Facebook and Twitter failing to tackle fake review factories, Which? finds

Facebook and Twitter are failing to tackle a fake and incentivised reviews industry that is easily infiltrating both online platforms and trading free items in exchange for bogus five-star ratings to boost the ratings of mediocre products sold on Amazon, a Which? investigation has found.

The consumer champion uncovered Facebook fake review groups with hundreds of thousands of members between them as well as fake reviews agents on Twitter attempting to get artificial positive ratings for tens of thousands of products across more than 130 brands being sold on Amazon.

Between June and November 2021, Which? found and joined Facebook fake review groups with relative ease by searching key terms in Facebook’s search bar such as “Amazon five star review” and “free AMZ” – terms people looking for freebies are likely to use.

There were 18 Facebook groups facilitating the trading of fake reviews that collectively had more than 200,000 members. Some of the groups found had been on Facebook since as far back as 2011 and 2014, despite the company having repeatedly made commitments to the Competition and Markets Authority (CMA), announced in January 2020 and April 2021, that it would crack down on this type of activity on its platform.

The groups targeted Facebook users by offering refunds for Amazon products in exchange for five-star reviews, which is against Amazon’s terms and conditions. Which? also found many more of these groups targeting shoppers around the world but focused only on UK groups in its investigation.

Trading, or facilitating the trading of, fake reviews, is a practice which is likely to be in breach of consumer law. The CMA first started looking at the issue in 2019 so it is disappointing that problems on Facebook continue, with implications for consumer trust in online reviews. More recently the government has also proposed measures for tackling the trading of fake reviews as part of its consumer and competition reforms.

Within minutes of joining the Facebook groups, Which?’s researchers were offered hundreds of free Amazon items in exchange for five-star reviews. Agents shared spreadsheets and documents of products on Amazon that researchers could review, and asked them to select which ones they wanted. Products offered varied widely, from hats and gloves to headphones, webcams, fairy lights, birthday balloons and dog beds.

Once within the groups Which? found agents who claimed they were acting on behalf of Amazon Marketplace sellers, often based in China, India and Pakistan, posting photos of products that they said they needed top reviews for. They often used easily decipherable cryptic messages in an attempt to evade detection, such as: “Ne3d R3vi3w Full Fr33 product”.

When Which? searched key terms in Facebook’s search bar, researchers were sometimes met with a message warning that the term was associated with fraudulent activity – but Which? simply clicked continue to carry on its search, raising questions about how effective such warning messages are.

Which? also uncovered evidence of evolving tactics from the review agents. This included ‘free Amazon products’ being offered on Facebook Marketplace, not just on the groups, as well as the sharing of QR codes within Facebook groups to draw potential reviewers on to other messaging platforms.

Which? researchers, using false names, actually bought four of the items on offer from different agents on Facebook – a pair of bluetooth headphones, a bluetooth speaker, magnetic eyelashes and a webcam with the coveted Amazon’s Choice endorsement and an overall customer rating of 4.2 stars out of 5 (Which? rated it 3 stars).

Researchers gave honest reviews for all of the products, ranging from two to three stars. As expected, none of the agents would reimburse Which?’s experts for the purchases, and they each tried to pressure them to change reviews to five-star. When Which? declined to do so, researchers were told they would only get their money back if they gave five stars. One of the agents said they would only earn commission themselves for achieving five-star reviews.

Facebook said that it had proactively removed many of the groups identified by Which? before it was approached by the consumer champion, and that it then swiftly removed the additional flagged groups that violated its policies.

Which? also found that fake review issues were rife on Twitter in a separate investigation in October 2021. Using a fake Twitter profile to pose as a potential Amazon product reviewer, Which? searched for phrases such as ‘Amazon freebies,’ ‘Amazon free product seller for good reviews’ and ‘free Amazon products for review’, and unearthed dozens of review agents.

Which? got in touch with 30 review agents, and while 11 did not get back to the consumer champion, the remaining 19 did. Plus, dozens of others followed Which?’s fake Twitter profile page within a few weeks of its first tweet being posted about wanting to become an Amazon product reviewer.

