Evolution of digital marketing campaigns with artificial intelligence and machine learning: Analysing success prediction capabilities

Abstract

This study aims to investigate the predictive capabilities of machine learning algorithms in forecasting the success of digital marketing campaigns. In addition, the study aims to evaluate the performance of machine learning algorithm classification models and to determine which classification model is more effective in making this prediction. In this direction, a classification analysis was performed with machine learning algorithms using a dataset of 10,001 digital marketing campaigns obtained from the Kaggle platform. The study's theoretical background is based on Attribution Modelling, the Technology Acceptance Model, Incremental Response Modelling, and the Diffusion of Innovations Theory. As a result of the analysis, it was revealed that the success prediction capabilities of machine learning algorithms for marketing campaigns are pretty high. In the study comparing the success prediction abilities of machine learning models, the Gradient Boosting model demonstrated the highest success prediction ability (93.31%). In comparison, the Logistic Regression model has the lowest predictive success rate (53.36%). The findings revealed that machine learning algorithms should be more widely incorporated into the marketing literature and that businesses can run more successful and efficient campaigns by utilising machine learning models in their marketing efforts.