A Systematic review on the usage of Deep Learning techniques in the marketing process

Abstract

This study aims to present a systematic review of research articles published between 2021 and 2024, found from the Google Scholar repository, and focusing on the application of deep learning (DL) algorithms in marketing. Two key research questions were examined: The first one identifies marketing activities and tactics within Kotler’s Marketing Process that have adopted deep learning solutions, while the second one explores the specific deep learning implementations employed for marketing purposes. Findings highlight the growing integration of deep learning across all steps of the marketing process, particularly in customer behaviour analysis, for the purpose of optimising promotion strategies. The field has moved from the use of basic deep learning implementations to the application of advanced, multi-layered deep learning systems that can support more dynamic, data-driven and personalized marketing approaches. Increasingly, hybrid models of CNNs, RNNs, LSTMs, Transformers and GANs are being used in real-world applications to analyse large-scale data and support real-time decision making, as well as to improve customer engagement. As for the future work, the search methodology should be refined further to incorporate more specialized keywords to expand the search and include a wider range of relevant studies, and thus, to further identify emerging trends and critical research gaps in the evolving landscape of deep learning enhanced marketing.