One of the most contentious issues in marketing today is the issue of privacy, especially in regards to data-gathering. A recent Pew Poll showed significant across-the-board concerns about the amount of private information that’s getting into the hands of business, and this is undoubtedly affecting customers’ attitudes towards the companies they deal with.
For example, many of you are probably familiar with the case of Target announcing a pregnancy via direct mail before the family itself knew, because their data-mining had produced a model of first-trimester purchasing habits which was TOO accurate. Instead of helping a newly-pregnant woman with her shopping, it was seen as creepy and intrusive – and helped spark the debate that’s still ongoing today.
Properly targeting a niche demographic without scaring them away is a tricky balancing act. Let’s look at a few ways to accomplish it.
Producing Direct Mail Which Is Personalized Without Being Creepy
1 – Keep things regional.
If you’re talking about things on a household level, that’s getting too personal. But talking about a neighborhood? Then it’s relevant. Talking about issues, demographics, trends, etc. on a neighborhood-level scale provides a good level of personalization, but without tipping over into stuff that’s too personal.
Anything that would be common knowledge to someone living in that area can be fair game. At that scale, just about anything you can reference will help add relevance.
2 – Only use personal numbers that are directly business-relevant.
It’s only natural that a business is going to be keeping close tabs on their business relationship with customers, so these numbers are almost all fair game. A good example here is how “frequent shopper card” promotions like to send out mailers informing people of how much money they, personally, have saved. That’s totally relevant use of data.
The only caveat here would be to show discretion in how many personal numbers are shown in your direct mail pieces. Don’t itemize when it’s not necessary. For example, saying a customer has purchased bananas “many” times this year is probably better than specifying 19 times.
3 – Personalized recommendations are almost always appreciated.
As Amazon has amply demonstrated, promotions in the form of “Customers who purchased X also purchased Y and Z!” are effective personalization that doesn’t cross the creepy line. It’s information that’s directly beneficial to them, without including anything too specific about the people involved.
Another nice thing about this particular promotion is that you never have to “show your work.” You could have incredibly detailed data driving these pieces, but as long as the customer only sees the final recommendation, they won’t know about it.
4 – Use representative images.
You probably have information in your database about the ethnicity and general family makeup of your customers. However, they’re likely to be creeped out if you directly talk about them being an Asian family with three children, or whatever.
Instead, include a photo depicting a family that matches their own -a generic Asian family with three children- but without commenting on it directly. This sort of drop-in imaging can easily be implemented with digital direct mail techniques, potentially from a library of dozens of stock images representing the range of family types your business services.
5 – The cocktail party guideline.
If in doubt, here’s our little rule of thumb: Would you talk about it at a cocktail party or similar social gathering? If it’s not a topic they’d want broached in a setting like that, they probably won’t appreciate it in a direct mail advertisement either.
What do you think? How have you been trying to leverage data without coming off like a stalker?