Most data analysis is concerned with understanding what the data tells us – for example understanding if business is growing or not and discovering how sales are developing in different territories or across product lines.
But sometimes the story the data is not telling can be of vital interest. If a customer has stopped ordering how would you know – and how long would it take you to find out?
The chart below is designed to answer this question. It allows the user to see which customers have not placed orders recently and a slider control allows the user to define how many months “recently” means.
The customers are sorted by the volume of sales placed before they stopped ordering and the user can select to pick how many to bring into the view by using a second slider control.
The challenge in designing this lies in the fact that datasets contain information on things that have happened and not on events that did not take place – in this case the orders a customer no longer placed.
To extract this information we have to find a work-around that lets us “read between the lines”. Here we look for the latest date that an order was placed and use a calculation to check if it falls earlier than our test date.
This example looks for customers that have stopped ordering, but the same approach can be used to discover other ways that inactivity may be a risk for business. Rather than looking at order dates you could choose to analyse contact activity from your customer relationship data. This could help discover if there are large accounts that have not been contacted recently or significant sales opportunities that need to be followed up.
How might an approach like this help you recognise trouble while there is still time to act?
If you have thoughts on this please share them with us in the comments section.