analysis
Kate Emery: Why it’s time to review the five-star review

If you catch a lot of Ubers you’ve probably given out a lot of five star reviews. That’s not because Perth customers are particularly generous or Perth Uber drivers are gifted with excellent manners and the ability to make it from the CBD to Fremantle in 20 minutes at rush hour. It is probably for one of a number of reasons. Possibly you have heard that drivers who are consistently scored below five stars can get penalised. Possibly you know that a rating below five stars means a follow-up question on what was wrong. Hit five stars and the job is done. Or possibly you feel that, in the world of online reviews, five stars is increasingly seen less as a sign of exceptional service and more a standard. If so, you are not alone. The tendency for consumers to give five-star ratings indiscriminately is so common that researchers have dubbed it “the positivity problem”, suggesting the trend robs reviews of their value. One estimate suggests 90 per cent of all Uber reviews are five stars. More than half of all Amazon reviews are five stars. In May, US author Sarah Stusek went viral on TikTok after lambasting a reviewer for giving her “only” four stars in a Goodreads review of an upcoming book. “She said it was ‘really great’ but ‘the ending was kind of predictable’ and yeah, well, it’s my life not a f..king murder mystery ‘but other than that it was incredible’ so you just gave me four stars why?” the author wrote. She has since been dropped by her publisher. If consumers are feeling the pressure to give even average service five stars, blame the companies behind them. Businesses from food delivery services, rideshares and hotels, to retailers and mechanics encourage people to regard a five-star rating as standard via follow-up prompts that encourage them to explain why the experience fell short. Some take the five-star review system seriously enough to penalise workers who score less than a nearly five-star average. That can include anything from the Perth car dealership where a customer was told by a mechanic that anything less than five stars would be harshly judged by their boss, to rideshare companies that penalise drivers for a rating less than high fours. Perth driver Alex Ironside, who works for Uber and DiDi, said Uber recently introduced a $500 bonus for drivers who maintain a star rating of at least 4.85 for three months and earn a certain number of points by accepting rides. The previous main incentive to keep ratings high was to avoid penalties, such as a reduced fuel discount. “Most people are nice, but it’s quite easy for a driver not to get a five-star rating,” he said. “It could be completely not the fault of the driver, but they have to wear that rating. It doesn’t take many bad ratings to take a driver down. Mr Ironside, who has a 4.86 star rating, said just three or four bad reviews out of 500 could have an impact. The Sunday Times spoke to Perth rideshare users and most said five stars was their standard rating, unless something was seriously wrong. Those who routinely rated their drivers five stars mostly fell into two camps: those who did it “so I don’t have to think about it” and those who did not want to endanger the driver’s rating. “It’s their livelihood,” said one in the latter camp. Another said: “There would have to be something quite serious to prompt me to give a less than five-star rating. I guess in terms of rating an Uber trip I’m kind of all or nothing, not interested in the nuance of 2-4 stars.” “My drivers have always been fine,” said another. “Never amazing, but do what I need them to, so I figure help a brother/sister out. Would give less if they creeped me out or murdered me.” In an episode of the dystopian science-fiction TV series Black Mirror, Bryce Dallas Howard plays a woman obsessed with bolstering her social status. In her world, people rate not just products and services, but each other, offering one to five stars based on positive or negative interactions. These can be converted into benefits, such as better housing options. That technology remains, for now, a cautionary sci-fi tale. But only just. Social media encourages people to rate each other via “likes” shares and follows, which can be cashed in for paid endorsements. The commodification of criticism online has ushered in the era of the influencer, not only stars such as Kylie Jenner and Selena Gomez (both are reported to earn north of $1 million for a single Instagram post endorsing a product), but up to 30 million others who consider themselves influencers. Meanwhile, the pressure to constantly rate the world is real. Pay for a shirt online and the company will follow up with a request to review the “service experience”. Take that rideshare, book an Airbnb or buy something on eBay and you will be asked to do the same. There are websites where students can rate teachers, patients can rate doctors and any author can torture themselves on Goodreads. Technology has turned anyone with an internet connection into a critic. It has also opened the door to scammers or activists to tank a business’ rating for profit or malice. One recent scam targeting restaurateurs involved making multiple one-star reviews online, before requesting money to remove the ratings. Closer to home, Perth chef John Mountain believes he was targeted by vegans online via multiple poor reviews after one vegan customer had a bad experience. His