Measuring Digital Technology Adoption - A Novel Text Mining Approach

Due to the relevance of digital technologies for economic prosperity, measuring their adoption is of particular importance. We show that the availability of firm website data and advances in text mining enable the creation of real-time and large-scale estimates for digital technology adoption. In order to learn the abstract definition of digitalisation, we train a random forest regression model on labelled newspaper articles. The trained model is then applied to firms’ website content to obtain a firm-level indicator of digitalisation. Plausibility checks confirm the link to established digitalisation indicators at the firm and sectoral level, as well as for firm size classes and regions. Lastly, we illustrate the indicator’s validity by analysing the link between digitalisation and firm resilience during the COVID-19 crisis to produce findings that are consistent with the related literature.

Axenbeck, Janna and Patrick Breithaupt (2026), Measuring Digital Technology Adoption - A Novel Text Mining Approach, Journal of Economics and Statistics (Jahrbücher für Nationalökonomie und Statistik)