Data study · Wikipedia readership in the AI answer era
Wikipedia's 'Comparison of' Pages Lost a Median 40.8% of Their Readers Since 2022, More Than Three Times the Drop for Regular Articles
Across 2.95 million English Wikipedia articles, side-by-side comparison pages lost a median 40.8% of their views between 2022 and the past 12 months, and 'List of' pages lost 24.7%. Regular articles lost a median 12.1%.
Analysis of 2,950,961 English Wikipedia articles grouped by page format, 2022 vs October 2025 to September 2026 · Wikimedia pageviews via BigQuery, October 2026
Key findings
- 1
The median 'Comparison of' page lost 40.8% of its views between 2022 and the past 12 months, across 504 pages; only 12.5% of them gained readers.
- 2
The median 'List of' page lost 24.7%, across 76,372 pages.
- 3
The median standard article lost 12.1%, across 2,843,616 articles, and 36.8% of standard articles gained readers.
- 4
The gap holds at the same popularity level: among pages with 5,000 to 50,000 views in 2022, comparison pages fell a median 42.1%, list pages 28.5% and standard articles 17.2%.
- 5
Timeline pages (median minus 33.3%) and outline pages (minus 29.7%) also fell more than standard articles; glossaries fell a median 31.2%, but their largest pages held up, losing 5.0% in total.
Summary
Picture a software reseller whose best-performing page, built years ago, is a big table comparing ten accounting tools side by side. For a long time it pulls steady traffic from buyers who want options in one place. Lately that traffic has thinned, as more buyers ask an AI assistant to compare the tools for them.
We grouped 2.95 million English Wikipedia articles by format and compared their readers in 2022 with the past 12 months. Pages titled 'Comparison of', which line up products, standards or software side by side, lost a median 40.8% of their readers. 'List of' pages lost a median 24.7%. Regular articles lost a median 12.1%.
Comparisons and lists are exactly what AI answers produce on request: a short table, a ranked shortlist, the top five. Some readers who used to scan those pages may now get the summary without opening them.
What we measured
We took every English Wikipedia article with at least 1,200 views in 2022 (about 100 a month), grouped it by the opening words of its title (such as 'Comparison of' or 'List of'), and compared desktop plus mobile web views in calendar year 2022 with October 2025 to September 2026. Articles whose views rose or fell more than twentyfold were set aside as likely renames, merges or deletions.
Because popular pages tend to lose more readers than small ones, we also compared formats within three popularity bands based on 2022 views. Medians are used so that a few very large pages do not dominate.
Comparison pages lost more than three times as much as regular articles
Median change per page by format:
Median % change
| Format | Pages (N) | Median % change | Total views change | Share that gained |
|---|---|---|---|---|
| Comparison of | 504 | -40.8 | -38.1% | 12.5% |
| Timeline of | 1,900 | -33.3 | -36.9% | 20.9% |
| Glossary of | 346 | -31.2 | -13.6% | 19.4% |
| Outline of | 731 | -29.7 | -33.6% | 26.3% |
| History of | 4,891 | -29.3 | -36.7% | 20.7% |
| List of | 76,372 | -24.7 | -27.5% | 25.1% |
| Standard article | 2,843,616 | -12.1 | -20.1% | 36.8% |
Source: aisearchoptimization.net analysis of 2,950,961 English Wikipedia articles grouped by page format, 2022 vs October 2025 to September 2026 · Wikimedia pageviews via BigQuery, October 2026
Every format built to organize information for the reader, whether as a table, a list, a timeline or an outline, fell further than ordinary articles. Comparison pages fell the furthest.
The gap is not just about popularity
Popular pages fall more in general, so we compared formats at the same size:
| Format | Under 5,000 views | 5,000 to 50,000 views | 50,000+ views |
|---|---|---|---|
| Comparison of | -37.1% (154) | -42.1% (299) | -42.7% (51) |
| List of | -17.6% (32,747) | -28.5% (34,616) | -34.7% (9,009) |
| Timeline of | -25.1% (1,047) | -41.5% (734) | -41.2% (119) |
| Outline of | -13.1% (403) | -43.7% (311) | -55.5% (17) |
| Standard article | -6.0% (1,538,678) | -17.2% (1,058,549) | -25.9% (246,389) |
Source: aisearchoptimization.net analysis of 2,950,961 English Wikipedia articles grouped by page format, 2022 vs October 2025 to September 2026 · Wikimedia pageviews via BigQuery, October 2026
In every band, comparison and list pages fell more than standard articles. Among smaller pages the gap is widest: comparison pages lost a median 37.1% while standard articles lost 6.0%.
