According to the July, 2010 issue of Scientific American, online ratings sites are not exactly fair, with the writer even declaring that their “judgments are inaccurate at best, fraudulent at worst.” Whoa, that’s a pretty hard slap in the face.
What drives this opinion is some pretty good research combined with some good old common sense.
Eric K. Clemons, Professor of Operations and Systems Management at the Wharton School of the University of Pennsylvania, says there are quite a few inherent biases in online ranking systems. The first one is that people who rate a purchase already made the purchase so they already are slanted toward liking the product. The best example of this is consumer reviews of books. In general, people who buy a book when it’s first released already have some knowledge of the writer or subject. They may even be a fan. So the first consumer reviews to come out are normally positive. These high ratings lead other people who may never have considered the book to give it a try. If they end up hating the book, it may be because their expectations were too high. But instead of leaving well enough alone, these dissatisfied readers might post a negative review just to balance out the positive ones.
Then there’s the point that “people tend not to review things they find merely satisfactory. They evangelize what they love and trash things they hate.” That’s why most ratings are not in the middle, despite the fact that true opinions follow a bell curve, where “ratings cluster around three or four, with fewer scores of two and almost no ones and fives.” So basically, those with the most emotional connectivity to a product tend to do the most reviewing.
But it gets worse.
In a 2009 study of more than 20,000 items on Amazon, researchers “found that a small percentage of users accounted for a huge majority of the reviews.” They’re called “super reviewers” and they’re even given a “Top Reviewer” badge by Amazon and then pitted against one another to promote even more participation. But who says this overly enthusiastic crowd is any good at what they are doing? What credentials do they have, other than a lengthy list of products they reviewed? And why do consumers see these reviewers’ badges and think they have some sound advice? In general, 95% of Amazon reviewers have rated fewer than 8 products, yet consumers put less weight in what this majority says.
Fortunately, these “Top Reviewers” are not paid by the companies whose products they review — at least as far as any researchers can tell. Plus, ratings sites in general use automated filters to search out and purge extremely positive or negative reviews, especially if the reviewer has little on his list of review. So while the reviews shouldn’t be trusted, it’s not because anyone is tampering with the system.
But the story points out that there are fair review sites out there, it just depends on the products and the audience. RateBeer.com has 3,000 members who have rated at least 100 beers each, with just about all beers being reviewed hundreds, if not thousands of times. These reviewers are passionate about their beer, and so are the readers of the site. In fact, the reviewers tend to post on all beers they try, rather than just those they love or hate.
So what’s a business to do about review sites? Take them seriously, regardless of how inaccurate they are, and treat review sites as an opportunity to engage readers in your brand. Because while the system may have flaws, most readers do not know it; thus, they read the reviews and believe what they read.
So treat review sites like a dinner party where the conversation turns negative about your company or product, and respond professionally.