LSE - Small Logo
LSE - Small Logo

Andrew Hosch

May 22nd, 2020

Shady sellers: Identifying counterfeits online

1 comment | 5 shares

Estimated reading time: 5 minutes

Andrew Hosch

May 22nd, 2020

Shady sellers: Identifying counterfeits online

1 comment | 5 shares

Estimated reading time: 5 minutes

Can you identify shady sellers of counterfeits on Amazon? Distinguishing between honest and dishonest sellers online is difficult, writes Andrew Hosch. 
You examine the seller feedback, of which there are hundreds or thousands of reviews. They are mainly positive, with only a few complaints. The seller reputation looks good, too; it is above 90 percent. You click the ‘Buy Now’ button, and, like many others, you just got scammed.

Online sellers of counterfeits are not limited to Amazon, and the stakes are quite large. U.S. online retail grew four times quicker than brick and mortar retail in 2019. Given the pandemic this year, that difference likely will only increase. Included in these growth numbers are counterfeit products being sold online by disreputable dealers. From unsafe bike helmets to knockoff car seats, examples proliferate the news. One U.S. government investigation found almost half the items they purchased online were counterfeit. National brands have threatened e-commerce platforms with lawsuits over the prevalence of imitations.

Sellers can hire services to create fake reviews, boosting scores in days instead of what would normally take years

To get beyond the anecdotal stories, we examined seller behaviour instead of the products, as it is the sellers of these fakes that create an untrustworthy market environment. How easy is it to find counterfeit sellers? How effective are consumer protections, such as ratings and comments? Do the dishonest sellers push out the honest? To answer these questions, we examined Amazon.com’s third party seller market, known as Amazon Marketplace. Sellers operating here generated $160 billion in 2018, from which Amazon collected base monthly fees, service plan charges, and a percentage of every transaction.

Finding counterfeit sellers

Our analysis rapidly identified 250 questionable sellers. These sellers had at least one, and usually multiple, complaints in their seller feedback pages about hawking fake goods. Grievances included products with misspelled logos, such as ‘Ralph Laurent’, outerwear with imitation The North Face tags, Canada Goose coats missing Canada Goose patches, and watered down Aveda shampoo that didn’t lather. To find these sellers, from December 2019 to February 2020, we collected over 175,000 seller comments via data collection automation and screened for complaints about counterfeiting, scams, bootlegs, etc. The search focused on sellers in product categories where fake and used-as-if-new goods would be likely, primarily luxury, name-brand products attractive to imitators. A range of product categories was examined, from Automotive to Beauty to Electronics, and included a variety of brands, from Apple (iPhone) and LG (smart phone) to NKG (spark plugs) to Fenty Beauty by Rihanna (lip gloss).

How effective are ratings and review counts?

As seller sophistication has increased, so too has rating and comment manipulation. Sellers can hire services to create fake reviews, boosting scores in days instead of what would normally take years. In the early days of e-commerce, when auction sites such as Ebay and Amazon Auctions dominated, ratings and comments were novel quality markers and touted as effective for building trust online. They acted as analogues for the hands-on experience of examining products in a physical store. High ratings indicated quality customer service. Large review counts indicated longevity and stability.

Given the ease to distort reputations online, it was not surprising to find counterfeit sellers and legitimate sellers having statistically equivalent ratings. Though counterfeit seller’s ratings were slightly lower on average, the difference was too small, less than two points on a 101-point scale, for a buyer to distinguish between a ‘good’ seller and a ‘bad’ seller.

Incentives exist for online sellers to push low-quality, counterfeit, and used-as-if-new goods as if they were genuine, authentic, and of high quality

Surprising, though, was finding a large review count did not indicate a high-quality seller. Counterfeit sellers actually had a larger number of reviews on average. Perhaps since seller feedback is tedious to access on Amazon Marketplace, bad reviews do not act as a deterrent. The only way to find negative feedback is to laboriously page through each and every review on the seller’s storefront page.

Are dishonest sellers price leaders?

In markets consisting only of third-party sellers, counterfeit sellers were price leaders a majority (almost 75 percent) of the time.  However, when Amazon was also selling the same product, these sellers rarely were the market leader (only about 35 percent of the time). Note that our sample was biased by design, so general conclusions such as “a majority of third-party vendors sell counterfeits on Amazon” cannot be drawn. Also, these sellers were less prevalent overall when Amazon was selling the same product. The underlying mechanisms are unclear: we suspect Amazon’s presence as a seller leads to additional monitoring by Amazon which deters counterfeits. What is clear is where incentives exist for online sellers to push low-quality, counterfeit, and used-as-if-new goods as if they were genuine, authentic, and of high quality, dishonest sellers will undercut on price and dominate the market.

Buyer beware

So, what is a shopper on Amazon to do when about to press the Buy Now button? The average buyer cannot rely on seller rating or the amount of seller feedback to choose a ‘legitimate’ seller. And while Amazon’s retail presence on a product page reduces the prevalence of counterfeit sellers, it does not eliminate them. Our study is only a first step, identifying the danger, but we do not have a failsafe way to help people avoid fraudulent sellers of goods online. For now, perhaps the only recommendation is a very old one: caveat emptor.

Here are some of the reviews I came across in my research:

“The frames delivered weren’t real. They were fake. The logo on my frame says “Ralph Laurent” with a ‘t’ at the end.” – 1 out of 5 stars.

“The item was a FAKE/COUNTERFEIT! I rarely use 3rd party sellers and I should have gone with my gut of the other reviews I saw before buying! I’m sure this return process will be a hassle as ive [sic] seen from the other reviews.” – 1 out of 5 stars.

“Very disappointed, I live in Canada and could not find what my 3 year old asked everyday from Santa. A Mr. Grouper (Bubble Guppies) toy to play with. Found this one and paid a premium. Absolute ripoff [sic] if you asked me but the things we will do for our children. The item that came was a 6″ bubble puppy (dollar store quality) Thanks for ruining her Christmas morning, not enough time to get one now!” – 1 out of 5 stars.

Notes:

  • This blog post expresses the views of its author, not the position of LSE’s Department of Management nor the London School of Economics.
  • Photo by Christian Wiediger on Unsplash.
  • Andrew Hosch is currently a student on LSE’s Executive Global Master’s in Management (EGMiM) programme. This blog is derived from Andrew’s dissertation, entitled ‘A Quantitative Analysis of Counterfeit Sellers on Amazon’.

About the author

Andrew Hosch

Andrew Hosch is Vice President of Engineering at a Seattle-based tech startup. He graduates in July 2020 from LSE's Executive Global Master's of Management (EGMiM).

Posted In: Management with Impact

1 Comments