EXPRESS: Measuring Business-Process Distance with Large Language Models

Traditional measures of cultural and contextual distance between countries typically rely on arithmetic or geometric comparisons based on survey-based cultural values or country-level statistics. While useful for explaining cross-national variations in behavior, organizational practices, and trade, these measures often lack the granularity and real-world applicability required by international business practitioners. This paper takes an exploratory approach, investigating whether the knowledge embedded in large language models (LLMs) can be used to assess differences in industry-specific business processes across countries. Rather than relying on aggregated national indices, the study examines how LLMs might reveal nuanced, domain-specific patterns in business practices. The goal of this proof of concept is not to replace existing distance measures, but to explore a complementary, AI-driven perspective that could yield more flexible, detailed, and actionable insights for managing cross-border operations.