It was a rainy Tuesday morning when the man approached Target’s support desk. In his hands was a pile of coupons. He was middle-aged with short hair, and he was visibly upset.
“Where is your manager?” he demanded.
The employee radioed for help.
When the manager arrived, the customer held up the coupons in a display of frustration.
“Why did you send these to my daughter?” he asked. “She’s still in high school! You’re sending her coupons for baby clothes and cribs? Are you trying to encourage her to get pregnant?”
The manager was caught off guard. Though bizarre incidents are common in customer service, this was a bit further out there. The manager promptly apologized and said he’d look into it.
Then, the manager followed up three days later.
Things changed that day. The angry customer’s tone had softened, with hints of remorse, “It turns out there have been some activities in my house I haven’t been completely aware of. She’s due in August. I owe you an apology.”
The crazy thing? His daughter hadn’t been shopping for any pregnancy products.
How This Technology Began To Take Hold
Pregnant women are a coveted demographic.
It’s a pivotal life change that upends a mother’s spending preferences. She’ll try new products and spend more money than before. Marketers know spending habits are very hard to change. Childbirth is their moment.
Target knew that if they could get a pregnant woman to come to their store for pregnancy items, they’d do a lot of other shopping there. Mothers are often too tired to go to multiple stores (as most people do).
Target figured out how to exploit this.
The Statistician Becomes a Data Magician
Andrew Pole was at his desk on a Monday morning when a fellow employee approached him and asked a very random question, “Would you have a way of figuring out if a woman is pregnant?”
Foregoing the opportunity to make a joke, he responded, “I could try.”
Pole is a talented data scientist, with a master's degree in statistics. His parents were both math teachers.
He started his project by pulling data on spending habits for millions of customers.
Each time you buy from a company, you are assigned a customer ID. It’s like a social security number, a data point that can be exported into various programs for analysis.
After seeing the data, Andrew realized there was a huge opportunity. “As soon as we get them buying diapers from us, they’re going to start buying everything else too.”
How He Pulled Off the Math Magic
Pole pulled the previous year’s worth of data on each expecting mother.
He found that women often buy lotion around the start of the second trimester to prevent stretch marks.
Then, when they’re closer to giving birth, they tend to buy cotton balls, scent-free soap, hand sanitizers, and washcloths.
This, along with 25 other product data points, allowed him to triangulate cross-sections of buying behavior. He created a Pregnancy Predictor score.
Eventually, Target could tell a woman was pregnant with greater than 80% efficacy — even if she’d never bought a maternity product.
When they rolled out this ad campaign, it was so effective that it creeped out many female customers, “How does Target already know I’m pregnant? I haven’t told anyone yet.”
The letters that were headlined, “CONGRATULATIONS” proved too bold. So they pivoted and were more subtle with their messaging.
For example, in their online ads, they even mixed in ads for products they knew a pregnant woman would never buy — just to make themselves look less stalkerish.
The payoff has been huge.
Upending the World We Live In
“Know your demographic” has long been a staple for advertisers.
Yet it’s now gone to a place not even advertisers could have predicted.
There are entire fields of science around habit formation at major universities, with massive grants being dolled out to professors. It isn’t without merit. More than 40% of the things you do daily are an iteration of a habit.
Corporations are headhunting mathematicians from major universities. Then, they are buying huge swathes of data about you regarding your income, the shoes you buy, your birth records, everything. Finally, they turn that data over to their number crunchers and figure you out.
Target’s “Guest Marketing Analytics Department” has played a key role in their $23 billion increase in its revenue in only eight years. They’ve used your history to predict your future. And they’re only getting better at it.