CFT Poland successfully designed and implemented data warehouses for numerous customers. One interesting project was a data warehouse designed and executed for a large insurance company (EDW warehouse). The EDW warehouse consists of ca. 170 tables with data and 30 dictionary tables. The main transaction table contains over 217 million lines of data. Around 100,000 lines of data are loaded to the transaction table every day. There are 15 input interfaces implemented in EDW providing the warehouse with data from other systems. These interfaces automatically load versioned data every day. There are six data marts created for the needs of CFT reports implemented in the EDW structure. The EDW warehouse is also a data source for BI. As part of the implementation of the EDW project, CFT also created five interfaces to export data from the data warehouse to other systems or customers.
Benefits for customers
- The warehouse became a single reliable source of information for all departments of the company, because data from numerous systems are collected in one base
- Improvement of reporting processes - reports are created from one source
- After launching the data warehouse, time for storing information and reports decreased, and their quality drastically improved
- Providing better, faster and more reliable information makes it easier to manage a company and make strategic decisions
- Optimal, customised architecture allows for practically non-stop access to the data, regardless of loading times
- The implementation of data marts, separating Business Intelligence systems and reports from the warehouse, improved analytical capabilities.
- Analysts have constant access to the latest data, regardless of the accessibility of the data warehouse.
- Implementing the BI tools allowed users to create new, specific 'on demand' reports by themselves, without involving IT resources
- Automation of the generation an distribution of standard reports allowed for relieving employees from the necessity of making reports manually and sending them to appropriate people
- The system of logging events and errors makes problem cause analysis simple and quick. Support is easier, which decreases downtimes in data warehouse accessibility
- The quality of cooperation with external companies is improved by the possibility of exporting necessary data from one location
