Whenever you make a purchase, you’re paying with more than just money; you’re paying with data. For statisticians like Target’s Andrew Pole, sometimes that data is worth more than gold. They know how to use your data to answer valuable questions.
Statisticians get asked some weird questions, and for Pole, his came from marketing: “If we wanted to figure out if a customer is pregnant, even if she didn’t want us to know, can you do that?”
What? Why would they want to know that? The marketers explained that new parents are simply a goldmine.
Most people go to different stores for different things. Electronics? Best Buy. Clothes? Sears. Toys? Toys R’ Us. So even though Target offers everything, their only visited when a customer needs something they associate with Target. They hate this; a primary goal of Target’s marketing is to become the one store customers think of for everything. But it’s quite difficult because ingrained habits die hard.
Everyone knows having a child is a life changing event; it challenges old habits and changes a persons mindset. And for a company like Target, selling everything from bread to Boggle, any lifestyle change is an opportunity to capture more of that customers business. If Target could find these customers during this period flux, then they could capture customers for the rest of their lives.
“We knew that if we could identify them in their second trimester, there’s a good chance we could capture them for years,” said Pole. “As soon as we get them buying diapers from us, they’re going to start buying everything else too. If you’re rushing through the store, looking for bottles, and you pass orange juice, you’ll grab a carton. Oh, and there’s that new DVD I want. Soon, you’ll be buying cereal and paper towels from us, and keep coming back.”
Timing is everything when it comes to marketing. Births are on the public record, so as soon as a child is born, the mother’s mailbox is flooded with coupons and advertisements. Pole and his team needed to identify expecting mothers long before the birth, and preferably in the second trimester. So they got to work, mining through vast data warehouses and public records. Every shopper who goes to Target gets a unique file tracking purchases, coupon usage, demographics, and much more; which is common practice for large retail companies. They explored past records of customers known to have had a child and looked for purchasing trends among them that could predict pregnancy. Soon the patterns emerged. Unscented lotions and soaps, cotton balls, zinc and magnesium supplements, and vitamins, these are among the 25 products Pole used to build a predictable pattern expecting mothers would follow. It’s the basis for Targets “Pregnancy Prediction” that gives a pregnancy percentage likelihood and even estimates a mother's due date. Soon after this was built, Target started sending ads tailored based on expected due date.
And this is where it gets really interesting, and how this story caught our attention.
One day at a Target outside of Minneapolis, an angry father burst in demanding to talk with the manager. He was furious, clutching a mailer sent to his precious teenage daughter. He didn’t understand how Target could be so immoral as to send teenage girls ads for baby clothes and cribs. “Are you encouraging her to get pregnant?” he asked. The manager looked at the mailer and quickly understood the father’s anger. Cute babies and pregnant mothers on the cover of an ad with coupons for maternity clothes and baby formula. The manager apologized profusely and assured the father he would look into it personally and it would not happen again.
A few days later the manager called the father to again apologize, but instead it was the fathers turn to say he was sorry. “It turns out there’s been some activities in my house I haven’t been completely aware of. She’s due in August. I owe you an apology.”
Target wouldn’t make this mistake again; they quickly realized that people don’t like the feeling of being spied on. So they stopped sending such targeted mailers, but they didn’t stop using the “Pregnancy Prediction” and data mining to gain customer loyalty. So, they started adding a touch of randomness to the mailers. With wineglass coupons next to vitamin supplements and garage door openers next to baby diapers, Target stopped creeping out customers and started winning loyalty.
If you’re interested in learning more about how this type of data mining is done, we just finished a book on this very topic that we're offering you for free! Data Science Secrets explores in greater detail the process's that enable companies to do this and other amazing things with their data. Free for for a limited time, get your copy today by signing up below: