The goal of the research is to create an algorithm for analyzing vulnerabilities discovered and to enhance the data protection technique based on system penetration testing as a means of evaluating data security. The primary responsibility is to modify established security assessment techniques to fit the evolving technical landscape, namely cloud technologies and security system requirements. The issue of data protection evaluation of electronic documents is discussed in the article, and specifically, an A/B test was conducted. To address this issue, an analysis was conducted on current data protection evaluation methodologies and techniques, and an A/B testing system was constructed. These methods help to determine the level of danger of vulnerabilities, which allows you to effectively and visually determine the overall level of system security by evaluating each vulnerability found. The article examines the practical use of the method, prospects for further development, and features of the effectiveness of penetration testing as a method of assessing the security of information systems and the development of an algorithm for evaluating the vulnerabilities found. Researching the methods and means of security assessment, the work was aimed at creating a convenient and effective tool for assessing the security of data and documents in the network. The theoretical significance of the work is determined by the expansion of scientific knowledge in the field of information and analytical systems of electronic document management. The research aims to explore and apply big data analysis and machine learning techniques to improve data protection. The novelty of the research lies in the synthesis of modern methodologies of A/B testing and analytics within a single platform, which will allow electronic documents to protect data and increase efficiency from hacker attacks. The principles of the method are described along with the main approaches and techniques that contribute to the analysis of hacker attacks on online systems and cloud environments. The components of the method, which provide for data protection, were separately analyzed, and an analysis of system testing was carried out to determine the requirements for the new system and increase the level of protection against data theft. The application of methods and tools for assessing the security of networks, which use a complex of general scientific, experimental, practical, statistical, and mathematical methods for the implementation of algorithms and the formation of the determination of the levels of danger of vulnerabilities, is considered. The method includes data analysis tools, as well as analysis techniques that provide an opportunity to conduct flexible testing of changes on the site with further analysis and impact on conversion, which is important for increasing data protection and meeting consumer needs.