subsequent decision to ban vegans made international headlines and ultimately led to vegan activist Tash Peterson being charged with trespass and disorderly behaviour after a series of fiery clashes in front of diners. Online, his Connolly restaurant Fyre is a battleground of competing for one and five-star reviews from detractors and supporters who may or may not have ever eaten there. ***** The idea of starred reviews dates back to the 1840s and the popular British Baedeker travel guide, which used such a system to rate various tourist attractions. Nearly 180 years later, the five-star review isn’t ubiquitous — numerical ratings or a simple thumbs up v thumbs down system remain popular for some professional critics or sites such as YouTube — but it is the overwhelming standard online. It comes with limitations, the most obvious of which is that it tends to be good at separating the very good from the very bad, but is not so good at differentiating good from great. That was the conclusion of Josh Breinlinger, Andrei Hagiu and Julian Wright, writing in Harvard Business Review. The trio found most users either opt for five stars or one. There is also the issue of what researchers behind a 2021 study published in Nature Human Behaviour dubbed “the positivity problem”. They found 80 per cent of online reviews were rated four or five stars. “Across disciplines, behavioural scientists are beginning to recognize the problematic nature of these ratings,” they wrote. “The high degree of positivity effectively makes the ratings ineffective signals for discriminating what are likely to be the best or most successful options.” There is also a distinction between starred reviews that include comments and those that do not. The former offers valuable context. TripAdvisor users discouraged by a one-star review for the Louvre Museum in Paris, for example, may wish to learn that a now infamous reviewer never made it inside. “We didn’t get in … the tour operator made a mess and we didn’t get tickets,” they wrote. One popular podcast, Beach Too Sandy, Water Too Wet, highlights some of the more ridiculous reasons behind online one-star reviews, from the nude beach with “too much nudity” to the lack of wi-fi available. . . in a cave. Violetta Wilk, a lecturer and researcher in digital and social media marketing at the School of Business and Law at Edith Cowan University, said the most helpful reviews were those that were neither great, nor terrible. “We’ve done some research in this area a couple of years ago looking at the helpfulness of ratings on Zomato, which is a restaurant review platform, and what we had found was that consumers who were interacting with the reviews, that is trying to find somewhere to go, would find the three or four-star ratings more helpful than the one, two and five-star ratings,” she said. “The tables have turned now where I think what’s happening is consumers are increasingly visual creatures, so looking at five stars means ‘that’s the right score, I’ve validated my choice’ without looking at the detail. I think we’ve become increasingly visual, but also lazy at interacting with information: we don’t want to be processing a lot of text, we just want that quick and easy information validation, which the star rating provides us. “It’s a real shame. As brands and businesses we’re competing for attention and consumers have become complacent … and I think this will increasingly be the case as we rely on digital communication and digital technology for our information. It’s very different to what we had found a few years ago when those three and four-star reviews were more helpful and received more engagement, via likes, rather than the five-star ratings, which were seen as less authentic and less trustworthy.” If the five-star review is going nowhere, can it be improved? Dr Wilk thinks so. Adding useful context, such as showing a reviewer’s average score or the distribution of scores could, she said, help to distinguish both “the consistent five-star rater” and someone who is consistently negative. That five-star review of a restaurant might not look so valuable if you know the person who left it has never rated any venue less than five stars. Breinlinger, Hagiu and Wright had similar thoughts, suggesting showing the distribution of scores within a category would make it easier to see how a supplier or product compares to others, or to indicate where suppliers are in the top percentile, the way Airbnb does with “superhosts” and eBay with “top-rated seller”. The more we outsource our decision-making to online reviews, the more pressing the need for a better system becomes. However, the limitations of the five-star system have been known for some time and little has improved. Five years ago a post from a San Francisco Lyft rideshare went viral for a chart taped to the back of the front passenger seat, claiming to explain the rideshare company’s rating system. Five stars “got me where I needed to go”, while four stars translated to “this driver sucks, fire him slowly”. Three stars meant “this driver sucks so bad I never want to see him again”, two stars suggested “maybe the car had something dangerously wrong with it” and one star indicated “threats or acts of violence possibly made”. It was an exaggeration for comic effect. But not, perhaps, by much.

analysisKate Emery: Why it’s time to review the five-star review

analysis
Kate Emery: Why it’s time to review the five-star review