What is Data Warehousing?

Businesses measure all sorts of things in order to better manage their business. They measure how much a customer has spent on their products, what products have the best margin and how long it takes to get a product delivered to the customer after the order is confirmed. You name it, and most assuredly someone or some business out there is measuring it.

But measurements for a business are only as good as the delivery of this information to the people who manage what is being measured. This is where business reports come into play. Reports offer consolidation of the items measured. Generally speaking, the higher up in an organization a manager is, the more consolidated his or her reports need to be.

Computerized database technologies have been developed over the last few decades to allow businesses to store their data in highly efficient electronic formats, a.k.a. databases. Businesses rely on database systems geared toward capturing data from the users. These ‘front-end’ systems are often referred to as “On-Line Transactional Processing” systems (OLTP). Most transaction systems, in addition to their data entry and capture capabilities, also offer some amount of reporting. Even the most basic of reports can offer consolidation and some bit of context. For example, “How much did we sell last quarter as opposed to our projections?” But even the best and most complex of OLTP systems will have some reporting shortcomings for one of several reasons.

The concept of a “data warehouse” was developed to combat some of these issues. A data warehouse can be thought of as another database with high volume reporting and analytics as its main purpose, as opposed to row-by-row data retrieval and manipulation. They typically contain copies of data already managed by transaction systems, but they are designed and indexed for efficient bulk retrieval and reporting. In the following sections, we’ll explore some possible shortcomings of OLTP systems in regard to reporting and analysis and see how the implementation of a data warehouse can alleviate them.