Government ISSUE

IT’S SURPRISING HOW MUCH INFORMATION is available to marketers these days from the government. Most of it is cheap, even if you have to buy it through a list broker or a data enhancement house like Experian or TransUnion.

If you are really clever about the way you pull the list, it may not even cost you any extra to break it down by demographics.

Government-supplied information enables you to discern the obvious, like ethnic group, age, income and education level. But you can also use it to infer the health of people living in a neighborhood, their propensity to buy certain kinds of products and services, and even their likes and dislikes.

Publicly available data can be found at all levels of government. In fact, there are probably hundreds of publicly available databases that even third party vendors are not aware of.

A look of some of these sources demonstrates the depth and breadth of information we can get about American consumers even before they become customers.

Census Data

Census data is now available online in a fairly raw form on CD from a number of third party vendors as well as from the Census Bureau.

Even if you only know your customer’s ZIP+4 codes, you can use the census data to figure out the demographics of your customer down to the block face in a specific neighborhood. Once you know the dominant demographics for the block face, you can easily infer the wealth, race, religion, country of origin and education of a specific set of customers residing within your database. Once you have this information appended to the database, then you can clone your file, so to speak, and look for like customers.

Mapping this information is pretty straightforward these days. Companies like MapInfo and DecisionMark make software products that merge your customer data with demographic information to provide a visual of where your target audience lives and what it looks like.

Census Bureau information also can be used to model either inside or outside lists by accessing the North American Industry Classification System (NAICS) or the older Standard Industrial Classification (SIC) information. This data can be cross-referenced by geography, gross sales and number of employees.

State Governments

While there are legitimate concerns about privacy, and states are beginning to restrict list access, you can obtain abundant data through third parties or from government agencies at a nominal cost.

The variety of information that is available is just plain amazing. For example, a colleague once determined that state governments provide information on:

Just about anything that requires a license generally means that the licensee information is available. Of course, driver’s license and motor vehicle registration data are now largely cut off. But many of the large data houses now have inferential formulas to get at the same information. The bad thing is that what was publicly available and cheap if you knew where to get it directly is now tougher to get and expensive.

Local Governments

Most local governments require business registrations at the local town hall or county level. But probably some of the easiest information you can obtain is from the deeds files that are kept by every city government concerning the purchase of commercial buildings, land, private homes, condominiums and co-ops.

While you can buy it through third-party sources, sometimes this data is a bit out of date. I know of one very small local bank that hired summer interns to glean from deeds people who owned homes in a certain price range for exactly one year. From this information, the bank created a highly researched list of potential home equity line of credit customers, and pulled an outstanding response. The tradeoff for the bank was the number of hours it took to compile the list. Larger organizations will find that it is far more profitable to rent a list through a broker.

The USPS

The U.S. Postal Service Walk Sequence file (showing the most efficient local delivery pattern within a town) is frequently used to minimize the time it takes to do door-to-door fund raising. When overlaid with demographic data on a block-by-block basis, this becomes a powerful tool that can significantly increase yields.

Just to show how even a simple alteration in requesting names can change response, let’s consider a campaign for the Brooklyn College adult education program.

The goal was to get as many people as possible to come to an open house. Previous mailings had gone to compiled lists of people within the entire New York City borough of Brooklyn, without regard to demographics or where most of the current student population lived.

A simple analysis of the Brooklyn College adult education database showed that 90% of the students lived within three to five ZIP codes. (For those not familiar with the borough, Brooklyn has many ZIP codes, and by itself would probably be the fourth-largest city in the United States.)

It was also discovered that several ethnic groups dominated the school population. In descending order, they were: Caucasian (mostly Russian), African American, Hispanic and Chinese. We also knew that the U.S. Census reports education and ethnic data down to the block level.

As there was no budget to do enhancements, some cheap, clever list pulls had to be executed. That’s when the fun started. First the list compiler was asked to pull only households that lived within the ZIP codes that produced 90% of the college’s adult education students.

Within these ZIPs, there were pockets that had a greater percentage of households fitting the demographic profile. Also, they included high school graduates who had no college training under their belts, or some college.

To hone the rental list down, the compiler was asked to pull only households fitting the demographic criteria