The purpose of data analysis is to uncover relationships and trends in the data. The analysis should give you new insights into
the situation that the data describes. To structure the analysis, you should ask questions like these:
| • | How did my organization perform this year as compared to last year? Which parts and personnel of the organization did better and which did not? |
| • | For a consumer business, what is my most and least profitable product/customer/salesperson/office/store? Which factor in my data trend in the same direction as the most profitable, which factors trend in the opposite direction, and which are neutral? How do those factors trend for the least profitable? |
| • | For a hospital, for instance, which patients are staying longer than is typical for their diagnosis? What symptoms and other diagnoses do the long-staying patients have? |
The questions you can answer with Jaspersoft OLAP depend on:
| • | The structure of the data in terms of OLAP; that is, the structure of your cubes, dimensions, and measures. |
| • | The starting OLAP view defines which cube you want to analyze and the metrics relevant to a particular need. |
OLAP views give business users a starting point for analysis that can then be sliced and diced to answer detailed questions. For a particular OLAP data set, there are usually a number of OLAP views defined as convenient entry points.
As an example, let's answer a specific question: “What is the quarterly sales dollar amount for the snack foods category in 2012 for stores in California?”
Recommended Comments
There are no comments to display.