International Activity Between the Lines: How Text Data Helps Measure the Level of Company Internationalization
On May 15, 2026, a study dedicated to measuring the level of company internationalization based on open text data was presented at the research seminar of the "International Academic Cooperation" project. The report was presented by Nikita Matkin, Aleksei Smirnov, and Mikhail Usanin; the study was co-authored by Iuliia Naidenova, Sofia Paklina, and Mariya Molodchik.
The study focuses on how corporate reports, non-financial disclosures, and digital footprints can be used to assess the international activity of a business when traditional indicators - such as export revenue, foreign sales, assets, or overseas branches - are not fully available.
The authors proposed a method for constructing a text-based company internationalization index using open data. The method is based on a dictionary of 137 terms and phrases related to exports, foreign economic activity, foreign markets, foreign direct investment, cross-border transactions, international partnerships, and global value chains. The method involves searching for mentions of these terms in open sources and subsequently transforming the resulting text indicators into a company internationalization index.
At the empirical stage, the researchers constructed an internationalization index for Russian companies for the period 2007–2021 using data from three sources. The analysis included 90,446 annual reports of companies from the website of the Interfax-CRKI Corporate Information Disclosure Center, 1,600 non-financial reports from the website of the Russian Union of Industrialists and Entrepreneurs (RSPP), and 74,528 search prominence results of companies obtained via the Yandex.Search API.
To test the validity of the constructed indices, the authors compared them with the companies' share of export revenue. A positive relationship was identified for all three sources: companies with a higher share of export revenue mention international activity more frequently in public texts and digital footprints.
The results of the study show that open text data can complement traditional financial indicators when assessing the international activity of a business. The proposed index does not replace direct indicators of internationalization but allows for the comparison of companies and industries, as well as the analysis of changes over time in situations where financial data are incomplete or difficult to access.
The project "Internationalization of Companies from Developing Countries: The Role of Intellectual Resources in Response to Exogenous Shocks" is carried out under the framework of the Joint Basic Research Program "International Academic cooperation".
#international_academic_cooperation
