ZIP+4 + 2 = ZIP+6
(Direct) How do you get an entire newspaper chain to improve its targeting? It’s not easy when you own 90 papers, ranging from small weeklies to big-town dailies. But Gannett took on the job last year. It launched a program called customer acquisition targeting. The purpose: to help its newspapers use Claritas’s Prizm clustering system at the ZIP+6 level.
First, Gannett set up a central ZIP+6 directory, allowing newspapers to select the addresses relevant to their markets and then apply Prizm codes to create target groups. It also conducted a Webinar on the process, says Karen Giroux, director for marketing information and assessment for the firm.
But Gannett didn’t anticipate the budgetary restraints faced by many newspapers. And it had to concede that the papers knew their markets better than the home office did.
Take the Arizona “Republic,” the largest paper in the chain, which like most other papers had been using the U.S. Postal Service’s ZIP+4. “It’s the most savvy when it comes to direct mail,” Giroux says. “They didn’t just take my word” that ZIP+6 would work.
To test the new program, Gannett took a small slice of the paper’s address database and picked three homes in the same ZIP+4 segment code. All three were classified as “big fish, small pond,” indicating they are empty nesters, belong to country clubs, have large investment portfolios, and spend on computer technology.
But the local managers saw them differently.
“Since they live there and I don’t, they made sure to tell me that only one of those homes truly fits that description,” Giroux says. “Using the Prizm code, they found distinct clusters assigned to each home and confirmed the accuracy of the codes.”
Why the discrepancy? Each ZIP+4 code contains an average of four households, although there’s “a boatload that are actually single households,” says Sandra Hamorsky, Claritas’s senior product manager for segmentation. ZIP+6 codes are more likely to contain single households, and they allow for more precise profiling, Hamorsky adds.
The paper made quick use of that information. It appended Prizm clusters at the ZIP+6 level, identified those most likely to pay by credit or debit card, and mailed to non-subscribers matching those criteria.
The result? General response improved from the 0.5%-2% range to almost 2%. “And it’s still improving,” Giroux says.
Then there was the Ithaca (NY) “Journal.” It used ZIP+6 and demographic data to send a jumbo postcard for Life, its new lifestyle section. It said: “Life. It’s all about you.”
And the paper did mailings to increase single-copy sales. That number has jumped by 103 copies a day, or just over 2% — a big boost, Giroux says.
The “Journal” also used Prizm ZIP+6 to find drop-off locations for “Buzz Entertainment,” its publication for 18- to 24-year-olds. The magazine’s pickup rate increased from 72% to 90.7%.
Another paper to hop on the bandwagon was the Salem (OR) “Statesman Journal.” That paper was averaging a 1% response with an ongoing sampling and direct mail campaign. Since adding ZIP+6 and target analysis, response has increased to almost 3%.
Not that ZIP+6 can solve everything. The Pensacola (FL) “News Journal” split its market into several target groups, then did a mailing to increase Sunday circulation. The headline: “I’ll get hooked on what?” But “results were not as high as they would have liked,” Giroux says.
That may have been due to the sign-up premium: a discount card for a local sports store. The paper will now offer tickets to a brand-new movie theater. “We’re hoping the change in the gift will make a difference in the response rate,” she says.
This article is based on a session presented at the National Center for Database Marketing Conference in Orlando, FL.
