AI as the new fourth “P” of marketing: A statistical and machine learning modelling approach to marketing performance

Abstract

This paper undertakes a statistical and machine learning modelling analysis to explore how Artificial Intelligence works as a new ‘fourth P’ for marketing and how it alters the classical formula for a data-driven process. This paper uses a mathematical statistical method to analyse business-level data from 281 companies in Saudi Arabia. This research draws upon theories such as Resource-Based View, Dynamic Capability Theory, and TOE framework. Through correlation analysis, regression analysis, and mediation analysis conducted using Random Forest analysis, Gradient Boosting analysis, and Artificial Neural Network (ANN) analysis, the impact of AI on marketing performance as a function of product innovation, pricing accuracy, channel flexibility, and promotional personalization was objectively quantified. The findings revealed a robust correlation between the implementation of AI and marketing performance (β = 0.842, R-squared = 0.72, p < 0.001), and good predictability of marketing performance using machine learning analysis (R-squared ANN analysis = 0.94). The SHAP evaluation disclosed that the factor of promotional personalization has the highest influence on marketing performance. The findings effectively verify that AI functions as a strategic and measurable integrator of the marketing mix. Through a strategic integration of mathematical findings and strategic marketing principles, AI is positioned as the new fourth ‘P’ of marketing.