Kashmir Hill at Forbes has a great article, "How Target Figured Out a Teen Girl Was Pregnant Before Her Father Did." Short summary: the company looks at buying patterns and can estimate the probability that a regular customer is pregnant. It sent coupons to a 17-year-old high school student. The girl's father complained to the store manager about coupons that might encourage a high school student to get pregnant. A few days later the father has a talk with the girl–and learns she is, in fact, pregnant.
I've written about the need for data analytics (Big Data: Analytics Are Essential to Corporate Profits, Growing Need for Business Analytics, and reviewed Competing on Analytics).
The Forbes article taught me something new:
What Target discovered fairly quickly is that it creeped people out that the company knew about their pregnancies in advance.
Their solution was pretty clever: send out a personalized coupon books that has prenancy-related products interspersed with other products, so that it does not appear that Target knows the customer is PG.
If you sell stuff, you are competing against companies that are pretty sophisticated in their analytics. You need to get with it, or fold up your tent and retire.