Using and misusing business data

In one paragraph

Data supports a decision; it does not make it. Knowing the limits of the numbers in front of you is as important as reading them.

Quantitative data is countable — sales, costs, market share, survey percentages. It is comparable and can be tracked, and it tells you what happened, not why.

Qualitative data is descriptive — opinions, complaints, interview answers. It explains why, and cannot tell you how widespread it is.

Where data misleads

Small or biased samples. Ten responses from your most loyal customers is not the market. Ask who was *not* in the sample.

Correlation is not causation. Two things moving together does not mean one caused the other. Ice cream sales and drownings both rise in summer.

Out-of-date data. Secondary data was true when it was collected. Check the date before you build a plan on it.

Averages hide the distribution. An average order value of £80 may be a hundred orders at £20 and ten at £700, and those are entirely different businesses.

The most useful question about any figure is: what would have to be true for this number to be misleading?

How to actually use it

• For any number you rely on regularly, write down where it comes from and how often it is refreshed. Half of business reporting is numbers nobody can source.
• Pair the two kinds deliberately. When a metric moves, ring three customers and ask why. The number tells you where to look; the conversation tells you what to do.

Where it breaks down

• Collecting good data costs time and money, and beyond a point the cost of more certainty exceeds the value of the decision it improves.

If you are being examined on it

Questions about data almost always want the *limitations* — sample size, bias, age, and the fact that data informs but does not decide.