When we have lots of data, we try to summarize it by calculating the average. As they say, averages are mean, they do not give away much.
I want to share with you an interesting example from Amazon.com on how they give more details by combining average with distribution.
As you might know, Amazon shows the rating of each of the products they sell. Customers & users rate the products from 1 to 5 stars. When you visit the product page you will see the average rating. But there is a small down-arrow next to it. When you click on it, Amazon shows you the break-up of that rating so you have a better idea of how the ratings are split.
Why show distributions?
Showing distribution of values corresponding to the average reveals important information about the data. We tend to use averages alone since they take very little time to compute and very little space to show. But adding the ability to show distribution of values (on demand) is a powerful way to let end-users understand the data better. [related: calculating frequency distributions in excel]
Below, I have shown a demo of how you can do this using Excel. Tomorrow, I will write a tutorial explaining the same.
On-demand Details in Excel Charts – a Demo:
What is your Opinion – Averages / Distributions / Both or Something else?
I prefer to use simple metrics like average, median, sum, count, min or max at the high-level in my dashboards or charts. But I always add extra detail by showing distribution of values or break-up by a different parameter so that my audience can understand the outputs better.
What about you? What metrics do you use at high-level and how do you add detail? Please share your thoughts and techniques using comments.
More Charting Principles:
There are tons of very good charting examples around us. Once in a while, I write about these on chandoo.org. I recommend reading the below articles if you make charts,