Joe's 'Net Promoter' score drops, but his ranking improves. What gives?

Satmetrix, a company that specializes in measuring customer satisfaction, just released its 2013 Net Promoter Industry Benchmarks study, and Trader Joe’s shoved the much-loved-in-the-Northeast Wegman’s chain off the top of the ‘grocery’ category list. That’s the good news, but the bad news is that while Trader Joe’s rose one position, its ‘Net Promoter’ score has actually dropped.

The Net Promoter score is based on a survey of 20,000 consumers. Each consumer ranks the brands in the survey from 1-10 based on their own arbitrary feelings of satisfaction. The number of consumers who rate a brand from 1-6 is subtracted from the number of consumers who rate it 9 or 10.

Trader Joe’s score was 63%, which suggests that for every person with mild-to-middling negative feelings about the brand, there are two people with strongly positive feelings. That was enough to put it at the top of the grocery category. The average of all grocery stores was a measly 36%, suggesting that overall negatives outweighed positives by 2:1.

Costco scored an impressive 78%, but Satmetrix classifies it as a department store.

Both Trader Joe’s and Wegman’s scores dropped in 2013; TJ’s fell by 10 points, while Wegman’s 20-point drop caused it to fall from first to fourth in the grocery sector. According to Satmetrix, an analysis of the full survey response suggests that Wegman’s relatively premium prices were at fault when customers said they would not recommend it.

Another factor in the fluctuating scores, however, are the statistical limitations of a 20,000-customer survey. That sounds like a significant sample. However, there’s ample evidence to suggest that most grocery shoppers do not travel far to buy food. Trader Joe’s has less than 400 locations, while Wegman’s operates in a strictly limited geographic area. Satmetrix’ report summary doesn’t cite the number of consumers who responded to questions about Trader Joe’s, but since TJ’s sells roughly 1% of the groceries in the U.S., the survey probably only reached a few hundred regular TJ’s shoppers, at most; with such a small effective sample size, it’s likely that the fluctuation is inside the margin of error.