Modern data analytics equips businesses to make data-driven decisions by revealing patterns and insights that enhance strategic planning, operational efficiency, and process optimization. Its applications encompass personalized marketing through customer segmentation, predictive modelling for fraud detection, and enhancing security. A significant methodology in this realm is the Cross-Industry Standard Process for Data Mining (CRISP-DM), where the Business Understanding phase aims to ensure data science projects align with overarching business goals. However, challenges arise when these business objectives are ambiguous, ill-defined, or evolving. The complexity of data analytics projects underscores the need for domain expertise and robust collaboration between data scientists, business stakeholders, and domain experts. The imperative is to bridge the technical and business perspectives, manage expectations, and define project scopes. The short paper at hand addresses the question how data analytic goals can systematically align with business objectives in data science projects. By incorporating methods from Enterprise Architecture Management, we propose a structured approach for goal determination in data science projects, ensuring business and data mining objectives are seamlessly integrated.