Innovation threshold: A decision-making framework for micro-enterprises

This study aims to develop an evidence-based decision-making framework for innovation in micro-enterprises in India. It focuses on identifying and analysing key innovation variables and their interactions using decision tree analysis, grounded in the resource-based view. About 435 responses were collected from micro-enterprises featured in four trade events in India. Six independent variables such as Open Innovation, Strategic Positioning, Technology Progression, Ideation, Innovation Orientation and Innovation Drive were used to predict Innovation Outcome. Python's scikit-learn was used for decision tree modelling, feature importance analysis and rule extraction. A heatmap was prepared to interpret results and threshold effects. Open Innovation emerged as the most influential predictor. Strategic Positioning and Technology Progression showed clear threshold effects. Results confirmed that high innovation outcomes depend on collective strength across variables rather than isolated performance. This study introduces a novel integration of path analysis and decision tree modelling to develop threshold-based decision rules for micro-enterprises.