An analytics-driven method for building ethical customer digital twins using neuromarketing and social media data

As markets become increasingly digitalised, understanding the emotional and cognitive drivers of consumer behaviour is essential for developing responsive and targeted marketing strategies. This study contributes to the development of Customer Digital Twins (CDTs) by integrating neuromarketing techniques with social media analytics, positioning the work at the intersection of data science and behavioural psychology. The proposed framework combines electroencephalography (EEG) recordings with sentiment analysis of user-generated content to capture complementary dimensions of consumer emotion and decision making. An experimental design was implemented in which participants’ emotional responses were monitored via EEG during an online shopping task, followed by an ethical priming intervention and sentiment-based analysis of post-task responses. The study focuses on a fast-fashion e-commerce setting in which ethically framed information is introduced to examine how value-laden cues reshape cognitive and emotional states. The results show that EEG-derived indicators, including engagement, workload, and emotional valence, and text-based sentiment measures provide partially complementary views of consumer states, while also highlighting significant challenges in their direct integration. The joint use of these modalities offers richer descriptive insight into customer motivation and behaviour, but simple fusion strategies exhibit limited predictive power under real-world, asynchronous conditions. The findings identify both technical and ethical challenges associated with multimodal consumer modelling and establish a foundation for future work on temporally aligned and dynamically adaptive customer digital twin architectures. A proof-of-concept modelling exercise further explores the feasibility of using combined neuro-sentiment features to predict valence-related indices, serving as an initial diagnostic decision analytics layer rather than a fully realised digital twin. • Integrate brain signal data and social media text to construct ethical customer digital twins. • Model emotional persistence to enhance the analysis of customer decision behaviour. • Predict emotional states using fused brain and sentiment features. • Develop a reproducible analytics pipeline for customer digital twin construction. • Support ethical marketing decisions through emotion-aware customer modelling.