Wikipedia Pageviews Scraper — Trending Articles by Day or Month

Pull the top-viewed Wikipedia articles for any project (en, es, de, ja, etc.) and any device segment (all, desktop, mobile-web, mobile-app) from the official Wikimedia Pageviews API. Each day Wikipedia publishes the top 1,000 most-viewed articles — a powerful, real-time signal of public attention.

What this Actor does

This Actor wraps the Wikimedia REST API's /metrics/pageviews/top/{project}/{access}/{year}/{month}/{day} endpoint, optionally iterates over a date range, and outputs ranked article-day records ready for trend analytics, news intelligence, search demand modeling, and audience research.

Use cases:

• News & trend intelligence: Detect breaking events the moment they spike on Wikipedia.
• Marketing & SEO: Identify topics with surging real-world attention to inform content calendars.
• Finance: Wikipedia pageview spikes correlate with consumer attention shifts ahead of earnings or product launches.
• Academic research: Build datasets for attention economics, cultural studies, or NLP topic modeling.
• AI training: Mine high-attention article titles to expand topic coverage in RAG indexes.

How to scrape Wikipedia pageviews

1. Pick a Project — en.wikipedia for English Wikipedia (default), es.wikipedia, de.wikipedia, commons.wikimedia, etc.
2. Pick an Access Type — all-access, desktop, mobile-web, mobile-app.
3. Set a Start Date and End Date (YYYY-MM-DD). Leave blank to default to 3 days ago (the API has a ~1-day publication lag).
4. Toggle Exclude Meta Pages to drop Main_Page, Special:Search, and other non-article entries.
5. Set Max Articles Per Day (1–1,000).
6. Run. Output is one record per article per day.

Output fields

Field	Description
date	Date (YYYY-MM-DD)
rank	Daily rank (1 = most viewed)
article	Wikipedia article title (underscored)
articleDisplay	Human-readable title (spaces)
views	Number of pageviews on that day
project	Wikipedia project (e.g. en.wikipedia)
access	Access type used (all-access, desktop, mobile-web, mobile-app)
articleUrl	Direct article URL

Pricing

Pay-per-result. The Wikimedia API is very generous; one HTTP request returns up to 1,000 ranked articles per day. Filtering and pagination happen client-side, so runs are fast.

Example input

{
  "project": "en.wikipedia",
  "access": "all-access",
  "startDate": "2026-05-01",
  "endDate": "2026-05-07",
  "excludeMeta": true,
  "maxPerDay": 200
}

Example output (single record)

{
  "date": "2026-05-01",
  "rank": 5,
  "article": "Kentucky_Derby",
  "articleDisplay": "Kentucky Derby",
  "views": 421003,
  "project": "en.wikipedia",
  "access": "all-access",
  "articleUrl": "https://en.wikipedia.org/wiki/Kentucky_Derby"
}

FAQ

Q: Why does today/yesterday return nothing? Wikimedia publishes pageviews on a 1–2 day lag. Default to daysAgo=3 if you want guaranteed data.

Q: Can I pull monthly totals? This Actor uses the daily top endpoint. For monthly aggregates, use a 28–31 day range and aggregate downstream.

Q: Are bot views included? The API filters out most automated traffic but is not perfectly clean. For research-grade exclusion, post-process against known bot lists.

Q: What's the difference between access types? all-access is the union; the others slice by device class.

Legal disclaimer

This Actor uses the public Wikimedia REST API in accordance with the Wikimedia API Terms of Use. Wikipedia content is released under CC BY-SA; pageview data is similarly open. Attribute Wikimedia Foundation in derivative analyses. Not affiliated with or endorsed by the Wikimedia Foundation.