In the context of e-commerce, reliable and fitting recommendations are of paramount importance. They help users make faster and more informed purchasing decisions while enhancing satisfaction, engagement, and loyalty. However, delivering such personalized and effective recommendations presents several challenges. Most recommendation systems (RS) rely on sentiment analysis, user goal guessing, and predictions, each of which comes with its own difficulties. We introduce GoRS (Goal-oriented RS), a neuro-symbolic, user-centric, and goal-oriented RS in the Do-it-Yourself (DIY) context. GoRS generates a step-by-step guide tailored to the user’s previously entered goal and recommends fitting products like saws, drills, and sanding machines according to each step and the specific restrictions of a user’s goal. Therefore, GoRS utilizes the flexible structure of Knowledge Graphs (KGs) and the reasoning capabilities of Large Language Models (LLMs), which makes it a novel neuro-symbolic RS. We evaluated our approach through a between-subject study. Users were asked to create and review their step-by-step guide and recommended products based on a predefined DIY goal and complete a questionnaire investigating, among other things, user satisfaction, workload, and efficiency. The results indicate a significant correlation between mental workload, efficiency, user satisfaction, and user experience—as mental workload decreases, efficiency improves, and user satisfaction and experience increase. Based on our findings, we suggest adopting our novel approach and encouraging the integration of goal-oriented RSs to enhance user satisfaction and experience, efficiency, and trust while providing more intuitive, personalized, and effective recommendations.