Website visit & organic traffic keyword data

What is Similarweb?

Similarweb is a digital intelligence company that estimates web and app performance for millions of sites and apps worldwide using algorithmic modeling rather than self reported analytics tags. Because the estimates come from Similarweb's own measurement rather than a company's internal dashboards, they offer an unbiased, objective view of real world web and app performance, useful for studying digital assets researchers otherwise have no visibility into.

‍

What academic researchers should know about Similarweb's website traffic and keyword data

On Dewey, Similarweb contributes data on website visits, desktop search keywords, and app engagement. The keyword dataset covers long tail search terms and the clicks they drive per landing page, updated monthly with coverage from 2023 to present. App engagement is available at the daily level for tracking usage patterns over time. Because Similarweb's numbers come from algorithmic estimation, values include decimal places rather than whole numbers, and the underlying data returns exact figures even for smaller sites below the visibility threshold shown on Similarweb's own platform, which is useful for aggregating small domains across time periods or in bulk. Similarweb data can also be joined with points of interest data on Dewey, connecting online behavior to offline location activity.

‍

Why academic researchers choose Similarweb on Dewey

Website and app traffic is one of the few windows into company performance that does not depend on what a company chooses to disclose, which makes it a useful alternative data source for studying digital markets, competition, and consumer attention at scale. The algorithmic, panel based methodology behind Similarweb's estimates is designed to stay consistent across millions of sites and apps, so comparisons across brands, countries, and time periods rest on the same underlying measurement rather than each company's own self reported analytics. Paired with points of interest data also on Dewey, Similarweb's traffic data supports research that connects a brand's online reach to its physical footprint.

‍

Similarweb academic research ideas and use cases

Financial performance and market efficiency research has already used this exact kind of data at scale. A 2025 study in The Accounting Review by researchers at Stanford and UC Berkeley used Similarweb traffic data covering the 100 largest firms in each of 11 major industry sectors, finding that website visits and page views predict quarterly financial results and that markets are slow to price in that information, particularly for companies whose sites directly drive sales.

Internet structure and platform concentration research can also draw on this kind of data. A recent analysis of global web usage used Similarweb traffic estimates across more than 250,000 websites to show how a small number of top domains, mostly search engines, social networks, and streaming platforms, capture a disproportionate share of all web traffic.

Marketing and SEO research can use the desktop search keyword data to study how search behavior shifts around product launches, news events, or competitive moves, and how organic versus other traffic sources respond differently.

Mobile ecosystem and app economy research can use the daily app engagement data to study usage patterns, retention proxies, and competitive dynamics across app categories over time.

Omnichannel consumer behavior research can pair Similarweb's website and app traffic with points of interest data on Dewey to study how online attention and offline visits move together for the same brands.

‍