Analysis: Amazon Fake Review Problem

Comments · 72 Views

Amazon is guided by four principles: customer obsession rather than competitor focus, passion for invention, commitment to operational excellence, and long-term thinking.

Here is my study evaluating how third-party sellers and fake reviewers hampering the recommendation system.

Summary

Amazon generated 53.76 billion in third-party service revenues. While almost all 95% of shoppers use ratings and reviews to evaluate or learn more about products, fake reviews are affecting their ability to confidently turn reviews as a trusted source. Research carried out by “The Washington Post” has determined that 61% of reviews in electronics are fake. Fake positive reviews on poor quality products are also harmful to competitors who offer average or good quality products but do not have so many reviews on them. To avoid being detected by Amazon’s algorithm, sellers do not directly give reviewers of the product link. The vast majority of fake review solicitations compensate the reviewer by refunding the cost of the product via a PayPal transaction after the five-star review has been posted. Sellers advertise that they also cover the cost of the PayPal fee and sales tax. observed roughly 15% of products also offer a commission on top of refunding the cost of the product. The average commission value is $6.24 with the highest observed commission for a review being $15. Therefore the vast majority of the cost of buying fake reviews is the cost of the product itself.

 

 

How it works:

Amazon Sellers typically advertise on Facebook, Instagram touting “FREE” and Sellers offers a full refund for the products but only after buyers have 5-star reviews and typically refund will be either Paypal or Amazon gift card. Sellers deny the refund if the payment is coming from Giftcard.

 

Common way to detect Fake Reviewers:

1) Star Rank: Usually customers on e-business are divided in ranks based on consumption. Based on the investigation, the more the customer buys for commodities, the higher the rank will be. Rank will go down if the customer posts the reviews in less than a day.

2) Review Similarity rate: Some reviewers often copy the other customer’s review with slight changes.

3) Review Relevancy Rate: The review has nothing to do with product description such as link, AD. The higher relevancy rate has more details about the product descriptions.

4) Content length: When the review is too short, reviewers didn’t seriously consider the product or experience.

5)Burst review: Customers review a high volume of reviews for the different products within a short span.

 

Recommendations:

1) Track the long-term trends for rating, reviews, and sales rank. 

2) Comparing the Facebook promotion effect with the post-promotion effect where fake recruiting stopped.

3) Amazon review deletion

 

Conclusion:

Fake reviews are large and fast-moving, with hundred of sellers approaching thousands of potential reviewers every day. Several interesting findings emerge from our analyses. First, we find that products promoted on Facebook groups span a wide variety of categories and they already have many reviews, with average ratings that are often higher than those of comparable products not recruiting fake reviews. Second, we find that the Facebook promotion is extremely effective at improving several sellers’ outcomes such as the number of reviews, ratings (average and displayed), search position rank, and sales rank, in the short-term. However, these effects are short-lived as many of these outcomes return to pre-promotion levels a few weeks after the fake reviews recruiting stops. This is explained by an increase in the share of one-star reviews once the Facebook promotion ends. Amazon is not able to completely eliminate the short-term effects that these reviews have on sellers outcomes. As a testament to this, in 2019 alone, Amazon spent over $500 million and employed over 8,000 people to reduce fraud and abuse on its platform.

 

References:

1)https://link.springer.com/article/10.1007/s00521-020-04757-2?shared-article-renderer

2)https://bearworks.missouristate.edu/cgi/viewcontent.cgi?article=4454context=theses

3)https://www.researchgate.net/publication/303499094_Fake_Review_Detection_From_a_Product_Review_Using_Modified_Method_of_Iterative_Computation_Framework

4)https://ieeexplore.ieee.org/document/8004349

5)https://www.cnbc.com/2020/09/06/amazon-reviews-thousands-are-fake-heres-how-to-spot-them.html

6)https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3664992

7)https://techcrunch.com/2019/08/06/facebook-still-full-of-groups-trading-fake-reviews-says-consumer-group/

 

Comments