The problems of development of audit activity have signified a drop of audit companies' business reputation. During the transformations of the global market economy of the last decades of the 21st century, the questions of minimizing costs amidst growing competition and the requirements for high professionalism of auditors, as well as the growing needs of investors and owners due to changing markets and legislation, become highly relevant in the future consideration of the issue. The article discusses and tests the methodology for using statistical research in audit based on audit materials where sampling is used as a main instrument. We analyze the possibilities of selective research utilization during an audit check of a huge amount of accounting data and reports, evaluation of the results of the check and their extrapolation to the entire population of data taking into account the correlation of the elements of the sample. In order to minimize the number of mistakes in the auditor's conclusion, we offer a sampling procedure that takes into account the points of highest risk and a method for risk evaluation. We have developed document templates to provide to auditors aimed to systemize the source data to select the sampling type, auditor's actions when processing accounting data and reports, the results of audit sampling and to reflect all the steps of sampling research at any stage of accounting process. Additionally, we have developed a set of indicators that make up the base for sampling research in audit. The method offered has been approbated on enterprise's data, which has confirmed that using these instruments of digital economics is possible, that it does not reduce the quality of the audit results, and its practical applicability from the point of view of minimization of labor. © 2019 Copyright held by the owner/author(s).