Wikipedia's Concept and Explainer Pages Lost a Median 41.9% of Their Readers Since 2022, Compared With 26.8% for Pages About People

Summary

Picture a bookkeeping firm that wrote a clear page called 'What is accrual accounting?' It ranked well and brought in new leads for years. Then the visits thinned out, even though the page did not change. The question is still being asked. The answer just shows up before the click now.

We took a random 1-in-10 sample of the 10,000 most-read English Wikipedia articles of 2022 and classified each one by hand by subject, such as people, places, pop culture or explainers. Then we compared each article's views in 2022 with the latest 12 months. Pages that explain a concept or a technology lost a median 41.9% of their readers. Pages about people lost a median 26.8%.

Every type of page in this sample lost readers, partly because pages that were popular in 2022 tend to cool off. What matters is the gap. Explainers lost far more than pages about people, and almost none of them grew.

What we measured

We ranked English Wikipedia articles by combined desktop and mobile web views in 2022, kept the top 10,000, and drew a repeatable 1-in-10 random sample using a fixed formula applied to each title. We classified each of the 1,032 sampled titles by hand into people, pop culture, events and history, places, organizations and brands, and four explainer types: general concepts, technology, health and science. Lists, meta pages and adult or piracy pages (46 titles) were excluded.

For each article we compared 2022 views with October 2025 to September 2026 views and report the median change per type, the share that lost more than half their views and the share that grew. The labels were assigned by hand, not by an automated classifier.

Explainers fell further than people, places or brands

Median change by article type, with the number of sampled articles in each type:

Pop culture pages fell most, but they are tied to releases. A TV season that peaked in 2022 would lose readers with or without AI. Explainer pages have no release date, which makes their drop harder to put down to normal fading.

Almost no explainer page grew

Looking at winners and heavy losers tells the same story.

One in five people pages gained readers. Among explainer pages it was fewer than one in twenty-five.

Tech and health explainers lost about half their readers

Within the explainer group, the drop varies by subject.

These are the kinds of questions people can now ask a chatbot or see answered at the top of search: what a PDF is, how penicillin works, what gout feels like. The pages still exist and are still accurate. Fewer people are opening them.

What this means

If your content plan leans on 'what is' and 'how does it work' posts, expect those pages to lose traffic faster than anything tied to your own business, people or news. Definitions are the easiest thing for AI to answer without a click. That is why most AI SEO services start by moving effort away from definition posts and toward pages only you can write.

Keep explainers, but rebuild them around what only you can say: your prices, your process, your local experience, real examples from your customers. That gives AI tools a reason to cite you by name. Whether they actually do is the first question in any AI search optimization plan, and our free AI visibility audit answers it for you.

Limits

Wikimedia pageview counts include some automated traffic that its filters do not catch, and Wikimedia changed how it detects bots in 2025, which lowers counted views from that point on. Some of the change in any period may reflect better bot filtering rather than fewer human readers.

We count views of titles that had at least 5 views in a given month on a given site, which drops a very small tail of rarely read pages.

The sample is drawn from 2022's most-read pages, so every type tends to decline (regression to the mean). English Wikipedia as a whole fell 8.1% over the same period. Compare types with each other, not with zero.

Classification is manual and some titles could fit more than one type.

Some 2022 explainer peaks were news-driven (for example a disease outbreak), which adds to their decline.

Medians are used because a few very large pages would dominate averages.

Frequently asked questions

Related studies

• Wikipedia readership in the AI answer eraWikipedia Traffic Since ChatGPT: English Views Down 8.1%
• Wikipedia readership in the AI answer eraProgramming Pages Lost a Median 57.9% of Wikipedia Readers
• Wikipedia readership in the AI answer eraSpanish Wikipedia Views Fell 36.3%; English Fell 8.1%
• Wikipedia readership in the AI answer eraPopular Wikipedia Pages Lost 24.5%; Niche Pages Gained 12.2%