Evan Schuman's StorefrontBacktalk

But wait—why exactly is there such a big gap between Avon’s Web and mobile satisfaction ratings? It’s mainly because Avon’s Web site is tied with Apple’s at #2 for satisfaction. Having a middle-of-the-pack mobile site and a great Web site should discourage customers from coming back via mobile or any other channel, according to the theory.

Except it doesn’t. According to more of ForeSee’s survey numbers, Avon’s middle-of-the-pack mobile site gets a middle-of-the-pack rating when customers are asked if they’ll return via mobile. But asked whether they’ll purchase from any other Avon channel in the future, and Avon came in at #3—right behind Amazon and Apple.

OK, having a big gap between mobile and Web satisfaction isn’t necessarily deadly after all. But it certainly doesn’t help. Wal-Mart’s sites also have a big gap. Its mobile satisfaction was rated low, but its Web score was middle-of-the-pack. The predictable result: Wal-Mart was rated by customers as one of the least likely retailers they’d return to through any channel in this survey.

Still, having a satisfying mobile site will help drive customers to other channels, won’t it? Not consistently. Dell and Netflix, which scored right behind Apple and Amazon for mobile satisfaction, were near the bottom of the list for customers’ likelihood of returning via another channel.

Having a great mobile experience will consistently line up with customers wanting to come back to the M-Commerce site, right? Don’t tell eBay and Best Buy, whose mobile satisfaction scores were almost identical—but although eBay tied with Apple as the top mobile site customers said they’d return to, Best Buy’s mobile customers rated their likelihood to return near the bottom. (Then again, Best Buy rated high for customers returning by way of other channels.)

That’s the problem with these types of metrics: Although there may be some very general correlations, the specifics frequently overwhelm the usefulness of those rules of thumb. Amazon will keep pulling customers back even if its mobile site is rated much less satisfying than its Web site. Dell and Netflix’s mobile sites are a strong draw for bringing customers back to their mobile sites—but not so much for their Web sites.

And Target, whose M-Commerce and E-Commerce sites rank near the bottom of the satisfaction list, still ranks high for “other channels”—presumably because people still want to go into the brick-and-mortar stores.

None of this means retailers shouldn’t make their mobile sites better and more satisfying to customers. Of course they should. Retailers should make their Web sites and stores more satisfying, too, especially in ways that drive customers to whatever channel is handy when they want to buy.

And it doesn’t mean analysts shouldn’t keep squeezing insights out of survey numbers. Someday they might make a breakthrough.

Until then, any theory more complicated than “make your mobile, Web and store experiences better and your customers happier” is likely to get you into trouble.