Sanmargar completed a project involving comprehensive and automatic improvement of the quality of data of the Bank’s customers, including master data and identification data, ID document data, address and phone details, data on links and relations between customers, and classification data. The number of bank customer data records involved in this project was more than three million.

The tasks performed by Sanmargar included: measurement of current data quality and classification of significant errors, an inventory of business requirements in relation to data quality, including correction rules and error level acceptability thresholds, development of recommendations for automatic data correction rules and running of business workshops to verify and approve those, configuration of Sanmargar DQS software for automatic data correction purposes, based on approved rules, and creation of report generation processes for manual correction.

The main part of the project consisted of automatic data correction and generation of reports for manual correction on project extract data (three iterations of algorithm tuning), and subsequent acceptance tests of this solution (with the participation of employees from more than 20 branches of the Bank). The finale of the project was the development of a report on work completed and delivery of a solution for the process of regular data cleansing, presented to the Steering Committee, including members of the Management Board.

The project was completed with the use of a solution consisting of an Operational Data Store which stores data from the source systems, based on Oracle technology, and a repository of business and technical data correction rules, created with the use of the Sanmargar Metastudio DRM application.

The key process of data correction and deduplication, and mechanisms of automatic generation of test cases and data reports for manual correction, were developed using proprietary Sanmargar DQS software.

The project was completed with due diligence and according to the bank’s expectations. Thanks to its completion, the Bank can significantly improve the quality of its customer data and boost the effectiveness of its business processes. In addition to that, Sanmargar provided the Bank with reference address dictionaries.

#smartdata #dqs #metastudio #dataquality