Zip Codes Tell You Little About Consumers And Their Buying Behavior.

By looking at your zip code, some companies believe they can figure out what you eat for breakfast or which political party you vote for. That was an actual headline in Self magazine.
The article was designed to warn readers that their privacy was being invaded, but marketers could read it as a ringing endorsement of zip-code marketing.  Indeed, the author quoted an exuberant pronouncement of Jonathan Robbin, a founder of Claritas, the firm that divided the country into cleverly named clusters; “Geography is destiny.  I can predict what you eat, drink, drive—even think.” Though there is a glimmer truth to this claim, the margin of error surrounding it is often too large to make the procedure of much practical value.

The idea is that, by using zip codes, companies like Claritas can zero in on “clusters” and categorize them according to 40 socioeconomic groupings.  “Young Influentials,” to name one cluster, for example drink fresh-brewed coffee, jog, travel abroad, eat yogurt and vote republican.

For 30 years American marketers have enthusiastically accepted the premises of zip-code marketing.  The theory works like this: Similar people tend to live together.  Drive around any American city and you’ll find affluent neighborhoods and humble neighborhoods, good school districts and poor school districts.  Similarly, people in neighborhoods tend to buy similar products.  Apartment dwellers do not buy power mowers; humble families do not buy BMWs.

Assume you can find those zip codes that contain a disproportionately high number of the best prospects for your product.  Once you have them, you can rank-order the country’s 43,000 zip codes based on that characteristic.  Voila!  Market to the winning zip codes, and ignore the losers!
On the surface, this is a good idea.  It makes more sense to market in those areas where you have a disproportionately high number of prospects than to spread your limited resources (resources are always limited) over the entire country.

The problem arises when we discover that even in the top 10 percent of all zip codes, the proportion of people who are hot prospects for a product or service is often no more than twice as high as in the population as a whole.  If 20 percent of the American people are prospects for your service, ordinarily the top zip code has no more than 40 percent.  Or, looking at it from the other direction, 60 percent of the people in the area with the most prospects are not interested in you or your product. Assume you are BMW, and only 5 percent of the people in the country are likely prospects.  It means that in the top 10 percent of zip codes, 9 out of 10 people you mail to are not your target (two times 5 percent equals 10 percent; 100 percent minus 10 percent equals 90 percent).

If you mail a letter to all of the people in that zip code, it’s more efficient than mailing to the entire country (indeed, it may increase the response rate by 100 percent), yet you are still wasting most of your effort on people who are not prospects for your product or service.  As we’ll talk about later, the future of marketing lies not in marketing to zip codes but in marketing to individuals—in finding the names and specific addresses of your prospects, wherever they live, and mailing to those individual people.

There are two simple reasons why zip-code marketing does not work as well in practice as its proselytizers say it does:

1. Demographics are generally poor predictors of consumer behavior.
   
   
2. People in a given zip code, are far more different than similar. They don’t even have the same demographics, let alone the same attitudes and behavior patterns.

So relax.  No company can, by looking at your zip code, figure out what you eat for breakfast, which political party you vote for, or where you will go on vacation.