Wikipedia Trends Scraper — Top Articles, Country Traffic & Edit Activity

Find out what the world actually looked up today. This Actor pulls Wikipedia's daily trend rankings straight from the official Wikimedia REST Analytics API — no API key, no login, no proxy required — and turns them into clean, ranked rows you can export as JSON, CSV, Excel or feed into your own pipeline.

Wikipedia is the largest unfiltered record of public curiosity on the internet. When a film drops, a politician is appointed, or a disaster happens, the pageview ranking moves hours before most trend tools notice. This scraper gives you that ranking, per day, per language edition, and per country.

---

How this differs from wikipedia-pageviews-scraper

We publish two Wikipedia analytics Actors. They solve opposite problems:

	wikipedia-trends-scraper (this one)	wikipedia-pageviews-scraper
Question it answers	"What was most viewed?"	"How did my article perform?"
Input	A date (+ project or country)	A list of article titles
Direction	Discovery — top-down rankings	Measurement — bottom-up time series
Output	Ranked lists (rank 1…N)	Per-article daily view series
Extra data	Top-per-country, most-edited pages	Per-article history, granularity options

Rule of thumb: use this Actor to discover which articles are trending. Use wikipedia-pageviews-scraper to track articles you already care about. Many users run both — discover here, then feed the winning titles into the pageviews Actor for a full history.

---

What you get

Mode top — most-viewed articles per day

The full daily ranking (up to 1,000 articles) for any Wikipedia language edition, with exact view counts.

Mode by_country — top articles per country

What readers in a specific country looked up, across all language editions mixed together. This is the interesting one: German readers hit de.wikipedia, en.wikipedia and fr.wikipedia on the same day, and this mode shows the blended reality. View counts are rounded (views_ceil) for privacy reasons.

Mode edits — most-edited pages per day

Editorial activity, not reader activity. A page with a sudden edit spike is a page where something is happening or being disputed — an early-warning signal that often precedes the traffic spike.

---

Example output

Mode top:

{
  "type": "top",
  "project": "en.wikipedia",
  "date": "2026-07-27",
  "rank": 4,
  "article": "The Odyssey (2026 film)",
  "views": 581815,
  "url": "https://en.wikipedia.org/wiki/The_Odyssey_(2026_film)",
  "source": "wikimedia",
  "scraped_at": "2026-07-28T14:13:05+00:00"
}

Mode by_country (DE):

{
  "type": "by_country",
  "country": "DE",
  "date": "2026-07-27",
  "rank": 7,
  "article": "Franziska Hoppermann",
  "project": "de.wikipedia",
  "views_ceil": 45800,
  "url": "https://de.wikipedia.org/wiki/Franziska_Hoppermann",
  "source": "wikimedia",
  "scraped_at": "2026-07-28T14:13:06+00:00"
}

Mode edits:

{
  "type": "edits",
  "project": "en.wikipedia",
  "date": "2026-06-30",
  "rank": 1,
  "article": "User:4meter4",
  "edits": 393,
  "url": "https://en.wikipedia.org/wiki/User:4meter4",
  "source": "wikimedia",
  "scraped_at": "2026-07-28T14:13:18+00:00"
}

One row per ranked article. Every field is nullable, so a missing value never breaks your downstream job.

---

Input

Field	Type	Default	Description
mode	select	top	top, by_country or edits
project	string	en.wikipedia	Wikimedia project host, e.g. de.wikipedia, es.wikipedia, commons.wikimedia. Used by top and edits.
country	string	DE	ISO 3166-1 alpha-2 code for by_country mode: US, FR, GB, IN, …
dates	array	["2026-07-26"]	One or more YYYY-MM-DD days. One API request per date.
maxItems	integer	200	Row cap across all dates (max 2000).

{
  "mode": "top",
  "project": "en.wikipedia",
  "dates": ["2026-07-26", "2026-07-25"],
  "maxItems": 200
}

---

Important: the data lag (and automatic walk-back)

Wikimedia does not publish trend data in real time.

• Pageview data (top, by_country) lags ~1–2 days. Ask for today and you will usually get a 404.
• Edit data (edits) lags much longer — it comes from a monthly mediawiki-history snapshot, so the newest available day is typically 3–6 weeks in the past.

This Actor handles that for you. If the requested date returns no data, it automatically walks back one day at a time (up to 5 days for pageviews, up to 70 days for edits) until it finds the most recent published day, and stamps the row with the date it actually used. You never get an empty run just because you asked for "yesterday". The run log tells you exactly which date was substituted.

If you need a guaranteed-fresh date, ask for two days ago for top/by_country, and last month for edits.

---

Use cases

• Trend research — see what the public was actually curious about on any given day, with no keyword guessing and no sampling. Great for spotting cultural moments before they hit mainstream analytics.
• SEO — Wikipedia rankings are a free proxy for real informational search demand. If an article is climbing the top-1000, the topic has genuine search volume behind it, and Wikipedia usually outranks you for it — which tells you exactly which SERP you'd be fighting for.
• Content planning — mine the daily ranking for topics with proven interest, then check by_country to see whether that interest is global or local before you commission a piece.
• News monitoring — the edits mode surfaces pages under heavy revision, an early signal of breaking or contested events. Combine with top to separate "people are reading about this" from "people are arguing about this".

---

Pricing

Roughly $0.50 per 1,000 results. Runs are fast and lightweight (512 MB, one API call per date), so a full 1,000-article daily ranking costs about half a dollar. No proxy costs, no API subscription — the Wikimedia Analytics API is free and its data is CC0.

---

Related Actors

• Wikipedia Pageviews Scraper — per-article daily/monthly view history for titles you specify
• Wikipedia Featured Scraper — featured article, picture of the day, on-this-day events
• Google Trends Scraper — search-interest curves to cross-check Wikipedia trends
• Wikipedia Scraper — full article content, summaries, infoboxes and links

---

Notes & compliance

Data comes from the public Wikimedia REST Analytics API. It is aggregate and anonymous — no personal data is collected, and country-level counts are deliberately rounded by Wikimedia. Content is licensed CC0 / CC BY-SA; attribute Wikipedia when you republish. The Actor sends a descriptive User-Agent as Wikimedia's policy requires.

Redirects, Main_Page, and Special: pages appear in the raw ranking exactly as Wikimedia publishes them — filter them out downstream if you only want article content.