Hierarchies [BIME TIP]

What is a hierarchy?

A hierarchy is a path though different attributes of a dimension. A typical example would be the dimension “product” with attributes that would be organized as Product Category > Sub Category > Product Name.

What is the point in such organization?

First, you can easily navigate through your data in a way that make sense from a business perspective. It is highly desirable to analyze product categories and then drill down on into the details of sub-categories. While this was possible before, it is now much easier and Bime does all the hard work for you.

Secondly, it helps to organize attributes in logical blocks. All your product-related attributes will be in one dimension, all your customer-related attributes in another, and so on. In the end you will have a naturally organized analysis canvas, allowing even those with minimal knowledge of the business or technical skill to a) understand the model, b) ask deep business questions, c) see what details are important, and d) drill down and explore those details.


Drag and Drop Product Category below Dimensions


It created a new Dimension of Product Category. Now Drag & Drop Sub category.


It added a new attribute in the dimension. Drag & Drop Product into the dimension.

You now have defined a dimension with 3 attributes ordered as: Product Category > Product Sub-Category > Product.

The last thing to do is to rename it:


This means that the level of analysis in your query can be changed up and down simply by clicking on the + or - signs on the attributes you have placed on the frame.


How does this change how you use Bime?

As with any feature, this would be useless if it were hard to use. Creating a hierarchy is done by simply dragging and dropping elements. The most popular new feature using this is the “decompose” feature on the cartesian and pie charts. This feature allows you to click on any data point of a chart and ask Bime to decompose it by any attribute available.