Searching for a Fix for Online Reviews

The five-star online review system has revolutionized consumer purchasing. Uber’s platform and its drivers’ livelihoods hang on it. Online shoppers almost always check the opinions of thousands of strangers before buying. For small businesses especially, a bad online review can be fatal.

The value of just about everything, it seems, can be boiled down to somewhere between one and five stars, as illustrated by the opening of Caroline Beaton’s recent piece in The New York Times, “Why you can’t really trust negative online reviews”:

"The Great Wall of China has more than 9,000 Google reviews, with an average of 4.2 stars. Not bad for one of the most astonishing achievements in human history."

2,300 years old. 21,000 kilometers long. 4.2 stars. (Critics complain about things like no Starbucks or USB charging stations.) An amusing example, if perhaps a bit absurd. My own observation from Amazon reviews is that a great product gets a 4.6 and a horrible product gets a 4.2. Is that really helpful?

Uber’s feedback system was a brilliant success. The system achieved very high response rates by making it difficult to book your next ride until you rated the previous driver. By allowing drivers to rate customers, responsible customer behavior was encouraged. In truth, the firm wouldn’t exist unless it had incorporated reviews into its app. But, some challenges have arisen. As customers became aware that any rating below five stars could result in that driver losing his or her job (drivers are deactivated in many markets with scores below 4.6), courtesy five-star ratings became the norm. Only a horrible experience earned less than five stars. The unintended consequence has been that outstanding experiences couldn’t be identified via the star ratings. Since loyalty happens only with outstanding experiences, maybe it’s time for Uber to consider adding a sixth star, something that can be given only once per month by top riders, and demands a serious verbatim explanation before system accepts it?

More than the star system, the written reviews are where the value is. But wouldn’t it be nice if more sites separated the comments from people who complain about the shipping from product reviews, for example? Or to have some way to knock out the people who have totally silly expectations, or for whom price is the only factor?

Bogus reviews from competitors, quacks and click farms are a big issue. Anonymity makes them hard to root out. Plus, a study of one company by professors at the Kellogg and MIT Sloan schools of management found only 1.5% of the population wrote reviews. That’s a sign of system weakness. 

Rankings may be flawed, but they’re also powerful and ubiquitous, and with some thought, they might reach their higher potential. Reputation, loyalty and recommendation are, after all, at the very core of good commercial relationships, good business and good society.

So it’s worth trying to strengthen online reviews and make them more useful. Airbnb has made some noteworthy innovations. Guests and hosts can now give one another private feedback not visible to the public or to Airbnb, unlocking some really helpful insights.

In my opinion, however, innovation in online reviews has generally stalled. Creative ideas are needed.

Maybe some leading platforms can help us spot like-minded reviewers. Perhaps there are incentives that could coax more of us to leave thoughtful and insightful reviews—time is precious after all—or reward us for providing useful and insightful feedback?

What if every customer or citizen gets a finite supply of rant-and-rave scores to allocate each month? We would benefit from a system that approximates grading on a curve, so we can identify the top and bottom 10% and 1% of experiences. Such online reviews might make it possible to see what products or services truly delighted customer. That would be a system worth five stars.

What creative approaches do you see that could help the star system shine and reach its full potential?