Where to Find Daily Wikipedia Page View Data for Stocks

Daily Wikipedia page-view counts per equity, with a rolling z-score that measures unusual attention, from Alphanume.

Retail attention moves stocks, and one of the cleanest public proxies for attention is how often people read a company's Wikipedia page. The raw Wikipedia pageview API exists, but mapping pages to tickers, aligning them to trading dates, and turning the counts into something comparable across names is a meaningful amount of plumbing.

Alphanume does that plumbing and publishes the result as the Wikipedia Views dataset.

What the dataset is

Wikipedia Views provides daily Wikipedia page-view counts per equity, with a rolling 30-day mean, standard deviation, and z-score that measures how unusual the current level of attention is. It is cursor-paginated and ordered by date descending and then ticker ascending. The z-score is the key column: it tells you whether today's attention is normal or a genuine spike.

Fields

• ticker, name, date
• views: that day's page-view count
• avg_30d, std_30d: rolling 30-day mean and standard deviation
• zscore_30d: how many standard deviations above or below normal today sits

What you can do with it

• Build attention-based factors.
• Detect anomaly days where the z-score spikes.
• Contextualize event setups by how much public attention a name is drawing.
• Backtest attention-regime strategies.

How to access it

Wikipedia Views is a Pro-only dataset, so requests require a Pro key (free-tier keys receive a 403 PRO_TIER_REQUIRED response). Query one ticker on one date, or filter on the z-score:

curl "https://api.alphanume.com/v1/wikipedia-views?ticker=AAPL&date=2026-02-06&api_key=alp_your_key"

See the API documentation, review Alphanume datasets to compare tiers, or grab a free Alphanume API key to explore the free catalog before upgrading.

From raw pageviews to a usable signal

Wikipedia publishes pageview counts, but turning them into a trading signal takes work: mapping pages to tickers, aligning counts to trading dates, and normalizing them so a mega-cap and an obscure name are comparable. Raw counts on their own are nearly useless cross-sectionally because the scale differs by orders of magnitude between companies.

Alphanume does that processing and adds the key column, a rolling 30-day z-score that measures how unusual today's attention is for each name. Because the z-score is normalized against each name's own history, a spike on a quiet company and a spike on a heavily-followed one are directly comparable, which is what makes the dataset usable as a factor.

Frequently asked questions

What does the Wikipedia Views dataset measure?

Daily Wikipedia page-view counts per equity, plus a rolling 30-day mean, standard deviation, and z-score that captures how unusual the current level of attention is.

Why use the z-score instead of raw views?

Page-view counts vary enormously across companies, so raw counts are not comparable. The z-score normalizes each name against its own 30-day history, making spikes comparable across the universe.

Is it free?

No. Wikipedia Views is a Pro-only dataset. A free-tier key returns a 403 PRO_TIER_REQUIRED response.

What fields are included?

ticker, name, date, views, avg_30d, std_30d, and zscore_30d.

How is it ordered?

It is cursor-paginated and ordered by date descending then ticker ascending, returning has_more and next_cursor for large pulls.