I Stole a Wall Street Trick to Solve a Google Trends Data Problem
This title could be clearer and more informative.Try out Clickbait Shieldfor free (5 uses left this month).
Google Trends data is normalized per-country, making cross-country comparisons impossible by default. After exploring and debunking naive approaches (like using worldwide subregion ratios), the author borrows a concept from finance: building a basket of representative search terms (similar to a stock market index like the S&P 500) to create a common benchmark. By calculating absolute search volumes using internet user population data and a basket of popular terms, the author discovers that when measuring motivation as a proportion of total search volume, all scaling factors cancel out — simplifying the final calculation while validating the methodology's soundness.
Table of contents
The Maths Behind Google Trends NormalisationTaking Inspiration from the Stock MarketTracking the Market Through IndicesWhy you’d estimate in this big data worldApplying ETFs to Google Trends DataMaking Search Data Truly Comparable Across CountriesWhen the maths simplifies itselfSort: