Gaining new insight into online customer behavior using AI

Customer satisfaction is a key success factor for any business. Moise Convolbo highlights the process to capture relevant customer behavioral data, cluster the user journey by different patterns, and draw conclusions for data-informed business decisions.

Customer satisfaction indicators, based on how customers behave with a service, are important factors of business output. For web services, especially from a product perspective, the way customers’ interact is in direct proportion to their level of satisfaction with the service. With the advent of big data, the information collected about customer interactions in a service provides valuable insights for decision makers. One of the biggest business challenges is data-informed decision making. In the case of web services, such as an ecommerce platform, a travel or hotel reservation site, social media, or online dating site, customers follow a path to achieve their desired goal. Accomplishing this goal leads to conversion. The information gathered through this customer journey constitutes a tremendous opportunity to ensure the customer satisfaction with a web service. Moise Convolbo presents a model for minimizing customer dissatisfaction, improving the quality of web services, using an internal tool from Rakuten called the PathFinder—an autohealing mechanism that allows stakeholders to detect customer struggles on a web service, add the business impact to these issues, then provide feedback on the implemented solution. This internal AI-Kaizen platform for path analytics in web services has assessed the performance of six different web services within Rakuten. PathFinder is able to automatically find and cluster the customer struggles. As a result, the company can detect and solve product issues that are attributed to customer behaviors leading to abandonment and lost revenue.