The Story
As one of the world’s largest food service companies, this company has over 30,000 associates and $2.7 billion in revenue. They offer travelers a blend of local, regional , and international brands at airports and travel plaza stops in North America. The Company collects huge amount of data related to stores and customers and uses analytics to understand the behavior of the customer. They had a significantly large database, which was consolidated from various systems using an ETL (Extract, Transform, and Load) tool to mine data from a database and store it in a data warehouse.
The Challenge
The company partnered with Centennial to outsource the reengineering of their data warehouse and its reporting structure. The goal was to create a modular, scalable architecture, considerably reducing the processing batch time. The solution was required to provide strategic and tactical decision support to all levels of management.
Our Solution
Centennial adopted a 4-phased approach: plan, design, implement, and operate.
Initiation:
Centennial performed a detailed study of the existing data warehouse, ETL model, reporting structure and out-lined key areas of improvement.
Design and Implement:
Centennial redesigned the data warehouse to use star schema. As a result, several data elements were added or eliminated. The new and redesigned warehouse data model may be the single most important aspect of the effort. The two primary components of the data model were the reporting data model and the metadata model. While the actual raw data was stored in the reporting data model, it was the metadata repository and its data model structure that allowed the various functional areas within the warehouse to communicate.
Reporting Data Model:
The reporting data model represented the actual data stored in the warehouse. It contained both the data and the relationships between the data. In the client warehouse, property information, lease information, financial costs, and personnel responsible for maintaining this data are all examples of entities that existed in the reporting data model.
Metadata Model:
Simply defined, metadata is “data about data.” Our premise was that Company data should be considered like any other Company asset with intrinsic value. Similarly, as the Company maintained knowledge about assets, the Company also needed to maintain knowledge about the data in the data warehouse. Questions such as, where does the data come from, when does it get loaded and transformed, who is responsible for maintaining it, and most importantly, how does it relate to the other data.
Results:
Ultimately, this solution provided the Company with a long-term flexible, scalable and maintainable reporting/analytical solution, and decision support system. The redesigned data warehouse:
- Enabled the Company to leverage rapidly changing technology, embrace best-of-breed products, and integrate legacy systems into a powerful information engine delivering timely and relevant reporting capabilities for their customers.
- Served as a much-needed platform to combine data not only from legacy systems, but also new applications.
- Provided a general reporting platform with analytical capability to serve as a foundation for the consolidation of enterprise data.
- Supported future reporting requirements and additional data sources, and it standardized technologies across the board with repeatable processes.
Curious to learn more? Contact Us!
