November 22, 2010 – 22:05
A few days ago I became aware of the Mozilla Open Data Visualization Contest that is being organized by Mozilla Labs and the Mozilla Metrics Team. The goal of this contest is to creatively visualize answers to the question “How do people use Firefox?”. For example by creating visualizations that investigate interesting usage patterns, reveal interesting user behavior, or explore browser performance.
I figured this would be a nice challenge, so I decided to download the data and load it into QlikView. Since I am always curious to see what other people can do with QlikView (it can be very educational) I have shared my load script and data cloud here, so that other developers interested in joining the competition can get a running start. read more »
November 18, 2010 – 12:30
With over 80% of data* having a spatial component,
geographical analysis can add a powerful new dimension to almost any reporting environment. In the coming time I intend to review the various methods of extending QlikView with geographical analysis capabilities, describing how to apply these methods and what their pro’s and cons are.
Read on to see the options I have identified so far. read more »
November 10, 2010 – 00:44
Do you know if your customers are profitable? All of them? Performing Customer Profitability Analysis can answer these questions and give you some amazing, and sometimes counter-intuitive, insights into your customers’ contribution to your bottom line.
This post describes one of the visualizations that you can create once you possess accurate data* on the profitability of your customers: the whale curve.
In a whale curve, customers are ranked by profitability, from highest to lowest, on the X-axis while their accumulated profit is plotted on the Y-axis. The curve that results can, with some imagination, be said to look like a whale coming out of the water. An example of a whale curve chart is shown below.
When you look at this chart, you may notice that the top 200 customers generate the bulk of the profit. You may also notice that the you are losing serious money on the bottom 100 customers and that the customers in the middle are more or less break-even.
Read on to learn how to create a whale curve in QlikView. Even if you’re not interested in creating a whale curve, you might still want to read on to learn more about the rank function and the continuous x-axis. read more »
Decile analysis is a popular segmentation tool. Where a pareto analysis splits the top 20% customers (or products, regions, etc.) from the bottom 80%, decile analysis divides them into equally sized groups of 10%.
The image below shows an example of a decile analysis.
The example shows how a group of 1.000 customers is divided into deciles of 100 customers. Lots of interesting things can be learned from this analysis, amongst other things:
- Your top 10% customers are generating profit that is significantly above average;
- Your top 30% customers are responsible for 80% of your profit;
- You are losing money on your bottom 20% customers (the so-called “bleeders”).
So, how do we create a decile analysis in QlikView? read more »
