When you're looking for information about the customers of a consumer products company (often such companies will sell directly to consumers, though not always), then there are three basic questions to research:
For finding some of this information, to start with there is the Census Bureau. Here are some of the main links:
Then there are commercial databases, or datasets ...
In a few databases, you get a combination of all types of information, but generally that means that you lose detail in one or more. For truly detailed information that combines demographics, psychographics and geography, often that means doing your own primary data collection and research*. The sources here are all commercial products (i.e. secondary data), and do not get to that level. See the boxes below for more detail!
* For more about primary vs. secondary data and research, see the Marketing Research article in the Encyclopedia of Management.
The phrase "direct to consumer" originated in the pharmaceutical industry (in the mid-1990's, when direct-to-consumer advertising of prescription drugs became legal).
Now it is also used in other industries and in other ways (especially for digital relationship marketing efforts, i.e. social media or Web 2.0). Essentially it means marketing efforts aimed directly at consumers and bypassing any middlemen (retailers).
(Demographically, that is ...)
For example:
Age, sex, income ...
These are usually easy -- we all know that people of different ages, sex, and so on often buy different types of products!
Lifestyle.
This is sometimes easy to figure out, sometimes not. For example, it's pretty obvious that athletic people will more likely buy sporting goods and clothes, or that people with young children will buy diapers and toys.
It is less obvious which people will buy PC's vs. Mac's, or which people are likely to be interested in the arts, sports, or business ... This is what sometimes is called psychographics, though the dividing line between demographics and psychographics is often blurry at best. (See the definitions to the left.)
For example:
A geographically limited area: a 50-mile radius, a 250-mile radius, or much more?
Some companies can realistically reach only a geographically limited population. For example, a house-cleaning service or a small restaurant will likely have most of their customers within a certain geographic radius. A large manufacturing company will probably sell to customers worldwide.
Urban / suburban vs. rural ... or, what population density do you need?
For example, an expensive, fancy restaurant needs a certain population density just in order to survive, let alone prosper. A septic tank maintenance service will find more customers in areas without a municipal water/sewer system.
A specific geographic region: hot, cold; wet, dry; young, old ... etc.
The products / services of some companies are more likely to sell in specific regions, based on a variety of criteria. For example, why try to sell snow-blowers in the sun-belt? And while sun screen is a seasonal product in the north, it is a staple in the south.
Here are some tools you can use for these questions:
"Psychographics, plural noun.
The study and classification of people according to their attitudes, aspirations, and other psychological criteria, especially in market research."
From the Oxford Dictionary of English (revised edition).
What do people think, feel, and do? That is a shorthand way to think about this. Here's a longer definition or discussion:
The best single source for this that we have is the Simmons NCS, period. However, because all of the datasets have unique material, you may need to check MRI also, and perhaps even some of the books ...
Because using Simmons is not easy (unless you're already familiar with the IMS software), I'm linking to the page with lots more information about the datasets, rather than directly.
Getting the actual addresses of people in order to create a mailing list generally costs money. For academic libraries, ReferenceUSA is the best possibility, though it has two major limitations: