I found myself on the Jersey side of the Hudson, riding past a row of new townhouses overlooking New York City. Each townhouse had a single carport facing the road.
As I rode past, I noticed that every single carport, more than forty in all, had a similar car parked in it. Either grey or silver or a mix of the two. Either a lower-model Mercedes or more likely, a Toyota or a Honda. Every single one. It wasn’t until the 40th unit that I saw a red car, and a few later that I saw a pickup truck.
This is the power of demographics. This is the power of data mining.
Your choice of car shouldn’t tell us anything about what sort of neighborhood you’d like to live in, or who you will vote for in the next election. The kind of clothing you wear should have no influence at all on the kind of wine you prefer. But it does.
I don’t know which came first, the car or the townhouse, but the co-incidence of the two is unmistakeable.
This matters. It matters because the marketers at the townhouse ought to seriously consider a co-promotion with certain car dealers, and it matters because it opens a window for marketers. If people who buy novels also buy red wine, marketing your red wine in a bookstore might not be so dumb.
If you’re marketing a product or service in a cluttered marketplace, it may cost too much or be too difficult to reach the right person at the right time. Marketing red wine in a bar is intellectually compelling, but awfully expensive. But if you understand the data mining implications of the other habits of your typical prospect, you can reach those people somewhere else or sometime else.
For example, when I do a search on "grey toyota new york", there is plenty of unsold Google inventory, which means you can buy clicks here really cheap. So why not run an ad that says,
LOOKING FOR A GREAT PLACE TO LIVE?
This violates some of the precepts of permission marketing. It’s not anticipated. But it is likely to be personal and relevant, and even better, you don’t pay unless someone clicks.
Will this work for you? I have no idea. The chances that you’ll find the perfect match are unlikely. But if you do, the pay off can be significant.