Treemaps are an interesting way to look at data. Here is the first preview of the treemap add-in for excel. Being a demo, the number of data points are limited to 250. Also, to keep things simple, the demo is being made available as a standalone excel workbook.

Usage

Download the excel workbook given at the end of the post. The demo will work only with Excel 2003 although the complete add-in has been tested to work with Excel 2003 upto 2010. This is more to do with keeping it simple – the ribbon in Excel 2007 and later asks for additions that I would like to avoid at this stage.

When you open the workbook, you will see a new menu item being added to the toolbar with two options to choose from : ‘Simple Treemap’ and ‘Change Treemap’

Simple Treemap

The simple treemap option allows you to create treemaps where similar nodes are shaded using the same color for ease of identification. You can choose to color nodes at any level of the input data. Bear in mind that the last column should contain numeric and positive values.

Here is a sample treemap created using the simple treemap option.

Change Treemap

This change treemap option allows you to create treemaps where nodes are colored based on the criteria specified in the last column. The second last column specifies the size of the node. Bear in mind that the last two columns should contain numeric values and the second last column should only have positive values. The last column specifies the color and it ranges from dark Red to White and then onto dark Green indicating a range from the most positive to the most negative. The last column can be anything that you would like to use – rate of growth, per capita income or anything that can help you analyze the data better.

Here is a sample treemap created using the change treemap option.

The final product and the documentation will be released soon but the standalone file should work pretty well (I hope) with Excel 2003.

Update: I finally mastered the squarification technique. Squarification produces an aspect ratio close to 1 for each node. To a great extent, a higher degree of squarification needs a higher computation time. The squarification algorithm employed in this version is an approximate one and uses the same time to run as the non-squarified one. This relies on a modified form of the “pivot-by-largest” algorithm. Of the two images below, the first one uses the approximate squarification logic and tries to give each node an aspect ratio close to 1. The second one is a treemap where the nodes are plotted in the order in which they occur. The squarification comes into play when the ‘Squarify Treemap‘ option is turned on.

Here the a demo file containing the treemap add-in for Excel. (This one has been limited to work with a maximum of 250 rows and is meant for evaluation purpose only. The full version does not have this restriction)

If you like the demo, Buy the full version here.