As part of the work on an intelligent text processing system for research projects, an interface for data search and analysis is being developed. One of the key tasks is to search for documents, taking into account both formal features (organization, scientific direction, authors, year, etc.) and the meaning of the text. In the future, it is necessary to implement a search for individual sections of the document, taking into account its internal structure. As part of the work, such solutions were tested as: categorization and filtering by metadata, simple word search, contextual query constructor, formation of contextual queries based on voice input; several options for presenting search results were considered. None of the considered options for interface solutions for organizing multi-criteria search and data analysis can meet the user's needs, and using all possible options together greatly burdens the interface and limits the user's workspace. The article discusses the user experience and promising solutions to the problem.