Yelp's answer to fake reviews: a badge of shame

According to a recent Gartner study, brands are increasingly paying for fake online reviews, and by 2014, 15% of all reviews will be fake.

And it's not too difficult to figure out why: research has shown that 88% of consumers turn to online reviews when making a purchase.

For obvious reasons, the scourge of fake reviews is a threat to popular reviews sites. After all, if fake reviews proliferate and consumers start to question online reviews in general, they could fall out of favor.

So one of the most prominent online reviews sites, Yelp, is fighting back. As detailed by the New York Times' David Streitfeld, the company, which went public earlier this year, thinks shaming businesses that purchase fake reviews might be the answer:

Like every Web site that depends on consumer critiques, Yelp has a problem with companies trying to manipulate their results. So it set up a sting operation to catch them. The first eight businesses — including a moving company, two repair shops and a concern that organizes treasure hunts — will find themselves exposed on Thursday.

For the next three months, their Yelp profile pages will feature a "consumer alert" that says: "We caught someone red-handed trying to buy reviews for this business."

Potential customers will see the incriminating e-mails trying to hire a reviewer.

Naturally, the businesses caught in Yelp's honey pot will probably be embarrassed by the badge of shame that now graces their Yelp profiles. But will it work?

The solution, or a can of worms?

Yelp's sting operation may give other businesses considering paying for reviews pause, but it's highly unlikely the company can scale these sting operations significantly. So business owners committed to faking it until they make it will probably assume, perhaps correctly, that the odds they'll be caught and punished are fairly low.

Other services have applied technology-based solutions to the challenge of weeding out fake reviews, but this opens the door to false positives. And without humans at the helm engaging in a law enforcement-like operation, sneaky business owners could theoretically try to frame their competitors. That, for obvious reasons, could create even bigger headaches.

You can't fool all of the people all of the time

So what's the answer? The concern over fake reviews is well-placed, but companies like Yelp may overestimate how much they need to do to protect users. Chances are most of us have spotted a review that looked suspicious and which was dismissed to some extent as a result. The ability of users to filter the wheat from the chaff will, obviously, vary from user to user, but it seems unlikely that every user believes what he or she reads.

Interestingly, businesses that would purchase fake reviews shouldn't underestimate users either: according to a study published earlier this year, the presence of bad reviews may actually improve conversions, suggesting that businesses trying to fool potential customers may ultimately be fooling themselves.

Patricio Robles is a tech reporter at Econsultancy. Follow him on Twitter.

Add your own

Reader comments (13)

  1. Graeme Benge Graeme Benge

    Digital Marketing Executive at Koozai

    5:13PM on 19th October 2012

    It's going to be interesting to see to what extent other platforms follows Yelp's lead.

    Fair play to them.

  2. Avatar-blank-50x50 Ben G

    #marketing #web at Freelancer

    8:39PM on 19th October 2012

    Great article!

    Indeed YELP would definitely benefit from upping their algorithm in order to offer a bit more to trustworthy reviewers. For example, a higher valuation for reviews posted by such trustworthy members could bring more transparency and reduce significantly at least the impact of paid/falsified reviews.

    YELP has great potential to become an industry leader once that the new iOS Maps Application has been reviewed and relaunched.

  3. Avatar-blank-50x50 Frank

    9:27PM on 19th October 2012

    Many websites have people rate the reviews, e.g. "was this review helpful?"

    Although it is not perfect and requires high volume of users, it does help

    I think it is important to have more than one field for a review, e.g. pros and cons. The reviews are much more helpful than "I like it" or "it sucks".

  4. Avatar-blank-50x50 Ben G

    #marketing #web at Freelancer

    9:44PM on 19th October 2012

    @Frank You might a great point. Nonetheless a possible limit to the proposed model is that website visitors do not always want to go through the thought process required to say why a review was helpful or not. An example of this can be found on Ads displayed in Facebook. In the Fb scenario there is a dropdown menu that is shown to the member asking why an Ad does not please them. Not many users now let Facebook know that a certain Ad does not appeal to them because it has become time consuming to do so! Facebook ends up missing out on valuable member info because of this bad UX.

  5. Avatar-blank-50x50 Tim Thompson

    10:13PM on 19th October 2012

    I don't think the badge will stop anyone. If there's a will there's a way. But improving the Yelp review filtering algorithm is probably the better way to improve the results.

    I've seen reviews for local businesses (auto repair) that come from people that don't live within 500 miles and the review says nothing about a vacation. There's other examples like that. Focusing on weeding out these types will improve the Yelp results.

  6. Avatar-blank-50x50 Cassie Anderson

    11:49AM on 20th October 2012

    We think Mark's suggestion - verified identities with some reputation vouching (Klout, Connect.Me etc) - would be an easy way to create trusted reviews upfront.

  7. Susie Samuel Susie Samuel Silver

    MD at VetHelpDirect.com

    11:44AM on 22nd October 2012

    We manage a vet reviews website VetHelpDirect. Each review is checked by a person and we request evidence from randomly selected and suspicous reviews that they are actually clients of the practice. I think these methods actually put of fake reviewers as we have not yet encountered anyone unable to give us evidence that they are a client.

  8. Laura Phillips Laura Phillips

    Digital Marketing Executive at Koozai

    4:53PM on 22nd October 2012

    That's a good idea Susie, I imagine word would get around that you were operating a system like that and the fakers just wouldn't bother. It'll be interesting to see what other ways companies come up with to tackle this sort of thing.

  9. Patricio Robles Patricio Robles

    Tech Reporter at Econsultancy

    6:00PM on 22nd October 2012

    Ben,

    One of the challenges Yelp faces is that business owners specifically target 'reputable' users for their fake reviews. So the assumption that a review is credible because it came from a power user may not be a good one.

  10. Avatar-blank-50x50 Ben G

    #marketing #web at Freelancer

    6:35PM on 22nd October 2012

    The way a power user is defined by the algorith behind review sites such as Yelp can determine if member reviews are worthwhile or not. The strength of your algorithm when well done can yield amazing success rates in flagging users that are brand ambassadors for example.

  11. Avatar-blank-50x50 Sherrilynne Starkie

    4:03PM on 28th October 2012

    Yelp is leading the way on this issue.

  12. Avatar-blank-50x50 Ben G

    #marketing #web at Freelancer

    12:27AM on 29th October 2012

    @Sherrilynne Starkie I would enjoy hearing more about the exact algorithm behind Yelp's review site. Cheers

  13. Avatar-blank-50x50 shane watson

    4:54AM on 21st November 2012

    its really a worthy move at the end of day what matters is customer reviews to our services if they are handled properly a perfect business approach is set

Log in to post a comment