How Target Knew a Teenager Was Pregnant Before Her Own Father Did

Every time you swipe your card at a store, you leave behind a trail of information. Retailers collect it, study it, and use it to predict what you might want next. But Target took this practice to a whole new level, quietly building a system capable of figuring out whether a shopper was pregnant, sometimes before even her closest family members knew.

The story begins with a statistician named Andrew Pole, who Target hired to help the company better understand its customers. His goal was straightforward: identify expecting mothers early, because new parents tend to become some of the most consistent and high-spending shoppers around. Catching them at the right moment meant earning their loyalty for years.

Target assigns every customer a unique ID that tracks everything they buy, along with any personal information the company has gathered or purchased from outside sources. Pole dug into this data and studied the shopping habits of women who had previously signed up for Target’s baby registry.

What he found was surprisingly revealing. Pregnant women, it turned out, followed certain patterns. In the early months, many began buying larger amounts of unscented lotion. Around the same period, purchases of supplements like calcium, magnesium, and zinc went up. Later in the pregnancy, shoppers would often stock up on scent-free soap, cotton balls, hand sanitizer, and extra washcloths.

By combining around 25 of these signals, Pole could calculate what he called a “pregnancy prediction” score for each customer, and even estimate when she was due. Target could then send out coupons timed to match each stage of the pregnancy.

To illustrate how the system worked, one Target employee described a fictional shopper named Jenny Ward, a 23-year-old from Atlanta. Based on a few recent purchases including cocoa-butter lotion, zinc supplements, and a large bag that could double as a diaper bag, the system put her odds of being pregnant at 87 percent, with a likely due date in late August.

The accuracy of the system eventually led to a moment that reads almost like fiction. A man walked into a Target store near Minneapolis, furious, demanding to speak with a manager. His teenage daughter had received mailers filled with ads for cribs, maternity clothing, and baby products. He wanted to know why Target was seemingly encouraging his high school daughter to get pregnant.

The manager had no answers and apologized on the spot. He called back a few days later to apologize again. That time, the father’s tone had changed completely. He had spoken with his daughter in the meantime. It turned out she was, in fact, pregnant and due in August. He was the one apologizing now.

Target quickly realized that being this accurate made people uncomfortable. If customers felt they were being watched and profiled without their knowledge, they pulled back. So the company adjusted its approach.

Rather than sending mailers packed exclusively with baby products, Target began mixing those ads in with completely unrelated items. A coupon for diapers might appear next to one for a lawn mower. An ad for infant clothing might sit beside a promotion for wine glasses. The randomness was entirely deliberate, designed to make the targeting invisible.

As one Target insider explained, as long as a pregnant customer did not feel like she had been monitored, she would use the coupons without a second thought, assuming everyone in her neighborhood received the same mailer.

Target’s revenue grew from 44 billion dollars in 2002 to 67 billion dollars in 2010, a period that coincided directly with Pole’s work and the company’s push to attract expecting parents. Company leadership publicly celebrated this focus on specific customer groups, including mothers and babies, as a key part of their growth strategy.

When journalist Charles Duhigg began investigating the story for The New York Times, Target refused to cooperate. When he flew to their headquarters anyway, he discovered he had been added to a list of prohibited visitors.

The broader lesson here is one that applies far beyond Target. Nearly every major retailer is doing some version of this. Your shopping history, your loyalty card, your online behavior, all of it feeds systems designed to know what you want before you do. For anyone uneasy about being this closely watched, the most reliable workaround remains one of the oldest: pay in cash.