In yesterday’s post, I commented on a scatter plot that was part of an online demo of Microsoft’s Power BI product in the Data Visualization section:
I’ll now say a few words about one of the charts in the bottom of the dashboard.
The stacked bar chart on the left is sorted by month by default. But look more closely – the criterion used for sorting is alphabetical, so our months are displayed in ascending alphabetical order, rather than as a time series! And, incidentally, February is spelt incorrectly.
Now let’s look at the components that are stacked in the chart. There is no legend for this graph, but the colours are the same in the graph on the right. The right hand graph does have a legend, from which we can infer that the components of the left hand chart are Profit and Sales. But why are these two items stacked on top of each other? The total has no meaning whatsoever!
The graph has a sort function that enables us to change the order of the months from alphabetical to order the months by the size of Profit or Sales, either ascending or descending. However, this function has little practical use as it is far more important to see Sales and Profit as a conventional time trend and with a context (such as budget or prior year) to compare against.
This demo is very much a case of style over substance – it looks quite nice at first glance but fails to answer any key business question.