Cloud computing has become the most popular concept for on-demand delivery of Cloud computing services. Due to its high flexibility, many Cloud computing services are designed and implemented to meet the users’ needs and expectations. As a result, new challenges have emerged in the search for relevant Cloud services. In fact, the description, discovery, and recommendation of Cloud services are the main challenges of Cloud computing. It stems from the lack of standardization for the Cloud services description and publication, as well as the exponential growth in the number and functionality of Cloud services. Our objective in this paper is to present a comparative study of the different approaches that address the Cloud services description, discovery, and recommendation issues. This comparison study’s findings are separated into three parts. First, the approaches to Cloud service description have several flaws, such as the lack of a unified description that encompasses all types of Cloud services, the lack of a definition for several properties (such as Cloud characteristics, actors, and pricing model), and the failure to consider several critical SLA elements (for example, QoS guarantee, compensation, monitoring, notification, and termination). Second, we determined that web-based approaches, also known as crawling approaches, are the most likely to be adopted in the field of Cloud service discovery, in order to keep up with the ever-changing nature of Cloud computing. Hence, the crawling approach’s main purpose is to update the Cloud service registry with new services that are available on the internet. Existing crawling methods, on the other hand, have a set of shortcomings, including the types and categories of discovered Cloud services, the constant increase in web-published Cloud services, and the automatic updating of Cloud vocabulary. Finally, major Cloud service recommendation issues such as cold start, data sparsity, attack resistance, and diversity remain unaddressed.