AI (artificial intelligence) and ML (machine learning) technology proliferation has dramatically changed the marketing field with the use of predictive analytics, allowing organizations to decode, anticipate and control consumer purchasing behaviour. This doctoral dissertation provides an in-depth research study on how predictive analytics (powered by AI) systems function in a marketing context, and the many different forms they can take to influence the decision-making processes and outcomes of consumers. The current research was conducted through an integrated review of the peer-reviewed literature published between 2019 and 2025 as well as by reviewing the relevant theoretical frameworks (Theory of Planned Behaviour; Technology Acceptance Model; Information Processing Theory) to synthesise the empirical evidence found in e-commerce, retail, and digital marketing environments. This study shows that many different types of advanced algorithms can produce excellent results at predicting what consumers will purchase (the algorithms include things like gradient boosting such as XGBoost and CatBoost, deep learning architectures such as LSTM, and various forms of recurrent neural networks). Previous research found some predictive algorithms produced results with F1 scores over 0.92 when working on consumer purchase data from large amounts of e-commerce sales transactions. AI personalisation can improve consumer engagement and satisfaction but also create a paradox between privacy and personalisation, resulting in consumers losing trust when they feel that they are being intruded upon by the predictive algorithms. The ethical, regulatory, and transparency considerations surrounding the use of consumer data for the purposes of advertising and marketing to consumers must be examined along with the technical aspects of using predictive algorithms to ensure accuracy and consistency in their application, specifically considering regulations such as the GDPR and CCPA. The findings of this research carry significant implications for India's Viksit Bharat 2047 vision — the Government of India's flagship programme to transform India into a fully developed nation by the centenary of its independence. As India's digital economy grows rapidly, with over 900 million internet users projected by 2025 and e-commerce revenues expected to reach USD 350 billion by 2030, AI-driven predictive analytics in marketing will play a central role in shaping the consumer economy. Responsible, transparent, and inclusive implementation of predictive marketing systems in Tier Two and Tier Three cities will help to accelerate digital financial access, empower rural entrepreneurs, and support Make in India objectives. Frameworks developed in this research for ethical AI governance and consumer trust-building are directly applicable to the policy context of Digital India, the National AI Strategy (NITI Aayog), and the India AI Mission, each of which identifies AI as a critical enabler of equitable, data-driven economic growth aligned with Viksit Bharat 2047. The evidence suggests that the best uses of predictive analytics in marketing are those combining advanced algorithms and appropriate consumer-centric data governance. The conclusion provides future research options regarding the application of causal methods of inference, the governance of generative AI in consumer analytics, and the cross-cultural elements of predictive marketing acceptance, including a specific focus on emerging markets such as India.