Identificación de técnicas emergentes de inteligencia artificial para gestión de fricciones en experiencia del cliente en canales digitales del sector Bancario Colombiano

Abstract

This research project explores mechanisms for artificial intelligence (AI) to improve the customer experience in the digital channels of the Colombian banking sector. The study stems from a common problem in financial institutions: digital friction. This friction manifests as difficulties navigating digital channels, long wait times, problems understanding or distrust of automated processes, and presumably negatively impacts customer satisfaction and loyalty. Using a quantitative approach, a structured survey with a five-point Likert scale was conducted with actual users of digital banking services. The results were analyzed using frequencies, cross-tabulations, interpretive graphs, and 3D visualizations to identify patterns for consideration. Among the most important findings, it was demonstrated that the perception of automation, user age, interaction with chatbots, and the ease and frequency of use of digital channels are directly related to the level of customer satisfaction. The discussion was further enriched by referencing frameworks such as Google's HEART model, which allowed for the evaluation of key elements such as user happiness, customer engagement or effort, adoption, retention, and friction indicators (transactional error rate). Based on this analysis, guidelines and ideas are proposed to optimize the digital experience, such as improving empathy in automated channels, implementing more human-centered metrics to measure the experience, and advancing AI learning within the Colombian banking context. In conclusion, this study not only provides practical evidence on user behavior in relation to digital channels but also offers methodological and strategic tools for banks to reduce friction, increase satisfaction, and build stronger relationships with their customers through technology.