Can Brands Trust Reviews

Jun 15, 2022
Can Brands Trust Reviews

In our last blog, we talked about how to grow your reviews. Potential consumers are more willing to purchase a product or service with reviews. What’s more, a healthy number of reviews typically leads to better search results, whether on a site like Amazon where more than half of all online product searches start, according to survey data by the digital marketing firm BloomReach. Landing on the first page of results on an Amazon search can drive an explosion in sales. Google search also uses reviews as a signal to increase optimization.

At 4Sight, we also know that consumer reviews hold great insights into how to grow your brand, and how to standout against the competition. Any great research project stands on the shoulders of its data, so a question we must answer first: Is the data robust and accurate?

In the case of reviews, the answer is emphatically yes.

There are have been counterfeit review scandals that have caught the public’s attention, such as with the skincare company Sunday Riley. In 2016, it was found that the employees of the company, at the behest of their founder, made fake profiles on the specialty retail site Sephora to boost reviews and star rating for the company. Sunday Riley settled with the FTC in 2019.

What Precautions Do eCommerce Sites Take?

After the Sunday Riley incident, Sephora introduced a new system for verifying the source of reviews and added badges to highlight reviews tied to purchases. Reviewers can post reviews for promotionally gifted products but must disclose that the product was gifted in exchange for the review.

Yelp is another popular site. They do not allow for solicited reviews of any kind. According the site, their “ghost” algorithm, beginning in late 2018, began aggressively weeding out counterfeit reviews.

Amazon is credited with introducing online reviews to their platform in the late 1990s. They claim 99.4% of reviews are organic reviews. Specifically, their policy is that a supplier can't solicit reviews by email nor can they ask for a positive review. The only pathway to solicit reviews is via Sellar Central, which provides no further consumer insights and no ability to customize the message. About this, Amazon has said:

“We have clear policies for both reviewers and selling partners that prohibit abuse of our community features, and we suspend, ban and take legal action against those who violate these policies.”

On the subject of counterfeit reviews, Amazon’s statement reads that they use “powerful machine learning tools and skilled investigators to analyze over 10 million review submissions weekly, aiming to stop abusive reviews before they are published. In addition, we continue to monitor all existing reviews for signs of abuse and quickly take action if we find an issue. We also proactively work with social media sites to report bad actors who are cultivating abusive reviews outside our store, and we’ve sued thousands of bad actors for attempting to abuse our reviews systems.” In 2019 alone, Amazon spent more than $500 million and employed more than 8,000 people to reduce fraud and abuse on its platform.

Third Party Resources

Aside from trusting the sites themselves, there are independent, third-party tools brands and consumers alike can use to validate the authenticity of reviews. It’s important to note that Amazon claims these sites do not have access to their internal data and therefor not fully accurate.

ReviewMeta is one site, which analyzes Amazon product reviews and filters out reviews that its algorithm detects as potentially unnatural. Fakespot is another whose systems look at language patterns, account creation dates, the types of items that certain accounts are reviewing, and more.

What Else to Consider

Amazon, Yelp, Sephora and other sites are all aiming for removing 100% of the counterfeit reviews on their sites, but if they cannot, there are other aspects to consider.

For example, fake reviews appear to be most common for a certain type of product and category. They tend to be small electronics (think: Bluetooth headphones, speakers) or items like weight-loss pills, and priced within the $15-$40 range. They have an average rating of 4.4 and an average of 183 reviews. Lastly, they’re not typically brand name products and vast majority of sellers are located in or around Shenzhen, China.

Incidentally, counterfeit reviews have been shown to boost sales in the short term, but after legitimate consumers begin to purchase and review the products, ratings, search rank and sales plummeted – typically within 8 weeks of the initial fake reviews.

It’s important to make sure that when performing a reviews analysis, that there needs to be a robust internal QA process and rigorous standards to ensure the reviews are authentic feedback. For example, 4Sight runs proprietary algorithms to identify similar language patterns, stars distribution, and timing. It’s also best practices to limit data from brand websites with uncertain review-acquiring and review-posting policies.

Something else to note: the typical analysis with reviews will have a massive sample size, sometimes with as many as150,000 reviews. With counterfeit or less genuine reviews at < 1%, the insights are not diminished.

It’s also important not to conflate fake reviews with promotional reviews, i.e. reviews that brands use to promote new products. Programs like Amazon’s Vine or Influenster’s VoxBox allow for free or discounted product in exchange for an honest and helpful review.

In our next blog, we’ll take a look at how (and if) these promotional reviews influence star rating and sentiment.