The Big Hidden Problem With Uber? Insincere 5-Star Ratings (original) (raw)

Last night, I rode home with a stranger. He shot me a cursory glance in the mirror as I climbed into the backseat, but otherwise didn't acknowledge me. For the next 13 minutes, he jabbered away on a hands-free phone and I sat white-knuckled as we whipped around corners and barreled through stop signs.

It was a five-star experience.

Or at least that’s what I reported to Uber when prompted to rate my ride. It may have been slightly harrowing, but I had been chauffeured door to door without incident---for half the cost of a cab and infinitely less hassle than the bus. Who was I to nitpick? Not to mention, this guy now knows where I live.

I’m not the only undiscerning reviewer out there. Like many transactions in the online economy, "perfect" has become the default rideshare rating. The majority of customers give 5 stars across the board, reserving one-star scores for only the most egregious experiences. There’s hardly any middle ground.

In an industry built entirely on trust between strangers, this is an alarming notion. The digital equivalent of a “Good Housekeeping Seal of Approval,” peer-to-peer platforms tout customer ratings as a fail-safe indicator of safety and satisfaction. They implore us to open our homes, hitch rides and leave our pets with people we've never met---all based on the aggregate endorsement of other strangers. And I do so, regularly and without pause.

Yet, despite a prevailing code of transparency and trust, these companies aren't exactly encouraging honest feedback. For instance, Lyft explicitly tells customers “anything lower than 5 [stars] indicates you were somehow unhappy with the ride.” There’s nothing wrong with striving for excellence, but, in essence, a 4-star review is a black mark on the record of your hardworking and arguably underpaid driver. No pressure, right?

Uber is even more emphatic about maintaining a five-star fleet. A few low scores can reportedly land drivers in costly remedial classes where they learn not merely safe driving techniques, but the finer points of earning flawless reviews, like providing riders with bottled water and maintaining good hygiene. Those who don’t boost their scores risk suspension or permanent deactivation.

With so punitive a code, it’s no wonder so many riders dole out inflated ratings. Besides, judgment is a two-way street. Ratings are solicited from both parties in each transaction. While the feedback is supposedly anonymous, customers feel compelled to pump up their reviews and hope they will be rated in kind. After all, no one wants to be flagged as difficult, especially when it is unclear how such data could be aggregated and used in the future. (Imagine a world in which your worst interactions became data points in a cumulative customer profile sold to airlines, hotels or other companies.)

That’s not to say rideshare ratings systems are all bad. On the upside, both Uber and Lyft require users to leave ratings after the ride ends and before they can book the next one, eliminating the trolls and selection bias that plague many other online review mechanisms. This approach certainly is better than the sector status quo. Complaining about a New York City yellow cab incident, for instance, requires an official report, corroborating evidence and testifying at a Taxi and Limousine Commission hearing.

But in my experience, the lack of ratings differentiation makes it difficult to distinguish between exceptional and marginally acceptable service. I take an embarrassing number of Uber and Lyft rides each week. I’ve ridden with 4.7 star drivers who wear gloves and open passenger doors and 4.7 star drivers who couldn’t pass a road test. It may seem a petty distinction, but in some cases, inflated ratings could threaten customer safety.

So what should a perfect score look like? I have no idea. That’s probably how some companies would like to keep it. More nuanced review systems would paint a more vivid---but perhaps less upbeat---picture of customers’ experience. For companies with a large and loyal base of users, the branding benefits of universally perfect reviews outweigh the costs of the occasional horror story.

Yet, community feedback is a critical mechanism for regulating the peer-to-peer economy. In some cases, where the regulatory landscape has not kept pace with technology, it is the predominant mechanism. As such, companies have a responsibility to continually assess their review systems and ensure they accurately reflect safety and satisfaction. Doing so would go a long way toward building trust with customers and the municipalities in which they operate. For their part, customers should think critically about what the ratings they both give and rely on actually mean---especially when riding with strangers.