What are some advantages and challenges of using a logic-driven analytics process rather than follow a data-driven analytics process?

Logic-Driven analytics is one based on experience, knowledge and logical relationships of variables and constants connected to the desired business performance outcome situation. This approach differs from data-driven analytics, where the story in the data drives the model building and analytics outputs. The goal of a logic-driven analytics is to put variables and constants together to create a model that can explain the past and predict the future. Such model building calls for a deep understanding of the business and the relationships of business variables and constants that impact a business performance outcome (KPI). 1. What are some advantages and challenges of using a logic-driven analytics process rather than follow a data-driven analytics process? 2. From your past and current experience, which approach might work better for a business analytics project and why?