A Methodological Prototype for Assessing SME Digitalization through Survey Data, Web Scraping, and Human-in-the-Loop Validation

Abstract

Reliable assessment of the digitalization of small and medium-sized enterprises (SMEs) remains a methodological challenge, as existing studies rely predominantly on survey-based instruments and self-reported information. This article proposes a methodological prototype for evaluating SME digitalization by combining automated web scraping, survey data, and Human-in-the-Loop (HITL) validation. Using Bulgaria as a case study, the proposed approach integrates publicly available digital indicators with manual expert verification to improve the accuracy and reliability of digitalization assessments. The empirical validation is based on a sample of 100 enterprises, representing 10% of the overall dataset, selected according to their online presence across major social media platforms (Facebook, Instagram, and LinkedIn). The multi-stage verification process compares declared digital practices with enterprises’ observable digital footprints, enabling the identification of discrepancies between self-reported and actual online activities. The study contributes to the methodological discussion on measuring SME digitalization by demonstrating how multiple complementary data sources can improve the assessments. The proposed framework may support future research and policy evaluation by providing a more reliable approach to assessing SME digitalization.