Did you know that I can tell the following about the people who have viewed my ads online:
Whether or not they are taking nonpresription drugs
Whether they listen to Pink Floyd or Taio Cruz
How much household debt they’re carrying (and conversely how much disposable income they have)
What brand of car they own, and how many miles are on it
Last week I had a spin through software that offers extreme audience targeting. With this software I can reach a target audience that looks something like this (and follow them from Website to Website): 55-64 year old men who have had at least one heart attack and are taking preventative medication. They must have a household income of above $80,000 with their mortgage paid down to 40 percent principal. They must have taken at least three vacations, one of which was international in the past four years. They need to live in the suburbs and drive BMWs that are more than two years old. They must like classic rock and have at least three married couple friends in their social circle.
In my mind it’s a short logical leap to also be able to target the color of their underwear.
The fine line between “data” and “privacy invasion” is called PII (Personally Identifying Information). The law goes like this: as long as the data companies obscure the PII so that the target described above is sold to an advertiser as “HIT 55443” instead of “Bill Jones, 1122 Smith Street, Anytown USA” it’s not considered a privacy violation.
Data companies argue that these are aggregated data points, not actual people. However, I can’t help but mention that the actions taken to affect these aggregated data points just happen to affect actual people.
Imagine yourself (we’ll call you “HIT 999”) encountering a financial rough patch during the recession, where you spend a lot of time looking for debt relief and using online calculators to determine how long your savings will last. You also check out a couple of foreclosure lawyers to see if this is an option for you.
Well congratulations HIT 999, you are now tagged with data points that will cause advertisers to continue sending you ads and messages targeted toward you as a member of your data set, which may include compulsive spenders, people with poor financial planning skills, and a low savings priority.
This means your ads may be about luxury cars that you can’t afford, unless you take a predatory loan. It could mean you become peppered with debt consolidation ads, or vacation packages based on payday loans.
It’s funny right? But not so funny if HIT 999 actually IS someone with poor financial planning skills and a compulsive spending disorder. The advertisers can target them precisely to drain what remains of HIT 999’s finances as fast as they possibly can.
So imagine this: If I target an online ad campaign to reach only 17 year old girls on birth control who attend public school and have a GPA below 2.0, listen to Anthrax, and live in my zip code, I can have a pretty direct conversation with them on whatever page they land on. Hmm. What shall I say to them.
And then if they would happen to fill out a form on my Web site that includes a name, phone number, email address, and physical address, well then I have successfully removed the PII, now haven’t I?
I realize that most advertising is relatively harmless, where manufacturers just want to sell some more cars and maybe some ice cream in the grocery store. But in an age of an obesity epidemic, financial irresponsibility, and extreme interaction with the Web, I just feel like someone needs to ring this bell.
Marci De Vries is president of MDV Interactive, a web consulting firm in Baltimore. Reach her at firstname.lastname@example.org.