Nowadays, online consumer reviews are used to enhance the effectiveness of finding useful product information that impacts the consumers’ decision-making process. Many studies have been proposed to analyze these reviews for many purposes, such as opinion-based recommendation, spam review detection, opinion leader analysis, etc. A standard model that presents the different aspects of online review (review, product/service, user) is needed to facilitate the review analysis task. This research suggests SOPA, a modular ontology for cloud Service OPinion Analysis. SOPA represents the content of a product/service and its related opinions extracted from the online reviews written in a specific context. The SOPA is evaluated and validated using cloud consumer reviews from social media and using quality metrics. The experiments revealed that the SOPA-related modules exhibit a high cohesion and a low coupling, besides their usefulness and applicability in real use case studies.