Anyone can change a Wikipedia article, and yet most articles are not nonsense. People find that hard to square, and the usual explanation, that Wikipedia has moderators who approve things, is wrong. There is no approval step. Your edit is live the moment you save it.
What actually keeps articles coherent is a set of rules about evidence, plus a process for settling arguments. Understanding both tells you why some changes to your article stick and others vanish within the hour.
The three rules everything else hangs off
Wikipedia calls these the core content policies, and they are meant to be read together rather than traded against each other.
Verifiability. Material must be attributable to a published, reliable source. The test is whether a reader can check a claim, not whether the claim is true. This is the rule that surprises people most: a fact you know to be correct about your own life can be removed for having no published source, and a claim you know to be wrong can survive because a newspaper printed it. Wikipedia is reporting what sources say, and the article about you is a summary of coverage rather than a record of events.
No original research. Editors cannot publish their own analysis, conclusions or synthesis. Putting two sourced facts side by side to imply a third thing that neither source states is specifically prohibited. This is why an editor cannot correct your article based on a document they have but nobody has published.
Neutral point of view. Articles represent significant published views fairly and in proportion to their prominence in reliable sources. Proportion is the operative word and the one most disputes are really about. Devoting a third of an article to a minor incident can breach neutrality even when every sentence is sourced and accurate.
Living people get a fourth layer. The biographies policy raises the standard: contentious material that is unsourced or poorly sourced should be removed on sight, and the burden sits with whoever wants it in.
What counts as a reliable source
The rules above are only as strong as this definition, and it is narrower than people assume.
Wikipedia looks for published sources with a reputation for fact-checking and editorial oversight, and prefers ones independent of the subject. Established news organizations, academic publishing and reputable books sit at the top. Self-published material, press releases, and company websites are usable for uncontroversial facts about the subject and carry little weight for anything contested.
Two consequences matter if it is your article. A press release announcing your award does not establish that the award is significant. And paid placements in outlets that will print what you send them are recognized and discounted, sometimes on a published list of sources the community has deprecated. Buying coverage to change an article is a route that the people who maintain articles already know about.
How disputes actually end
When editors disagree, the mechanism is consensus, which on Wikipedia means something specific. Nobody counts hands. Consensus emerges from discussion on the article’s talk page, where an argument grounded in policy carries weight and an argument grounded in preference does not. Ten people saying they prefer a version count for less than one person citing the sourcing rule.
If reverting goes back and forth, that is edit warring, and Wikipedia treats the behavior as a problem independent of who is right. The three-revert rule is a bright line: more than three reverts of a page within 24 hours will usually get an editor blocked, whichever side they were on. Beyond that there are noticeboards, requests for comment that draw uninvolved editors, and eventually arbitration.
None of this is fast. A contested sentence can be argued over for weeks. That is the system working as designed, and it is the main reason getting to a dispute early matters more than arguing well later.
Why two similar people have very different articles
One executive has a detailed, current, carefully sourced article. Another with a comparable career has four paragraphs, two of them about a lawsuit, and a photograph from 2009. Neither outcome was decided by anyone.
Wikipedia is written by volunteers who choose their own subjects. Coverage follows their interests, and their interests are uneven. Topics with an enthusiastic community get thorough articles that stay current. Topics without one get whatever a passing editor wrote years ago, plus whatever has been bolted on since.
This produces two effects worth understanding if it is your article.
Sourcing shapes the article more than events do. A career phase that was covered heavily by the press becomes several paragraphs. A more consequential phase that nobody wrote about stays invisible, because there is nothing to cite. The article is a reflection of published coverage, and where coverage was lopsided the article inherits the lopsidedness and calls it neutrality.
Quiet articles drift. An article nobody watches accumulates whatever gets added. A single editor with an interest can shape a low-traffic biography substantially over a few months, one plausible edit at a time, and no policy is violated at any step.
Neither is a conspiracy and neither has a villain. It is what a volunteer encyclopedia produces, and it is why the useful question about your article is not whether it is fair but whether anybody is looking at it.
When a page gets locked
Sometimes the answer is to restrict editing rather than settle the argument. Protection comes in levels: semi-protection blocks new and unregistered accounts, extended-confirmed protection requires a longer track record, and full protection limits editing to administrators. Pending changes is a softer option where edits by unregistered users are held until a reviewer accepts them.
Protection is applied to a problem, not as a favor, and it is normally temporary. You cannot request that your own article be locked because you dislike its contents. You can request it when there is demonstrable sustained vandalism, and the request goes to a public noticeboard where an administrator decides.
What this means if it is your page
Three things follow from all of the above.
Sources are the only currency. Your position on an article is worth nothing without published material behind it. Gathering the sources is the work, and it is worth doing before you say anything on a talk page.
Timing beats argument. A change challenged in its first hours is a conversation with one editor. The same change challenged after it has been stable for two months has become the status quo, and the burden has quietly moved to you.
You are not the right person to make the edit. The rules discourage editing articles you have a stake in, and doing it anyway usually costs more than it gains. The rules, and what you can do instead, are here.
The first two both come down to knowing early. We watch the page and email you within 60 seconds of a change, with the diff attached, so you are reading edits while they are still fresh rather than inheriting them as settled text. See how it works.