This paper presents a systematic effort to prepare base data for benchmark tests to evaluate the quality of fatigue prediction methods and solvers. The output of the work is the FinLiv database. The use of FinLiv is based on data sheets with a predefined structure, saved as *.csv files. Two applications take care of handling the data. The FinLiv.VBA application under MS Excel is focused on new data input. It allows users to categorize various aspects of fatigue tests, save them to a csv data sheet, send them to a standardized xls document, and finish the regression analysis there. The final results of the regression can be further elaborated or processed to generate inputs for various fatigue solvers. The other part of the solution is the Internet FinLiv application, which takes over the csv files and gathers them into MySQL databases. The front-end of the database is available on the www.fadoff.cz website. It allows users to look for particular types of data, and to view them or compare them. A csv file can again be generated from the database record and reused for any other goal in research or engineering analyses.