In total, 53,065 listings and 132 brands were sent to Which?’s researchers by the review agents on Twitter. Products included beauty items, children’s toys, clothing, earbuds, fake hair, smartphone cases, stationery and underwear. Out of the dozens of products Which? specifically liaised with the review agents about – at least seven had earned the coveted Amazon’s Choice endorsement.

The majority of profiles said they were based in China. Others appeared to be based in India, Pakistan and the USA and were looking for reviewers in countries including Australia, Canada, France, Germany, Italy, Japan, Mexico, the Netherlands, Russia, Saudi Arabia, Spain, Sweden and the UAE.

One Twitter review agent shared a spreadsheet of over 22,000 Amazon listings with Which?’s researchers, and was following almost 800 other Twitter profiles. Another, based in India, told Which?’s team they had ‘more than 10,000’ products for sale on Amazon that they were looking for positive feedback for.

Three profiles were suspended from Twitter for ‘violating’ rules before Which? flagged them as part of its investigation. However, many profiles remained active until Which? flagged them – with one profile listed as having been live since November 2017. Twitter said it has since suspended these profiles though.

Which?’s latest research primarily shows the scale of fake and incentivised review trading on Facebook and Twitter. It also shows that review platforms, such as Amazon, are making it far too easy for unscrupulous firms and sellers to game the system. CMA research suggests that £23 billion a year of consumer spending is influenced by online reviews which is why it is vital consumers can trust them.

Having previously committed to taking action to banish fake review trading, Facebook must now explain why such activity still appears to be rife, and the CMA must challenge the company to provide evidence to show that the action it is taking is effective. Otherwise, stronger action should be taken against the platform to ensure that consumers’ trust in online reviews does not continue to be undermined.

The regulator should also consider investigating Twitter over incentivised and fake reviews.

However, the CMA’s work alone is not enough. The government has proposed measures for tackling the trading of fake reviews as part of its consumer and competition reforms and should bring forward legislation to address such exploitative online practices as soon as possible.

Rocio Concha, Which? Director of Policy and Advocacy, said:

“Facebook and Twitter are failing to adequately tackle fake review factories on their platforms, making it easy for unscrupulous firms and fake review agents to evade weak checks by some of the biggest online platforms and shopping sites. This risks seriously undermining consumer trust in online reviews.

“Facebook must prove that it is taking effective action having repeatedly made commitments to the regulator that it would crack down on fake review trading. The CMA should also consider investigating Twitter over this issue.

“The government plans to tackle fake reviews as part of its consumer and competition reforms and should bring forward new laws to banish these exploitative practices as soon as possible.”

– ENDS –

Notes to editors

Fake reviews: how to avoid them

Read a mix of reviews 

To get a nuanced view of how good a product really is, make sure you read a range of reviews. Which?’s investigations have shown that fakers typically request five stars, so stick to looking at the four star ratings and under. You can choose which reviews appear by clicking the star rating in the customer reviews section.

Check the most recent ratings

Fake reviewers have ways of manipulating reviews so they appear more helpful, and move to the top of the default list where they are most likely to be seen. You can change the sort order from ‘Top reviews’ to ‘Most recent’ to get a more reliable idea of the most recent reviewer experiences.

Look for suspicious review patterns

If there are lots of five star reviews posted on the same day that use unnatural or overly celebratory language then be wary. If a seller gets a bad review followed by a flurry of unrealistic positive reviews, it is possible they are trying to bury the bad ones and bump up their average score.

Look for signs of incentivisation

Spotting fake or incentivised reviews is not easy. But there are things you can look out for – including whether someone mentions being offered a gift card or refund in the review, or shares an unusual amount of photos of the product.

Some fake reviews are sophisticated, and appear natural, but others have some telltale signs to look out for: if it is an unknown brand, for example, with an almost perfect score of 5 star reviews, be wary and do a bit more digging.

Facebook investigation verbatims

Webcam

The £25.99 Amazon’s Choice webcam Which? purchased had a majority of 5 star ratings on Amazon but Which? rated it three stars as it thought it was just OK.

“Dear I need [a] five star review,” said the agent who sold it to Which?. “My boss won’t accept it unless it’s a… five star review.”

Magnetic eyelashes

The agent who sold Which? the magnetic eyelashes, that Which? rated two stars, said: “please give me a 5 star review.”

Which? was surprised to see that the product had a very high rating on Amazon – with a whopping 92% 5 star reviews.

