Nowadays online social networks represent important platforms for communication and influence. Engagement metrics such as likes are crucial for measuring content popularity and user interaction. However, traditional like counts offer limited information since they do not take into account several factors that may increase or decrease the significance of each like. This paper proposes a novel method for weighting social media likes to derive a score that reflects the actual engagement of a post. Our approach considers multiple features including user influence, engagement history, geolocation, and temporal dynamics. Additionally, we sketched a validation process involving a representative sample of users who can help assess the weights associated with each like by trial and error. Overall we aim to provide a method to gain valuable insights to several actors, such as content creators, marketers, and researchers.