Every consulting work undertaken goes through a rigorous data mining process. The initial stages are quite common across clients, but going forward they become very specific. Initially before starting to analyze the data, all the datasets related to a particular project is collected and a data mart is prepared. The data mart is prepared in such a way that necessary subsets of data can be picked out as fast and as easily as possible.

For every client proposal a spec sheet is created with the project details such as kick off date, delivery date, client update dates, deliverables, etc. The timeline sheet is made ready, and the analysis is all set to start. The data mart is also checked for whether the necessary inputs are present.

• Top down Approach: A hypothesis is made on prior business knowledge and the hypothesis is validated using the data.
• Bottom Up Approach: The data is first looked at for unseen patterns; then a hypothesis is created based on these patterns and later the hypothesis is validated.

To further know about the data mining approach, please click here.