Customer Voice Impact Matrix (CVIM): Structuring Product Feedback with BERT Sentiment Analysis and Topic Modeling

The Customer Voice Impact Matrix (CVIM) is a framework that combines sentiment analysis and topic modeling to convert unstructured customer reviews into actionable insights. A BERT-based model is used to assess the sentiment polarity of each review, while BERTopic, leveraging sentence embeddings, identifies key themes within the text. These topics are plotted on a two-dimensional matrix: the X-axis indicates average sentiment polarity, and the Y-axis represents how frequently each topic appears. The resulting four-quadrant layout namely Fix Urgently, Promote Strength, Hidden Gem, and Minor Issue, offers a clear basis for evaluating product features and guiding communication strategies. The framework is applied to reviews of an eco-friendly toothbrush, demonstrating how CVIM helps businesses detect critical issues, highlight valued attributes, and recognize overlooked strengths, supporting more targeted and effective AI-driven marketing decisions.