechCrunch or The Verge (editorial pieces, not contributed posts!); for retail, the established trade journals; for pharma, the industry press. What matters is that the outlet has a real editorial team and a reputation in its niche.
Books and academic publications. If the subject appears in a monograph, a serious history of the field, or a peer-reviewed paper, that's a top-shelf source. Rare, but it carries a lot of weight.
Government and official records. Registries, court decisions, regulator data, statistics. Fine for confirming facts (incorporation date, licenses), but not for proving notability — more on why those are different things below.
Quality regional media. These count, with a caveat: proving notability with regional press alone is hard. A piece in a local paper will verify facts, but clearing the notability bar usually takes coverage a level higher.
The red zone: what never countsThis is where it starts to hurt. Brace yourself: this list probably includes almost everything you're proud of.
Your own website. Good only for trivial things: how the name is spelled, who runs the organization, basic biography. Notability — never. The logic is simple: the subject is an interested party.
Press releases. They don't count no matter where they're published — on your site or on a newswire. A press release is a reference letter you wrote for yourself. Roughly like a CV where the "weaknesses" field says "works too hard."
Press-release reprints (churnalism). The sneakiest trap of all. A news item on a decent-looking site, dressed up as journalism — but it's your own release with a light comb-over. Moderators spot these in seconds: identical wording across five outlets on the same day gives the source away instantly.
Any paid publications. Advertorials, "partner content," native ads, articles marked "sponsored" — all out. And an experienced reviewer will recognize a paid piece even without the label: by the tone, the absence of criticism, and ten brand mentions per paragraph.
Social media. Your YouTube channel with 500,000 subscribers impresses investors, but not Wikipedians. Facebook posts, LinkedIn updates, Instagram stories, X threads, TikTok videos — all of it is self-publishing with no editorial oversight or fact-checking. The same goes for other people's posts about you: an influencer is not a newsroom.
Directories, aggregators, listings. A profile on a review site, a company-directory card, a marketplace listing — these pages are generated automatically or created by the subject itself, and they prove nothing.
Wikipedia citing itself. You can't support a claim in one Wikipedia article with a link to another. And then there's the classic "citogenesis" trick: someone inserts a claim into Wikipedia, a careless outlet reprints it, and the reprint is then used to "verify" the original claim. Moderators know this move better than the people who invented it.
The grey zone: it depends on how you use itInterviews.
The most common point of dispute. An interview with the CEO, the founder, or the person the article is about is a dependent source: everything they say is treated as the subject speaking about itself. It won't prove notability. But it can verify biographical facts ("born, graduated, founded") — and if the journalist added their own analytical framing around the interview, that framing is independent material.
Passing mentions. "Exhibitors included companies A, B, and C" — that's not a source, that's a list. The rule of thumb: if you can't write at least a paragraph about the subject from the text, the coverage isn't significant. You need pieces where the subject is the protagonist, or at least one of the main characters.
Industry awards and rankings. Depends on reputation. A spot in a Forbes or Financial Times ranking is an argument. The "Golden Palm of Quality" from an organization even Google hasn't heard of is not. Especially if entering the award cost money (yes, moderators check). The same logic applies to people: a national industry award is an argument, a paid "Top 100 Entrepreneurs" plaque is not.
Wayback Machine and archives. An archived copy inherits the authority of the original: a saved Reuters article is still a Reuters article. An archive of a junk site is still an archive of a junk site. No miracles occur.
The Daily Mail story, or how an entire tabloid got banned
To appreciate how seriously Wikipedia takes source quality, here's a true story. In 2017, the English-language community held a major discussion and prohibited the use of the British tabloid Daily Mail as a source — citing its "reputation for poor fact checking, sensationalism, and flat-out fabrication." This was a newspaper with a circulation in the millions and one of the most-visited news websites in the world. Didn't help.
Wikipedia co-founder Jimmy Wales backed the community's decision, remarking that the paper had mastered the art of clickbait — and, sadly, the art of running stories that simply aren't true. Since then there's been a public perennial sources list, where the community keeps score year after year: green means reliable, yellow means "it depends," red means forget it. At the time of the ban, the Daily Mail was cited in tens of thousands of articles — and editors manually cleaned out most of those links.
The moral for business: if Wikipedia didn't hesitate to ban a newspaper with 130 years of history, it will toss your corporate blog into the bin without blinking.
