If you're a multichannel marketer based in New England, you probably don't spend a lot of time thinking about customer behavior in Utah, Colorado, New Mexico, or Arizona.
Hillstrom's Zip Code Forensics (free to marketers who contribute anonymous zip code data, $5,000 for all other marketers --- compare to Claritas Prizm NE here) has a favorable view of the Southwest United States.
Multichannel evangelists suggest that direct marketing works well among customers who have access to retail.
Zip Models suggest otherwise. Notice that the bright orange (traditional direct marketing / catalog) and bright green (online marketing / e-commerce) areas are largely outside of urban areas, away from concentrated retail strongholds.
Honestly, folks, we've been led astray by mainstream multichannel "best practices" thinking. We survey a thousand customers who buy across channels, those customers tell us that they love multiple channels, then we tell retailers we should pursue these customers via $379 research reports and vendor white papers. Who benefits from this form of marketing? You know the answer.
We have an opportunity to think more about what a customer "needs", thinking less about what we want to "give" to a customer. What does the customer living fifty minutes outside of Flagstaff "need"? Does our industry explore this angle? Have you read a research report on the needs of the rural mountain customer in the Southwest United States?