The data in this example represents retail sales. Data has been recorded about the sales channel each customer responded to, as well as a promotional deal. In addition, there are some basic customer demographics; gender, age, and the type of property they live in. You can download the data as an Excel workbook here.
This company has a customer loyalty problem, with almost no customers still giving them repeat business after a period of six months or so. You can clearly see this in the @SK results; this graph shows sales volume by date for customers likely to make a repeat purchase.
Conversely, here is the corresponding result for customers not likely to make a repeat purchase.
Of course, @SK will also show you customer demographics for both groups of customers, but the most useful part of this analysis is the extra data that @SK shows you on the original worksheet.
Now, this company can design a marketing campaign that precisely targets customers on the point of defection. You just have to choose a confidence interval (we suggest maybe 90% or so), then contact customers with loyalty flagged lower than this value with an offer. It would be good sense to set a lower limit (maybe 50%), below which you consider that customer a "lost cause", and save the cost of the offer. The point is, you have a rigorously-computed statistic, and how you use it is completely up to you.
These are real results. If you want to try this for youself, all you need to do is load up the test data, push the Start button and select Lifecycle; @SK does all the rest.
Don't take our word for it. Download the trial version of @SK today, and see its powerful algorithms in practice for yourself, either using this data, or your own.