Construção de um processo ETL automatizado em dados de campanhas de uma empresa no setor bancário

Abstract

The ETL - Extraction, Transformation and Load (Extract, Transform and Load) are steps of a procedure for systematizing the treatment and cleaning of data that have as a source, several organizational systems (OLTP) for insertion, normally, in DW (Data WereHouse) or Data Mart. The objective of this technological report is to present the construction of an automated ETL process, built to optimize data processing activities that monitor the performance of the commercial team in incentive campaigns, launched by a company in the banking sector. As a problem situation, we have a report that monitors the performance of the commercial team in incentive campaigns, launch of new operations, monitoring of production indicators, inspection of forecast targets, supervision of regions of the country with below-expected performances that need more attention to understand which decisions must be taken by managers and the BackOffice team. However, the report in question was previously developed manually, extracting data from different sources and manipulating them in a tool that made it impossible to automate due to the large volume of data per day and later for its consolidation and availability for the interested. As a solution adopted, we chose to use the SAS Guide tool because it is renowned in the market, has statistical resources for data analysis, robust in processing large volumes of data. The results obtained with the implementation of the process brought an increase in productivity, greater power in decision making and consistent and complete access to information for the commercial team, less time to process information, previously what was done in 4 hours by an analyst in the day, it is now being accomplished in between 30 and 40 minutes with the automated flow. The increase in the approval rate, which was previously at 19%, now reaches close to 24%, also making it possible to have a view by region of the country.