Joomla ServiceBest Web HostingWeb Hosting
Home Services Data Warehouse Development

Requirements Analysis

Before embarking on a Data Warehousing project, it is important to understand your objectives and how the environment will be used.

There are a great many questions that need to asked, for example:

  • How accurate is the information in your source systems?
  • How do you combine information from databases, flat files and Excel spreadsheets?
  • What reporting are you actually going to be conducting?
  • How frequently should the information be updated?
  • Who are the users going to be?
  • How are reporting requirements going to vary between different user groups?

Data Warehouse development thinking used to follow the line of 'identify every data source in the organisation; conduct a Dimensional Modelling exercise to understand how to combine the data; and build a database based on everything'. However, this approach avoids the crucial question "What are you going to use the data warehouse for?".

Again, the thinking on this used to be that it was difficult to predict what the usage would be - often users don't even know what questions they may want to ask of the data until it is in front of them. However, unless the data warehouse can be of immediate assistance in every day reporting, then it is unlikely that users will graduate to the level of asking new questions and exploiting sophisticated query tools. The risk then is that the data warehouse will fall into disuse and the entire project - whilst being a success technically - will be a failure from a business perspective.

Performing a requirements gathering exercise - focusing on reporting requirements and involving the potential end users of the system - is imperative for a sucessful Data Warehousing project.