QualityKiosk Technologies Drives 45% AI-Led Lead Conversion for a Premier Hong Kong Insurer

While the client was shifting from a rule-based engine to an AI-based Insurance product Engine, QualityKiosk felt that we would have to move from a Quality Engineering approach to a Quality Intelligence Approach and model to ensure that the AI-led algorithms were validated. The current model did not have documentation around the rules or specific scenarios to be covered. Hence, the canvas for QualityKiosk Technologies was vast and fuzzy.

• Persona analysis: We collected data on the kind of personas the company was serving, and based on the segmentation, we outlined the parameters of a persona-based selling strategy.

• Personalized product recommendations: We leveraged the mind map modeling for all kinds of customer segmentation, agent segmentation, and the parameter of the products to be offered to create a comprehensive test set.

•  Continuous learning model: Basis the result of the learning model on the above recommendation, the data was again fed back to the learning system to get more persona-to-product mapping data. This was an iterative process where the test set was getting richer. It took multiple testing cycles to fine-tune our strategy and maximize its effectiveness.

• Continuous monitoring: Even after the go-live, persona-based selling is a continuous process, and the model is getting validated.