The Great Wall of China has more than 9,000 Google reviews, with an average of 4.2 stars. Not bad for one of the most astonishing achievements in human history.
But you can’t please everyone.
“Not very tall. Or big. Just sayin. I kinda liked it. Sort of,” wrote one ambivalent visitor of the structure, which stretches thousands of miles. Another complained, “I don’t see the hype in this place it’s really run down and old … why wouldn’t you update something like this? No USB plug ins or outlets anywhere.” Someone else announced that he’s “Not a wall guy. Laaaaaaaaammme.”
Even Shakespeare can’t escape the wrath of consumer scorn. One reviewer on Amazon awarded Hamlet just two stars: “Whoever said Shakespeare was a genius lied. Unless genius is just code word for boring, then they’re spot on. Watch the movie version so you only waste two hours versus 20.”
It’s no wonder why we live and buy by online reviews: The Washington Post recently reported that a third of American adults use a computer or phone to buy something at least once a week — “about as often as we take out the trash.” Last December, 75 percent of Americans said they would do “most of their holiday shopping on Amazon,” according to CNBC’s “All-America Economic Survey.”
We use reviews to vet our options. In 2016, the Pew Research Center foundthat 82 percent of American adults say they sometimes or always read online reviews for new purchases. And more than two-thirds of regular review readers believe that they’re “generally accurate.”
Marketing data indicates that negative reviews in particular dramatically influence our buying behaviors. But research on the biases and demographics of online reviewers — and our own, often errant interpretations — suggests that our faith in reviews is misguided.
Why we care so much about negative reviews
There are many more positive reviews online than there are negative ones, studies show, which creates a scarcity of negative reviews that we associate with value.
For instance: In a data sample from Amazon, just 4.8 percent of reviews with a verified purchase were rated one star, whereas 59 percent had five stars, according to a study published in 2014 by The Journal of Marketing Research and led by Duncan Simester, a marketing professor at the M.I.T. Sloan School of Management.
“The infrequent nature of negative reviews may help to distinguish them from other reviews,” Dr. Simester wrote in an email. We consequently pay more attention to them.
We also think of negative reviews as windows into what could go wrong. Is this camera’s memory card going to go kaput in the middle of my honeymoon? Are these socks scratchy? Dr. Simester pointed out that people may see negative reviews as more informative, and therefore more valuable, than positive ones because they highlight defects — even if they’re not actually more accurate.
“We want to feel secure in our decision-making processes,” said Lauren Dragan, who analyzes consumer feedback as the audio tech products reviewer at Wirecutter, a New York Times company that reviews and recommends products. We use negative reviews to understand our risk and reduce our losses, studies show.
Plus, after reports that five-star reviews are frequently fake, people may depend on negative reviews more than positive ones because they see them as more trustworthy.
Online reviews are less trustworthy than we think
The credibility of all reviews — even real ones — is questionable. A 2016 study published in The Journal of Consumer Research looked at whether online reviews reflected objective quality as rated by Consumer Reports. The researchers found very little correlation.
Why?
Reviews are subjective, and the tiny subset of people who leave them aren’t average.
People who write online reviews are more likely to buy things in unusual sizes, make returns, be married, have more children, be younger and less wealthy, and have graduate degrees than the average consumer, according to Dr. Simester’s 2014 study. Online reviewers are also 50 percent more likely to shop sales, and they buy four times more products.
“Very few people write reviews. It’s about 1.5 percent, or 15 people out of 1,000,” Dr. Simester said. “Should we be relying on these people if we’re part of the other 985?”
What’s more, reviews are often capricious and circumstantial. For example, the sentiment of travelers’ reviews hinges on their companionship. A study published last fall in Electronic Commerce Research and Applications, looking at 125,076 online reviews, found that people traveling with significant others wrote the most positive reviews, followed by those traveling with friends or family. Reviewers traveling alone or for business were the most negative. Our experiences change depending on our expectations, travel expertise and who we’re with.
People’s motivations also taint their neutrality. Take TripAdvisor’s “Super Contributors,” whose reviews tend to be more negative than those by less active members, according to a forthcoming study from Ulrike Gretzel, a communications professor at the University of Southern California and the director of research at Netnografica. Having formed identities around being expert travel reviewers, Super Contributors may “write more critically to appear more professional,” Dr. Gretzel said. Nevertheless, consumers disproportionately value and trust reviews professing expertise.
Put simply, we should distrust online reviews “because emotions are involved,” Ms. Dragan said.
Another reason to be wary is roughly one in 15 people review products they haven’t actually purchased or used, according to Dr. Simester. These “self-appointed brand managers” write speculative, unsolicited negative reviews to offer the company “feedback.” The problem is consumers are bad at determining which reviews are based on actual experiences and which aren’t, said Dr. Simester. “We are easily fooled.”
Get savvier about how you read reviews
Still, reviews can be helpful gauges when you’re buying stuff — so long as you keep in mind all the caveats around them.
First, weed out the most polarized perspectives. People are much more likely to write reviews if they have extreme emotions about something, said Eric K. Clemons, who teaches information management at the University of Pennsylvania’s Wharton School. This is why you see so many rave reviews and so many rancorous ones.
Even people who don’t initially have strong feelings often develop them in response to survey questions — something called the mere-measurement effect.
“We are socially conditioned to give answers when someone/something asks us a question,” Dr. Gretzel wrote in an email. So if we don’t have a pre-existing, well-defined opinion, we make one up.
When you’re reading reviews, try to find ones that are closer to the median, Ms. Dragan advised. She deliberately looks at three-star reviews first because they tend to be more moderate, detailed and honest. Unfortunately, research suggests that most of us instinctively do just the opposite: We prefer extreme reviews because they’re less ambivalent and therefore easier to process.
Second, ask yourself: “Is this person like me? Are the problems mentioned ones I care about?” For example, Dr. Simester recently bought a pair of ski pants online. He read the reviews and most people liked them, but one guy didn’t. “It turned out his body shape wasn’t the same as mine,” Dr. Simester said, so he disregarded the review.
Dissecting people’s preferences can be useful even if you don’t agree with them. Dr. Clemons, an I.P.A. fan who uses RateBeer.com, said, “If a Scandinavian who really likes lagers complains that a beer tastes way too hoppy, that may mean I should buy it.”
Finally, pay attention to contextual details and specific facts rather than reviewers’ general impressions and ratings. The number of stars someone selects often has “very little to do with” their review text, Dr. Gretzel said. People have different rating standards, and written explanations are inherently more nuanced.
Focusing on the most thorough reviews may also protect against getting duped by fake ones. In experiments where Dr. Gretzel and her collaborators presented both real and fake reviews, readers distinguished between the two better when reviews were longer.
And if you’re still not sure whether a review is fake, scan the reviewer’s profile. Dr. Clemons said that “someone who’s paid to write reviews probably isn’t doing a lot of writing under the same name.” His own research omitted reviews from profiles containing fewer than 10 reviews, “and that took care of a lot of paid nonsense,” he said.
All that said, real reviewers are usually genuinely trying to help: Research consistently shows that people are most motivated by helping others make decisions.
“They feel that they have benefited from other people’s reviews, so they want to give back,” Dr. Gretzel said. “They think it’s for the greater good.”
Originally published on – www.nytimes.com