The Trouble with Five Stars
Rating systems were supposed to make choosing easier. Instead of relying on advertising, we could see what thousands of other customers thought of a hotel, a phone charger or a plumber. In principle, that is a remarkable shift of power from sellers to buyers. In practice, anyone who has scrolled through online reviews knows that something has gone wrong.
The first problem is inflation. On many platforms, the vast majority of ratings cluster at four or five stars. When almost everything scores between 4.6 and 4.9, the number stops carrying much information. A difference of one tenth of a star may reflect a single unhappy customer rather than any real difference in quality, yet it can decide which product appears at the top of a search.
Part of the reason is social rather than technical. Many people feel that anything below five stars is a punishment, particularly when they are rating an individual, such as a driver or a host, whose income depends on the score. Giving "only" four stars to a perfectly adequate service feels mean, so people round up. Over time, five stars comes to mean "nothing went wrong" rather than "this was excellent".
The second problem is who writes reviews at all. Customers who were delighted or furious are far more likely to share their opinion than those who were merely satisfied, which pushes ratings towards the extremes and leaves the ordinary experience underrepresented. Add to this the reviews written or paid for by sellers themselves, and the picture becomes even less reliable.
None of this makes reviews useless, but it does change how they are best read. The written comments are usually more informative than the average score, especially the moderate ones, which tend to describe specific strengths and weaknesses. Looking at the three-star reviews, rather than the five-star ones, is often the quickest way to discover what a product is really like.