Some brands are making names for themselves on Amazon /zigman2/quotes/210331248/composite AMZN +1.04% by spamming their products with fake five-star reviews.
That’s according to an investigation of hundreds of tech products in 14 popular categories conducted by the U.K. consumer advocacy group Which? , which reported on Tuesday that the world’s biggest online retailer is “plagued with thousands of ‘fake’ five-star reviews.”
It found that “unknown brands” with names like ITSHINY, Vogek and Aitalk had thousands of unverified reviews playing up their products, meaning there is no evidence that the reviewers bought or used the products that they’re raving about. The investigation also questioned the authenticity of many of these companies. For the headphone maker called Cquang, for example, the Amazon seller profile was the main Google search result, and the business did not appear on Google at its listed address.
Smaller personal tech products appeared prone to phony reviews. Seven out of 10 (71%) headphones that the researchers scrutinized had suspicious five-star customer ratings, including a pair by the brand Celebrat that had 439 unverified five-star reviews that all posted on the same day.
And when Which? shared its findings with ReviewMeta, the review experts also believed that each unverified five-star review of the top 10 pairs of headphones, when sorted by average customer review, was phony. A rep from ReviewMeta said in a statement to the report that, “I’m shocked we’ve been seeing this so much on Amazon — seems so obvious and easy to prevent.”
An Amazon spokeswoman responded in a statement to MarketWatch that, “Amazon invests significant resources to protect the integrity of reviews in our store for our customers. We are able to assess review authenticity by looking at data points like reviewer, seller, or product history which websites like ReviewMeta do not have access to and therefore cannot concretely determine the authenticity of a review. ReviewMeta itself admits that they do not have the data to know, or even say with high confidence, whether a review is fake ... we have data that disproves with high confidence many of their ‘unnatural’ categorizations because we have the customer history behind each review.”
She also explained that Amazon “investigates suspect reviews, works with social media sites to stop inauthentic reviews at the source, pursues legal action to stop offenders from planning reviews abuse, and feeds new information into our automated systems so it continues to improve and become more effective in catching abuse.” Another rep told the Wall Street Journal (which is owned by the same parent company as MarketWatch) that e-retailer is “fighting a barrage of seller scams on its website,” and has deleted thousands of suspicious reviews; fired workers who gave sellers inside information; and nixed some techniques that might have helped products rise higher on searches when they shouldn’t have.
Smartwatches were also marked by an abundance of suspicious reviews in this Which? report. When sorted by customer review, every device on the first results page was from an “unknown” brand, and unverified reviewers made up 99% of the endorsements for the top four products.
The analysts also found the same suspicious review pattern in action cameras , fitness trackers and wireless security cameras : more than 9 out of 10 of the top-rated products in each category were manufactured by “unknown” brands, which were propped up by unverified reviews.
But larger electronics such as TVs, tablets and radios were more likely to come from large, familiar brands like Toshiba /zigman2/quotes/204149068/composite TOSYY +0.71% , Insignia, LG /zigman2/quotes/204466928/composite LPL +1.58% and Samsung /zigman2/quotes/202367843/composite SSNLF 0.00% , which appeared to weed out the phonies. Indeed, the first several five-star reviews on a 32-inch Toshiba smart LED TV on Amazon come from verified buyers.
The problem is, most online shoppers (97%) say that reviews influence their buying decisions, according to Fan & Fuel Digital Marketing Group , which also found that 92% of consumers will hesitate to buy something if it has no customer reviews at all. And Harvard research has found that every one-star increase in favorability can lead to a 5% to 9% increase in revenue. Or, spamming a product with fake one-star reviews can hurt sales.
And a deluge of five-star reviews doesn’t just raise a product’s rating; research has shown that if two similar products have the same rating, online shoppers will buy the one with more reviews. Around 20 (and running up to 50) is the optimal number of reviews for consumers to feel confident that the item in question has been tried enough by enough people. But only 3% to 10% of shoppers actually post reviews of their purchases.
Amazon isn’t the only online company to be pestered with fake reviews; Yelp /zigman2/quotes/201334325/composite YELP -1.35% and Google /zigman2/quotes/205453964/composite GOOG +0.21% have also had issues with fake endorsements, as well, so software company Algorip has developed a widget for these sites that uses an algorithm recognizes and flag phony reviews.
It’s harder for shoppers to spot a suspicious review, however; 84% of consumers in a 2017 survey admitted that they don’t always catch a phony review. So click here to see 10 ways to spot fake reviews online.