Days later, Which? received an email from another agent offering it a refund plus a £15 gift card if it deleted its negative review.

Which? was then sent two more emails, claiming to be from the seller: “Hello I’m the seller. Please change it to five stars or delete the review. After that, I’ll give you a refund,” they said.

Wireless headphones

The person who sold Which? the wireless headphones asked Which? to amend its review. The product had 4.5 stars from 296 ratings – a high score for headphones Which? felt had relatively poor sound quality (one review called them “absolutely terrible”). He said: “Please try to understand they [don’t] refund without 5 star review and also don’t give me my commission.”

Bluetooth speaker

After purchasing the bluetooth speaker (which had 94% 5-star reviews) Which? gave it an honest review of 3-stars because while the sound quality was surprisingly good, the radio did not work. The agent messaged Which? directly on Facebook asking it to delete the review.

Rights of reply

Facebook (Meta)

“Fraudulent and deceptive activity is not allowed on our platforms, including offering or trading fake reviews. We proactively removed many of the groups identified by Which? before they approached us, and we swiftly removed the additional groups that violated our policies. We’ve been working collaboratively with the CMA to tackle this across our platforms – and Which? research confirms that those measures have been effective. In the last year, we have removed more than 16,000 groups that were trading in fake and misleading reviews. While no enforcement is perfect, we continue to invest in new technologies and methods to protect our users from this kind of content.” – a Meta spokesperson

Twitter

According to a Twitter spokesperson: “We want Twitter to be a place where people can make human connections, find reliable information, and express themselves freely and safely. To make that possible, we do not allow spam or other types of platform manipulation. We define platform manipulation as using Twitter to engage in bulk, aggressive, or deceptive activity that misleads others and/or disrupts their experience.

“Using both technology and human review, we proactively and routinely tackle attempts at platform manipulation and mitigate them at scale by actioning millions of accounts each week for violating our policies in this area. We are constantly improving Twitter’s auto-detection technology to catch accounts engaging in rule-violating behaviour as soon as they pop up on the service.

“We’ve suspended all the referenced accounts for violations of the Twitter Rules.”

Amazon

Responding to the Facebook findings, an Amazon spokesperson said:

“When we detect groups on social media platforms soliciting fake reviews, we quickly report them to that site to have them taken down. For example, in the last year alone, we have reported more than 6,000 abusive groups to Facebook.

“In 2020, we stopped more than 200 million suspected fake reviews before they were ever seen by a customer in our stores, and more than 99% of reviews enforcement was driven by our proactive detection.”

Responding to the Twitter findings, an Amazon spokesperson said:

“Fake reviews mislead consumers, disadvantage small business sellers and foster an underground economy. We innovate and invest to help ensure that only authentic reviews appear in our store. We have clear policies that prohibit reviews abuse and we take action against those who violate these policies.

“We take our responsibility to monitor and enforce our policies seriously, so customers can shop in our store with confidence. When we detect groups on social media platforms soliciting fake reviews, we quickly report them to that site to have them taken down.

“However, Amazon and other retailers cannot do this alone. This industry of fake review brokers needs to stop now. Only when regulators, law enforcement, social media sites and retailers work together will these fraudsters be stopped. We have won dozens of injunctions against providers of fake reviews across Europe and we won’t shy away from taking legal action.

“Customers need to be able to trust the reviews they see online and the systematic manipulation of reviews needs consistent enforcement and global coordination with stronger enforcement powers given to regulators against bad actors. We continue to work to protect the authenticity of customer reviews.

“We advise customers who doubt the credibility of a review on a product to click the ‘report abuse’ link available below each review. We will then investigate and take necessary measures.”

About Which?

Which? is the UK’s consumer champion, here to make life simpler, fairer and safer for everyone. Our research gets to the heart of consumer issues, our advice is impartial, and our rigorous product tests lead to expert recommendations. We’re the independent consumer voice that influences politicians and lawmakers, investigates, holds businesses to account and makes change happen. As an organisation we’re not for profit and all for making consumers more powerful.

The information in this press release is for editorial use by journalists and media outlets only. Any business seeking to reproduce information in this release should contact the Which? Endorsement Scheme team at endorsementscheme@which.co.uk

Press Release