Earlier this year, I was inspired by a David Streitfeld story in the New York Times to write a post about all those "5-star" reviews littering the Internet (blog post here; Streitfeld's story, here). Mr. Streitfeld has done it again, with an article in today's Times about an attempt by Yelp to halt deceptive reviewing practices.
Yelp, an online hub for reviews of local businesses (dry cleaners, restaurants, jewelry stores, you name it), realized that it had a problem with false reviews, and set up its own sting operation.
For every five new notices that are submitted, one is determined by internal filters to be so dubious — either highly favorable or highly critical — that it is banned to a secondary page, which few users bother with, instead of appearing on the business’s profile page. Many of the reviews tagged as fake are written by people new to Yelp.
As a result, some businesses have sought out Yelp's so-called "elite" reviewers -- those who have a track record of posting reviews -- and offered cash payments for positive reviews.
How do you find an "elite" reviewer? On Craigslist.
So a Yelp employee went to Craigslist itself, and posed as an elite reviewer.
The initial sting uncovered eight businesses ready to pay for excellent reviews. Which, given that Yelp has a reported 30 million reviews, doesn't sound like a lot. (A Yelp vice president is quoted as saying, "It is safe to say that this is just a sample" of businesses seeking to influence their status.)
Payments reportedly ranged from $5 to $200; reviewers were sometimes given boilerplate copy to post and sometimes asked to write their own (positive) copy.
For the next three months, the Yelp profile pages for those eight businesses will features a "consumer alert" that says, "We caught someone red-handed trying to buy reviews for this business."
This kind of public shaming may help, but I'm not convinced that it will make a real dent in the deceptions. A Cornell doctoral student in computer science who has been studying reviewer deception noted,
My intuition is that public shaming would increase the risk and therefore the cost of posting fake reviews, which could reduce the prevalence .... [But:] What’s to stop someone from going and soliciting fake positive reviews for a competitor’s restaurant, in order for them to be publicly shamed?
Let me reiterate my earlier closing:
Maybe we should rethink hive-mind reviews altogether. I have more confidence in the Times' restaurant reviews than in Yelp, and more confidence in what Consumer Reports says than in glowing reports on Amazon. Because even if I don't know those reviewers personally, I know what they stand for.