Notability vs. verifiability — don't mix them up
Here's the subtlety almost everyone trips over. Sources on Wikipedia do two different jobs:
Verifiability — confirming a specific fact. For this, your official website works (for the name and leadership), a registry works (for the incorporation date), an interview works (for a biography).
Notability (WP:GNG) — proving that an article about you deserves to exist at all. And here, only independent reliable sources with significant coverage count. No registries, websites, or press releases will help.
The typical mistake: bringing 30 links, of which 28 verify facts but not one proves notability. The article gets deleted — and by the rules, rightly so. To clear the bar with confidence you need at least 2–3 strong independent pieces that cover the subject in substance. Five is better. Quality beats quantity: one analytical article in the business press outweighs twenty press-release reprints.
The most common point of dispute. An interview with the CEO, the founder, or the person the article is about is a dependent source: everything they say is treated as the subject speaking about itself. It won't prove notability. But it can verify biographical facts ("born, graduated, founded") — and if the journalist added their own analytical framing around the interview, that framing is independent material.
Passing mentions. "Exhibitors included companies A, B, and C" — that's not a source, that's a list. The rule of thumb: if you can't write at least a paragraph about the subject from the text, the coverage isn't significant. You need pieces where the subject is the protagonist, or at least one of the main characters.
Industry awards and rankings. Depends on reputation. A spot in a Forbes or Financial Times ranking is an argument. The "Golden Palm of Quality" from an organization even Google hasn't heard of is not. Especially if entering the award cost money (yes, moderators check). The same logic applies to people: a national industry award is an argument, a paid "Top 100 Entrepreneurs" plaque is not.
Wayback Machine and archives. An archived copy inherits the authority of the original: a saved Reuters article is still a Reuters article. An archive of a junk site is still an archive of a junk site. No miracles occur.
The Daily Mail story, or how an entire tabloid got banned
To appreciate how seriously Wikipedia takes source quality, here's a true story. In 2017, the English-language community held a major discussion and prohibited the use of the British tabloid Daily Mail as a source — citing its "reputation for poor fact checking, sensationalism, and flat-out fabrication." This was a newspaper with a circulation in the millions and one of the most-visited news websites in the world. Didn't help.
Wikipedia co-founder Jimmy Wales backed the community's decision, remarking that the paper had mastered the art of clickbait — and, sadly, the art of running stories that simply aren't true. Since then there's been a public perennial sources list, where the community keeps score year after year: green means reliable, yellow means "it depends," red means forget it. At the time of the ban, the Daily Mail was cited in tens of thousands of articles — and editors manually cleaned out most of those links.
The moral for business: if Wikipedia didn't hesitate to ban a newspaper with 130 years of history, it will toss your corporate blog into the bin without blinking.
Notability vs. verifiability — don't mix them up
Here's the subtlety almost everyone trips over. Sources on Wikipedia do two different jobs:
Verifiability — confirming a specific fact. For this, your official website works (for the name and leadership), a registry works (for the incorporation date), an interview works (for a biography).
Notability (WP:GNG) — proving that an article about you deserves to exist at all. And here, only independent reliable sources with significant coverage count. No registries, websites, or press releases will help.
The typical mistake: bringing 30 links, of which 28 verify facts but not one proves notability. The article gets deleted — and by the rules, rightly so. To clear the bar with confidence you need at least 2–3 strong independent pieces that cover the subject in substance. Five is better. Quality beats quantity: one analytical article in the business press outweighs twenty press-release reprints.
The 5-minute checklist: audit your sources
Take your list of publications about the subject of your future article — a company, a person, a product, an event — and answer honestly for each one:
Don't publish the article "and hope for the best" — that's the worst strategy: a deleted article stays in the logs, and the second attempt gets reviewed more strictly. The right move is to build the source base first. There are two ways to get there — a slower one and a faster one.
Path 1: Earn the coverage.
The classic route is organic media work: getting journalists at reputable national outlets, major business publications, or respected industry media genuinely interested in the story. What you're after is two to three full-length, reporter-written features where the subject of your future article is the story itself — not the garnish.
