Thoughts on Demographic Targeting
By Heidi Tolliver-Walker on July 20th, 2012
Sometimes reading the comments posted after an online news story can be more enlightening than the story itself. Have you ever noticed that?
I experienced this phenomenon just last night when I read about a study conducted by Catalina Marketing that looked at a cross-section of CPG brands and found that 64% of the ad exposures delivered by advertisers in the category are reaching households that account for only 2% of the company’s sales. Meanwhile, only 15% of those exposures are reaching the consumers that account for 80% of sales.
What a waste! Or is it?
After reading the synopsis of the research, my first inclination was to write a post asking whether your clients know which clients are accounting for what percentage of sales. After all, the research seemed like a cry for basic data analysis. Then I read the comments at the end of the article. One person — not someone writing a press release or an editor putting together a news story but a researcher from another market research firm — made the following observation:
Just a thought, but maybe the 80% that make up sales are “pretty loyal” and are repeat purchasers so that the strategy is now to expand and start fishing where the new fish are.
This took me back to several other posts I’ve written lately.
The first is last Friday’s post about naked 2D barcodes. After a series of comments debating whether a Sherwin Williams direct mail campaign using a floating Microsoft Tag with no text and no incentive to scan the code was well done or stank, one of the commenters noted, “It really doesn’t make any difference whether naked worked for Heidi (or Michael) or not. What matters most is whether it worked for [Sherwin Williams].”
In other words, before we pass judgment, do we know what Sherwin Williams was trying to accomplish? If not, we really don’t have much insightful to say.
The second was the post I wrote in which John Leininger told the story of a wasted 1:1 campaign that looked great, was highly personalized, but didn’t create any lift over traditional static campaigns of the same type at all. Why not? A little investigation showed that the way the piece was personalized was very cool but created no value to recipients.
Circling back to this data, the surface take-away, and the comment, it takes us to the same place. Doing a basic data analysis is a really good idea, but before you can put that data to use, you need to understand who your target audience is and why, what you are trying to accomplish, and how to use the data to get what you want.
I think about an interview I recently conducted with Quad/Graphics about NFC communications. It’s expensive to produce NFC ads, especially inserted into magazines. How do advertisers know where the most valuable regions are to insert? Often by running AR campaigns in those magazines first. The goal isn’t to sell product. It’s to qualify the audience so they know what regions in which to run their more expensive NFC campaigns.
So perhaps those CPG advertisers know a little more what they are doing than folks are giving them credit for. Maybe they don’t. But maybe they do.