Predicting Business Growth

Forecasting business growth is a persistent and significant challenge in economics and management research. Despite extensive study, empirical research has struggled to identify consistent drivers of business growth, with most conventional models delivering very low predictive accuracy due to substantial heterogeneity across firms, industries, and institutional contexts. This report directly addresses these challenges by providing a clear, evidence-based assessment of the difficulties in predicting growth and evaluating how recent methodological advances may be redefining the field.

Drawing on leading academic and grey literature, this review critically examines both traditional econometric approaches and emerging machine learning techniques for modelling business performance. It synthesises findings across growth metrics, predictors, and modelling strategies, identifying the limitations of established methods and highlighting the predictive accuracy achieved by newer, data-intensive approaches. The report offers an accessible yet rigorous overview of current advancements in business growth modelling.

Importantly, the report extends beyond methodological comparison to consider practical implications for research, policy, and decision-making. By emphasising the complementary strengths of econometrics and machine learning, as well as the trade-offs between interpretability and predictive power, it presents a framework for more robust, transparent, and actionable growth forecasting. This report offers timely and essential insights for those seeking to understand not only whether firms grow, but also how growth can be anticipated and supported.

Download the Executive Summary and Full report via the buttons on the right