The method for displaying hierarchical data using nested rectangles is called tree mapping. The different branch of the tree is given a category as a separate area rectangle whose sub-branches are represented further by small rectangles within it. The greater the size of the rectangle, the greater will be the value of that category and vice versa.
Why do you need it?
Using a treemap, you can visualize hundreds of points in a single visualization. By this, we can visualize patterns of the data. A treemap gives us the ability to visualize a single dimension with two measures. One measure could visualize the size, and others can visualize color.
What kind of data you can visualize with it?
It is best suited for categorical-data with one or two measures. If you have a dataset with categories of different ship modes like first class, second class, standard class, and their size of profit respectively, then treemap is an alternative. It will display which category represents higher profit, as the area of rectangle indicating that category will be maximum.