Not all press is created equal here. A quick funding-round blurb, an interview with the subject, a seasonal roundup, an affiliate listicle — none of it moves the needle, for reasons we covered above. What does: articles built around independently verifiable milestones (growth, funding, study results, career achievements, market share), voices of researchers and analysts with no ties to the subject, and a real look at where it stands among competitors or peers — not a tour of product features or a recitation of the bio. In other words, the kind of piece a journalist writes because the story is worth telling, not because you asked nicely.
Path 2: Accelerate it with qualified partners.
If time matters more than budget, the same result can be reached faster: coordinated outreach through media partners whose publications are consistently treated as reliable in Wikipedia sourcing discussions. To be clear, this is not buying advertorials — sponsored content marked as such carries zero weight. The difference is that a genuine editorial process still stands between the pitch and the publication: an editor decides whether the story runs and how it reads. You're paying for access and speed, not for the byline.
Want the details on the second path? Email us with the subject "Media Campaign for Wikipedia purposes" and the message: "Interested in details about a media campaign focused on Wikipedia-cited publications."
Whichever path you take — start Wikidata in parallel. While press coverage accumulates, a Wikidata entry can be created right away: the bar there is lower, and it benefits your Google Knowledge Panel regardless of the article.
And only then — publish. Once you have 3–5 solid independent pieces, the article is written neutrally, every claim is backed by a citation — and it quietly lives on for years.
A source audit is exactly where we start every project: we look at what's already been written about you, tell you honestly whether you clear the bar today, and if not — build a plan for getting there. It's less exciting than "published in three days," but the article doesn't turn into a pumpkin afterwards.
Take your list of publications about the subject of your future article — a company, a person, a product, an event — and answer honestly for each one:
- Did you pay for this piece — with money, barter, or "as a favor"? If yes — cross it out.
- Was the text written by a journalist on their own initiative, not from your press release? If it's from your release — cross it out.
- Is it secondary — does it analyze and interpret, rather than just print your own words or raw data? If it's an interview transcript or a registry entry — set it aside for facts, not for notability.
- Does the outlet have a real editorial team and a reputation? If it's a blog, an aggregator, or an "everything" site — cross it out.
- Is the piece substantively about the subject — could you write several paragraphs about it from the text? If the subject is a single line in it — cross it out.
- After all the crossing out, are 2–3 pieces still standing? Congratulations — you have a basis for an article. Zero left? That's a useful result too: you just saved the money you'd have spent on a publication that would've been deleted within a week.
Don't publish the article "and hope for the best" — that's the worst strategy: a deleted article stays in the logs, and the second attempt gets reviewed more strictly. The right move is to build the source base first. There are two ways to get there — a slower one and a faster one.
Path 1: Earn the coverage.
The classic route is organic media work: getting journalists at reputable national outlets, major business publications, or respected industry media genuinely interested in the story. What you're after is two to three full-length, reporter-written features where the subject of your future article is the story itself — not the garnish.
Not all press is created equal here. A quick funding-round blurb, an interview with the subject, a seasonal roundup, an affiliate listicle — none of it moves the needle, for reasons we covered above. What does: articles built around independently verifiable milestones (growth, funding, study results, career achievements, market share), voices of researchers and analysts with no ties to the subject, and a real look at where it stands among competitors or peers — not a tour of product features or a recitation of the bio. In other words, the kind of piece a journalist writes because the story is worth telling, not because you asked nicely.
Path 2: Accelerate it with qualified partners.
If time matters more than budget, the same result can be reached faster: coordinated outreach through media partners whose publications are consistently treated as reliable in Wikipedia sourcing discussions. To be clear, this is not buying advertorials — sponsored content marked as such carries zero weight. The difference is that a genuine editorial process still stands between the pitch and the publication: an editor decides whether the story runs and how it reads. You're paying for access and speed, not for the byline.
Want the details on the second path? Email us with the subject "Media Campaign for Wikipedia purposes" and the message: "Interested in details about a media campaign focused on Wikipedia-cited publications."
Whichever path you take — start Wikidata in parallel. While press coverage accumulates, a Wikidata entry can be created right away: the bar there is lower, and it benefits your Google Knowledge Panel regardless of the article.
And only then — publish. Once you have 3–5 solid independent pieces, the article is written neutrally, every claim is backed by a citation — and it quietly lives on for years.
A source audit is exactly where we start every project: we look at what's already been written about you, tell you honestly whether you clear the bar today, and if not — build a plan for getting there. It's less exciting than "published in three days," but the article doesn't turn into a pumpkin afterwards.