Comparison pages lost readers on desktop and mobile alike
Unlike many topics, comparison pages fell about equally on desktop and mobile web.
| Format | Desktop change | Mobile web change |
|---|---|---|
| Comparison of | -38.6% | -37.2% |
| List of | -24.9% | -29.2% |
| Timeline of | -31.9% | -42.4% |
| Standard article | -25.6% | -17.2% |
Source: aisearchoptimization.net analysis of 2,950,961 English Wikipedia articles grouped by page format, 2022 vs October 2025 to September 2026 · Wikimedia pageviews via BigQuery, October 2026
Comparison pages are often read at a desk while someone is choosing between products or standards. That is the setting where AI assistants now sit one tab away.
What this means
If your content plan leans on 'best X' lists and 'X vs Y' comparison posts, the Wikipedia pattern suggests those pages may lose readers faster than the rest of your site. The format itself is what AI answers copy. A buyer who asks for a comparison gets one in seconds. An AI search optimization agency would point you toward pages AI cannot copy: your own prices, results and customer stories.
Comparisons still matter, but they now need to feed AI answers rather than replace them: clear, current facts about your own product or service, honest pros and cons, prices and who you are best for. That gives AI tools something specific to quote. Run a free AI visibility audit to see whether you show up when buyers ask for options, since that is where AI brand visibility now gets decided.
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.
Format is taken from the title only; many list-style articles do not start with 'List of'.
Comparison pages on Wikipedia mostly cover software, file formats and technical standards, so they skew toward technical readers.
Regression to the mean affects all formats; the popularity-band comparison controls for it only roughly.
Embed this research
Add the charts from this study to your page. The embed links back here.
<iframe src="https://aisearchoptimization.net/embed/data-studies/comparison-and-list-pages-wikipedia-decline" title="Comparison Pages Lost 40.8% of Wikipedia Readers" width="100%" height="720" style="border:0" loading="lazy"></iframe>
Cite this study
APA
Grenyo, K. (2026). Comparison Pages Lost 40.8% of Wikipedia Readers. AI Search Optimization. https://aisearchoptimization.net/resources/data-studies/comparison-and-list-pages-wikipedia-decline
Plain link
Comparison Pages Lost 40.8% of Wikipedia Readers - AI Search Optimization - https://aisearchoptimization.net/resources/data-studies/comparison-and-list-pages-wikipedia-decline
Frequently asked questions
Are comparison pages losing traffic to AI?
On Wikipedia, comparison pages lost a median 40.8% of their views between 2022 and the past 12 months, across 504 pages, more than three times the 12.1% median drop for standard articles.
Do 'list of' pages lose more readers than normal articles?
Yes. The median 'List of' page lost 24.7% of its views across 76,372 pages, against 12.1% for standard articles.
Is this just because comparison pages were more popular?
No. At the same popularity level (5,000 to 50,000 views in 2022), comparison pages fell a median 42.1% and standard articles 17.2%.
Did any page formats hold up?
Standard articles held up best, and 36.8% of them gained readers. The largest glossary pages also held steady, losing 5.0% in total.
Related studies
- Wikipedia readership in the AI answer eraAI Topics Draw 2.9x the Wikipedia Readers of Search Engines
- Wikipedia readership in the AI answer eraSEO vs GEO on Wikipedia: SEO Page Views Down 51.9%
- Wikipedia readership in the AI answer eraHome System Pages on Wikipedia Lost 50.4% of Readers
- Wikipedia readership in the AI answer eraLegal Concept Pages Lost 46.9% of Wikipedia Readers
See where AI search leaves you out
Run the free AI visibility audit. If you want the fixes done for you, the plan is a flat $299/month.