Presented By O’Reilly and Cloudera
Make Data Work
September 11, 2018: Training & Tutorials
September 12–13, 2018: Keynotes & Sessions
New York, NY

Anxiety at scale: How Investopedia used readership data to track market volatility

Masha Westerlund (Investopedia)
4:35pm–5:15pm Wednesday, 09/12/2018
Data science and machine learning
Location: 1A 06/07 Level: Beginner
Secondary topics:  Financial Services, Text and Language processing and analysis
Average rating: *****
(5.00, 2 ratings)

Who is this presentation for?

  • Data scientists, business managers, and product leads

What you'll learn

  • Learn how Investopedia is harnessing user data to build educational tools for users


While companies have thought of a myriad of creative ways to harness user data for sales, marketing, and products, it’s more difficult to consider how that data can be used to share knowledge back with the users themselves.

Investopedia has tens of thousands of evergreen articles covering everything from personal finance to technical trading and draws over 20 million unique readers a month. This scale in both content and traffic allows the company to track changing interest in financial topics across the country over time. Data science in digital publishing often revolves around using data to power content recommendation engines, guide editorial strategy, and target ad sales. However, because Investopedia is primarily an educational resource, rather than only using data for those purposes, it also wanted to harness readership data to further educate its visitors.

Masha Westerlund discusses the creation and thought process behind the Investopedia Anxiety Index, a novel, public, and free tool hosted on the site that distills raw proprietary user visits into an index that tracks market anxiety. The index tracks trends in visits to certain “anxiety-reflecting” words such as “short-selling” and “default,” deseasonalized and normalized by overall traffic. Despite including only visits to Investopedia and no actual market data, the Anxiety Index is remarkably similar to the CBOE’s VIX “fear index,” which measures implied volatility on the S&P 500. Even more intriguing, the Anxiety Index occasionally shows movement prior to large VIX or market movements, suggesting that it is sensitive to early changes in investor sentiment.

Join Masha to learn about the Anxiety Index as well as other initiatives at Investopedia to use data to help educate readers.

Photo of Masha Westerlund

Masha Westerlund


Masha Westerlund is director of the data science team at Investopedia, where she works to answer questions such as “What can Investopedia’s readership tell us about current market sentiment?” and “What financial concepts are most interesting to American investors, from Wall Street to Silicon Valley?” Masha holds a PhD in cognitive science from New York University.