Evaluation is an important step in the development of information retrieval systems, as it  allows for the successful measurement of a system designed to help users meet their information goals. Several scholars have conducted a series of tests to calculate the efficiency of information retrieval systems. Among the developed methodologies, the Cranfield model remains the dominant approach in evaluating information systems. This paper presents the steps to be taken to test an information system, the indicators used to evaluate the effectiveness of information systems (recall and precision). Studying the evaluation and its role in the functioning of information retrieval tools can essentially contribute to improving the performance of these tools and the quality of their results.