Data-Driven Business Decisions

CHAPTER 1

How Are We Doing? Data-Driven Views of Business Performance

1.0. Introduction: What Is the Issue?

Very few business outcomes are completely predictable. If they were, decision making would be so much easier. But as in life, we have to look at what the world throws at us and try to make sense of it.

If we collect data on a business variable then, because of the many factors that are not under our control, we will see a range of values. Here are some examples:

• Weekly sales figures over a year. The 52 numbers will vary greatly. Some of the variation we might attribute to calendar events like Christmas. But much of the variation is not predictable or easily explicable.
• Daily returns of a stock. In a year there might be 250 trading days, and so there will be 249 daily growth figures. These numbers are almost completely unpredictable. Typically just over half of them are positive and just under half of them are negative. You might think that good trading days occur in runs that can be used to predict tomorrow's result. You would be wrong.
• The sale prices of a large number of apartments sold in a particular area. Certainly we can anticipate that some properties will be worth more than others, but it all depends on who attends the auction and the available stock, and we will often be surprised at the price that is actually achieved.
• Satisfaction ratings of a surveyed list of customers. Every customer has a different experience with your business. Even if the experience is the ...