Data-Driven Business Decisions

CHAPTER 2

What Stands Out and Why? Who Wins? Data-Driven Views of Performance Dynamics

2.0. Introduction: What Is the Issue?

Describing performance data by its distribution, perhaps summarized by the mean and standard deviation, is fine as a first step but is not much use in itself. We want to know what factors drive high and low performance. We want to see how our management interventions affect performance. We want to investigate and quantify performance differences across market segments.

Here are a few more specific examples.

• You have daily sales figures for a supermarket over 10 weeks. How do sales figures compare across the 7 days of the week? Different days are associated with higher and lower sales.
• A telephone company mails advertising brochures to a randomly selected group of customers. How effective is the mail-out? This involves a comparison of the behavior of customers who received the brochure with customers who did not.
• You want to know how the purchasing habits of male and female customers differ. How many products do they typically purchase on a shopping trip? How much do they spend per purchase?

Questions such as these motivate the methods and ideas in the chapter. The key idea is breaking data up into meaningful subgroups and comparing the performance across the groups.

But there is another issue that we have completely ignored up to now. Not all business performance can be summarized in a single number. It is part of the complexity of modern work that we ...