Q. What do racism and classical music marketing have in common?
A. Old white people (and a troubling grasp of statistical analysis).

Generalizations can be helpful, but they can also lead us astray if we don’t understand what we’re doing with them. Let’s look at a few plausible statements that you might hear in a marketing meeting, backed up with some authoritative-sounding research from somewhere like the NEA. They might individually be true, but if you’re going to base decisions on them, it helps to know how little information they really contain.

Most classical music fans are well educated.

Most classical music fans are old.

Most people don’t like contemporary music.

Most people come to at least two concerts a year.

Each of these statements tells us something about the makeup of the audience, but combining these insights to create a picture of a typical customer can be disastrous, misusing demographic data to create a stereotype that doesn’t exist.

It might be that only 1% of the audience is both old and well educated. It might be that most of the people who come to more than three concerts a year are really into contemporary music. Even if most of the “mosts” do travel together, you only need a tiny percentage of the total audience to fill the hall or buy a record.

The best-selling record of all time is Michael Jackson’s Thriller. Worldwide, more than 110 million copies have been sold. There are 6.9 billion people in the world. That means that for everybody that bought it, 61 people didn’t. Put another way, 98.4% of people have not purchased the most popular album of all time. In the US, where most people have access to a stereo, nine out of ten people didn’t buy it.

Forget about most people. You don’t need them. Nobody sells music to most people. The New York Philharmonic can fill the hall playing Varèse and Ligeti.

The audience comes to an average of 4.7 concerts a year.

The average record buyer purchases 1.7 records a month.

Most people have two arms. A few people have one. I’m not sure it’s even possible to have three. If Thomas Pink got their customer research from arts organizations, then none of my shirts would fit: I have an above average number arms.

Some people buy one ticket for themselves. Some people take a date. Others take their whole family. Group them together, and you’ll completely misunderstand the purchasing process.

When your variables only exist in whole numbers, nobody conforms to the mean. You’d get more accurate results by asking just one of your customers. Even a stopped clock tells the right time twice a day.

Classical music purchasers are more likely to care about audio quality than other music consumers.

12% of the US population is black. 44% percent of the US prison population is black. There is a higher incidence of incarceration among black people than among white people.

It’s not ok for me to assume that black people are criminals, but even if it were, it wouldn’t be sensible: it’s alarming that one in 36 black people is in jail, but this doesn’t tell us anything about the other 35.

The same is true of the audiophiles. They might be more prevalent among classical music fans, but it doesn’t mean they’re the majority, and it doesn’t tell you anything about the others. Build a store for the audiophiles and you might find yourself catering to a very small audience.