Vote 2020 graphic
Everything you need to know about and expect during
the most important election of our lifetimes

Study on Amazon and Fake CDs Makes Sweeping Claims After Testing Fewer Than 100 Albums

Illustration for article titled Study on Amazon and Fake CDs Makes Sweeping Claims After Testing Fewer Than 100 Albums
Photo: Ross D. Franklin (AP)

It’s no secret that Amazon struggles to keep fake products off its online marketplace, and it has for a while now. So when a new study made headlines this week for supposedly finding that a quarter of CDs “Fulfilled by Amazon” are knock-offs, it wasn’t exactly a shocker. But this alarming statistic becomes decidedly less so when you realize how many counterfeits these researchers actually found: nine. In total.

Advertisement

The Recording Industry Association of America recently released the results of its annual study on the state of counterfeit and pirated goods, and quite frankly their methods seem a little less than scientific. After all, the organization only purchased 80 CDs for the test, according to the report.

Nine of these CDs were fakes, roughly 11 percent of those sampled, and the study only looked at albums released by U.S. record labels. The report doesn’t specify how many of these 80 CDs were “Fulfilled by Amazon” purchases, only that 25 percent of those that were ended up being fakes. The company’s Fulfillment by Amazon program allows third-party vendors to sell their wares on the site and use Amazon’s distribution centers for shipping.

Advertisement

So, based on this examination of fewer than 100 CDs, the RIAA then extrapolated their findings to the entire stock of “Fulfilled by Amazon” CDs worldwide. Now I’m not a data analyst, but considering Amazon receives 5 billion product updates every day, this sample size seems a tad small.

The RIAA’s findings concerning counterfeit CD sales on eBay came from a similar methodology (only 79 purchased this time around, though). So you may want to take the organization’s findings with a grain of salt. Gizmodo reached out to the RIAA for comment and will update this article with their response.

We reached out to Amazon too, and while they didn’t immediately respond, one of the sites that originally covered this story, Digital Music News, reported that the company provided the following:

Our customers expect that when they make a purchase through Amazon’s store—either directly from Amazon or from one of its millions of third-party sellers—they will receive authentic products. Amazon strictly prohibits the sale of counterfeit products and we invest heavily in both funds and company energy to ensure our policy is followed.

Advertisement

The full statement is ridiculously long, but you can check it out here.

So in short:

Does Amazon have a problem with third-party sellers hocking pirated goods in its marketplace? Absolutely; everything from knock-off books to fake Snuggies to pairs of eclipse glasses that are absolutely not safe to wear during an eclipse have been sold on the platform. It’s a problem so pervasive, Amazon sued two of its vendors in 2016 for allegedly selling counterfeits in effort to crack down on the practice.

Advertisement

Does Amazon have a problem with third-party sellers hocking pirated CDs specifically? Maybe. But these RIAA’s statistics are based on a sample size that hardly seems to reflect the company in its billion-dollar entirety, and so we can really only apply them so far before things get skewed beyond usefulness. In any objective sense, at least.

Gizmodo weekend editor. Freelance games reporter. Full-time disaster bi.

Share This Story

Get our newsletter

DISCUSSION

jkmcsw
Hans Bubbe

The fact that the sample size is small relative to the population size does not make the sample inherently bad - that’s a common misconception. The absolute sample size is really all that matters; a sample of size 100 is essentially just as good for doing inference on a population of 5 thousand or of 5 billion CDs. Now I’m with you in thinking that the study is fishy, because - regardless of sample size - it doesn’t seem like they went to any pains to get a representative random sample (e.g. they only looked at albums on US record labels).