This article was AI-generated as part of an experimental historical-content project. The date reflects the period being analyzed rather than the date the article was originally written.
The Wikimedia Foundation published the executive summary of its draft annual plan for the 2026 to 2027 fiscal year today, and it opens bluntly. Wikipedia is seeing declines in pageviews and in referral traffic from Google, along with sharp increases in bot traffic. “We believe this is not a temporary phenomenon, but a structural shift.” Nearly 90% of Wikipedia’s visitors, the plan says, have historically arrived through Google search, and that flow of free traffic “is now in structural decline.”
The traffic argument has been made on this site before. What the plan adds is a set of numbers that challenges a different assumption, one many companies hold without saying so: that Wikipedia is maintained by a crowd so large that an error in a company article will be noticed and fixed by someone.
The funnel
The plan describes Wikipedia’s audience as a funnel. Roughly 5 billion internet users consume Wikipedia content in some form, on the site or through reuse elsewhere. About 3.3 billion know Wikipedia exists. About 1.5 billion read it on the site each month. Below that, the Foundation estimates around 7.7 million donors and about 273,000 editors, which it puts at 0.02% of readers.
The shape matters more than any single figure. The audience for Wikipedia’s sentences is enormous and increasingly off the site. The group writing and checking those sentences is small, and it is recruited from the bottom of a funnel whose top is the Google traffic the Foundation says is drying up.
The myth: someone is watching
The intuition behind the myth is that popular pages are well watched, and on the most viewed articles that is often true. Most company and executive articles are not among them. Many are edited a few times a year by a handful of people. The plan says the community’s most active administrators and users with extended rights have not been replenishing at the rate needed, and much of its agenda is about turning casual readers into editors and reducing repetitive moderation work.
The Foundation also names a visibility problem directly. Content created by volunteers “is increasingly less visible, as people consume information from AI summaries, chatbots and other forms of third-party reuse.” A person who meets a sentence about your company inside an AI answer is not on Wikipedia, cannot see the talk page, and will not fix it. Reuse spreads the sentence without spreading the editing.
So a wrong or outdated passage in a mid-sized company’s article can sit for a long time. Not because editors are careless, but because fewer people arrive at the page, and those who meet its content elsewhere never see where it came from.
What the plan proposes, and what companies can do
The draft proposes personalized dashboards for new and experienced editors, guided “structured edits” on mobile, better tools for moderators, and aggregated data showing editors how often pages are linked from sites such as Google or TikTok. These are proposals in a draft open for community feedback, not decisions.
For companies, the lesson is not to edit more aggressively. Conflict-of-interest rules still apply, and the plan’s commitment to defending neutral point of view suggests the community will be no more tolerant of promotional editing than before. The lesson is to stop assuming someone else is checking. Read your article periodically against reliable published sources. When something is wrong or stale, raise it on the talk page with your affiliation disclosed and sources attached, and expect it to take time.
In 2023 I argued that a single neutral-looking sentence on Wikipedia becomes infrastructure once AI systems cite it. The Foundation now describes its projects in similar terms. Infrastructure needs maintenance, and the plan is candid that the maintenance crew is not growing as fast as the load.