The business market is different today than it was 20 years ago when BI got started. We're just beginning to grasp how to work within the new economic and communication models. Companies can't rely solely on financial and operational metrics any more, and need to analyze customer behaviors in more detail.
The big change in analysis is a move from mass market metrics to individualized data, no longer analyzing or managing by averages. The stream of events and observations available from applications today combined with new platforms for collecting and processing data enables (relatively) easy analysis.
Despite this, many companies struggle to analyze customer data. This talk will describe a handful of customer metrics and models that are (relatively) easy to do, yet are often not done. It's often easier to succeed by stringing together a handful of simple techniques rather than applying advanced techniques.
Expect to come away from this session with:
- a little history of customer data use by marketing and how that has changed in the last 10 years.
- the most common behavioral data sources you have available.
- some of the basic questions that often go unanswered, and data that is not assessed in the proper context.
- some basic analyses you can perform.
44. Why simple might be better: Video Store case
Key Figures:
▪ 200 stores, 200.000 transactions per week
▪ Appr. 1 million active customers, 1.3 million inactive
▪ Average frequency 8-10x per year
Adapted RFM Model: FMR
▪ 'F' determines 'M', 'M' class influences 'R' class
▪ M = Revenue in 52 weeks prior to last visit
▪ Less than 52 weeks of data: weighted total
Assumptions (tested):
▪ No visit in last 52 weeks : inactive customer
▪ First 12 weeks: new customer
Similar model results, 2 minutes vs 2 days, SPSS consulting and
licenses and server vs 0 recurring external costs, better overall result