Amazon Problems: Fake Product Reviews

az5starqFor while, I was convinced that one day I would only shop at two places: Costco and Amazon. But recently, I’ve been concerned that Amazon’s quest for growth has hurt their customer-centric reputation. Here’s one part in what unfortunately may become a multi-part series.

Discounts for reviews. In the past, I wrote about a few websites that provided you with discounted products from Amazon in exchange for an “honest review”. I originally felt that since Amazon itself does this type of thing with its Vine program, it should be fine for the makers of a product to send out a few samples for review.

In the months since, I’ve gradually changed my mind. This practice is no longer a way to “jumpstart” your product with a few reviews. Instead, it has basically resulted in a race to the bottom where other sellers feel they must give out tons of free product and end up with hundreds of 5-star reviews with the disclaimer “I received this product at a discount in exchange for an honest and unbiased review.” This screenshot from Reddit is a perfect example of the bottom of this downward spiral:

az_bias

ReviewMeta.com analyzed 7 million reviews and found that the reviews with disclaimers had nearly all 4 and 5 star ratings, with a much lower occurrence of 1 and 2 star ratings.

I think where I messed up was that when a regular website like CNET does a gadget review, they have a readership and reputation to maintain. An anonymous individual is probably busy and just wants to satisfy the review requirement in order to keep getting discounted stuff. There is no feedback system to keep them accountable.

Fake reviews. What’s worse is that some reviews are simply fake with no such disclaimer. As someone who has to deal with spam comments every day, I see the same language patterns in Amazon reviews. Another website called FakeSpot.com also analyzes the quality of reviews:

Fakespot utilizes numerous technologies to validate the authenticity of reviews. The primary criteria is the language utilized by the reviewer, the profile of the reviewer, correlation with other reviewers data and machine learning algorithm that focuses on improving itself by detecting fraudulent reviews.

Check out this Micro SD memory card, and see it’s FakeSpot analysis.

fakespot1

ReviewMeta adjusted their 4.8 star rating down to 1.3 stars:

reviewmeta1

Survivorship bias. As someone who has bought a few “discount-for-review” items in the past (I no longer participate), I can tell you that when I left a less-than-positive review and others did as well, the item simply disappeared from Amazon shortly thereafter. It’s quite easy for a factory to simply slap on another made-up brand name and thus create a separate, new listing. This makes it quite easy to just start over, seed with positive reviews, and keep selling until too many “real” reviews start lowering the average rating. Repeat as necessary.

As always, buyer beware. Amazon has already filed some lawsuits over paid fake reviews, but they still have a lot of room to improve.


<!–
BA
–>

Amazon Problems: Fake Product Reviews from My Money Blog.


© MyMoneyBlog.com, 2016.


Read More: Amazon Problems: Fake Product Reviews

Leave a Reply

Your email address will not be published. Required fields are marked *