The primary reason that corporations introduce Six Sigma into data warehousing boils down to cost reduction. Large corporations are incurring huge expenditures, most of the times running into millions of dollars, which eats into stakeholders' margin, in creating and maintaining data warehouses. The criticality of data warehouses can be understood by their vital role in support to prediction of business performance.



There is no denying the fact that data warehousing is in a way, the powerhouse of Six Sigma deployment. In early stages of projects, data warehousing allows for better planning of deployment, design and tuning of the production environment.

Data Warehousing Basics

Data warehousing components are complex in nature and are multifaceted. The various components are either developed in house or by a third party or in joint development at the party's place of business. Typically, designers focus on functional and business needs and not on performance constraints faced by the production environment. The consequence of this costly mistake is the possibility of missing deadlines and reworking the project, which are manifestations of operational inefficiencies.

Challenges to Data Warehouse Design

It is not new that modern day data warehouses are built for auto refreshing and/or compatible for at least real time updating. ETL, as extraction, transformation and loading of data flow is a very resource-consuming exercise in data warehousing. The importance of data warehousing increases several times, considering the fact that data structures are both strategic and functional.

Even the real time refreshing of data becomes a daunting task with the refresh window getting clogged straining server resources. Then there are some other factors that have a play in affecting the performance of ETL.

Meeting the Challenge to Quantify the Data Warehouse Effect

Quantifying the effects of data warehouse is to project whether challenges can be scaled. The recent trend in data warehouse development is to treat them as belonging to the same family or group. Consider dedicating each family to a particular geographical location, and other subsets of respective hierarchical data. Warehousing modules for individual data groups (families) are developed at their initial stages and new ones are taken care off as and when they arise and are just plugged into the main data warehouse. The database could contain three fundamental tables such as tables to store attributes of data; storage of linking information; and finally, aggregated data ready for use.

Applying Six Sigma Elements into Software Development

Applying Six Sigma elements into software development typically helps in identifying potential problems in production if the development is done in the early stages of the project. Secondly, the mammoth task of data warehousing can return positive results if deployment plans are fine tuned before implementation.

The self-assessing nature and the provisions for internal auditing shed light on the course of implementation. At the same time, one cannot forget that databases developed remain tied to the system architecture on which they are built and bear heavily on the accuracy of predictions in a fluctuating business environment, ironically for which they are built.