Wikipedia Pageviews: daily views per article, any language

Daily or monthly page views of Wikipedia articles in any language, per article per day, with the change against the previous day, from the Wikimedia statistics API. Give titles or article URLs, choose the period, device and human or all traffic. Up to 500 articles a run. Pay per day.

Why this scraper

• Only delivered results are charged. Inputs that fail come back as clear error records at no cost.
• Straight from the Wikimedia Foundation's own statistics API, the source of the numbers on every Wikipedia page-views tool. Measured on the platform: 14 days each of four articles across three language editions, 56 rows, in 4.5 seconds for a seventh of a cent.
• One row per article per day, newest first, with the previous day's views and the percentage change already computed, so a spike is visible without a spreadsheet formula.
• Articles can be given as plain titles, as de:Berlin for another language, or as the article URL copied from the browser; mobile URLs and percent-encoded titles are understood.
• Human readers only by default (agent: user), so bots and crawlers do not inflate the numbers; desktop and mobile can be split; monthly granularity for years of history in a compact table.

Who this is for

Put articles in articles (up to 500 per run), set days (default 30, up to 3,650), and optionally project for titles without a language, granularity, access and agent. Built for trend and attention research (which topics, companies, people or products are rising), PR and marketing measurement around a launch or a news event, SEO and content teams comparing interest across languages, and anyone who needs a time series of public attention without Google Trends.

Who this is not for

Page views count article visits, not searches; a topic that people read about is not necessarily one they buy. The API has data from July 2015 and ends at the previous day; today's views come tomorrow. A title must match the article exactly as Wikipedia names it (redirects are separate pages with their own, usually small, counts); a title with no data in the period comes back as a free NOT_FOUND row. The Wikimedia API is rate-limited per client and asks for polite use; the actor identifies itself and stays far below the limit.

Input fields

Field	Type	Required or default	What it does
articles	list of text	required	One per row, up to 500: a title (Bitcoin), a title with language (de:Berlin) or an article URL (https://en.wikipedia.org/wiki/Bitcoin).
project	text	en.wikipedia	Used for titles without a language, for example en.wikipedia, de.wikipedia, nl.wikipedia.
days	number	30	How many days back from yesterday (the API has data up to the previous day). Up to 3650.
granularity	text (daily, monthly)	daily	daily or monthly.
access	text (all-access, desktop, mobile-web, mobile-app)	all-access	all-access, desktop, mobile-web or mobile-app.
agent	text (user, all-agents)	user	user for human readers only, all-agents to include crawlers and automated traffic.

Input example

{
    "articles": [
        "Bitcoin",
        "Python (programming language)",
        "https://de.wikipedia.org/wiki/Berlin"
    ],
    "project": "en.wikipedia",
    "days": 30,
    "granularity": "daily",
    "access": "all-access",
    "agent": "user"
}

Output example

Field	Type	What it holds
article	text	The article title with spaces, as it reads on Wikipedia.
project	text	The wiki the views belong to, for example en.wikipedia or de.wikipedia.
date	text	The day (YYYY-MM-DD) or, with monthly granularity, the month (YYYY-MM).
views	number	How many times the article was viewed that day, as Wikimedia counts it.
previousViews	number	The views of the day before in the delivered series; empty on the oldest day.
changePercent	number	The change against the previous day in percent.
granularity	text	daily or monthly, as requested.
access	text	all-access, desktop, mobile-web or mobile-app, as requested.
agent	text	user (human readers) or all-agents (including crawlers), as requested.
url	text	The article on Wikipedia.

Error codes: INVALID_ARTICLE, NOT_FOUND, BLOCKED.

One delivered row looks like this:

{
  "article": "Python (programming language)",
  "project": "en.wikipedia",
  "date": "2026-09-24",
  "views": 6088,
  "previousViews": 5697,
  "changePercent": 6.86,
  "granularity": "daily",
  "access": "all-access",
  "agent": "user",
  "url": "https://en.wikipedia.org/wiki/Python_(programming_language)",
  "status": "ok"
}

Related actors from steadydata

• google-trends-compare: search interest for the same topics, from Google
• google-news: the news that caused a spike
• wayback-history: how a page changed over time

Pricing

Pay per event: one day-listed event per delivered result. No charge for inputs that fail, no separate platform-usage surcharge.

Free Apify plan: this actor delivers up to 25 rows per run for accounts on the Apify free plan, and then stops with a message. That limit is set by us, not by Apify. It exists so the actor keeps paying for itself for the people who do pay. Any paid Apify plan runs it at full size, billed per delivered row, with failed rows never charged.

Reviews: if this actor saves you time, a short review on this page is the one thing that helps most. Ratings are what other buyers look at first, and we have no other way to ask.

FAQ

Is personal data collected? No. The rows are aggregate view counts of public articles.

How do I compare topics across languages? Give the same subject as en:Bitcoin, de:Bitcoin, ja:ビットコイン; each comes back as its own series with project on every row.

Can I get years of history without thousands of rows? Yes: set granularity to monthly and days to 3650; a decade becomes 120 rows per article.

What is the difference between user and all-agents? user counts human readers as Wikimedia classifies them; all-agents adds crawlers and automated traffic, which can be a large share for some articles. Use user for attention research.

Why does a title come back as NOT_FOUND? Because the API knows the exact page title only: Python is a different page from Python (programming language). Copy the title or the URL from the article itself.

What does a run cost when an article is unknown? Nothing. NOT_FOUND, INVALID_ARTICLE and BLOCKED rows are free; only delivered days are charged.

What happens when the source changes? Sources change from time to time; that is the nature of this work. The actor is monitored daily and fixed fast, and while it is broken you are not charged, because only delivered results cost anything.