As much fun as old superstitions can sometimes be, very few entrepreneurs would use bad omens such as dodgy horoscopes and broken mirrors to guide business decisions. But what if there were scientifically vetted signals that could help you predict ahead of time which products are likely to tank, saving you a whole lot of time and money in the process?

That’s exactly what new research out of Wharton is promising. Rather than offering tarot card readings or advice to avoid black cats, the study from professors Eric Anderson, Song Lin, Duncan Simester, and Catherine Tucker sought to identify more objective "harbingers of failure" for retail products. The research found them--in the form of customers whom you really don’t want buying your new product.

It’s natural to celebrate when customers buy a product, but be cautious when some sorts of people buy something new on your shelves, the team behind the study warns.

"There is a very real class of customers who are uncannily attracted to products that will never catch on," explains a write-up of the research in Knowledge@Wharton. So who are these customers of doom?

Once a 'Weirdo,' Always a 'Weirdo'

When it comes to life, being a nonconformist can be great, but when you’re doing market research, those with odd tastes should be viewed with a hefty dose of caution. The researchers found that customers who loved one failed product are far more likely to like others that are unlikely to catch on. If you can identify these "bad omen" customers who loved your other duds, you can get an early-warning sign of what future products might fail--just ask your weirdest customers if they like the new thing you’re stocking.

"Take, for instance, Diet Crystal Pepsi, a notorious product failure. Customers who purchase a product like this are also more likely to purchase, for instance, Frito Lay Lemonade, another product that eventually failed," reports the research write-up. Be especially mindful of customers who really loved and repeatedly bought previous flops. "Harbingers with a history of making four or more repeat purchases of a failed product are nearly twice as likely as other customers to buy another product that fails."

Identifying Your Customers of Doom

To avoid costly product failures and put your bad-omen customers' wacky buying impulses to good use, you need to first identify these "harbingers of failure,” advises the Wharton team: "Firms should ask customers not just whether they would purchase the product in question, but what other products they regularly purchase." If the customer’s usual basket of products is thoroughly mainstream, his or her positive opinion of a new product is good news.

But if you find that one guy who is still pining for that weird-flavored beer that collected dust on your shelves for months on end, view his enthusiasm for your latest offering as a definite sign of trouble ahead for the product. "You may not want to launch this product, because it’s probably not going to have the mainstream appeal that sustains products for the long run," cautions Anderson.

Have you identified any customers of doom at your business?