We Need to Rethink How We Evaluate Reviews

Nobody likes to guess at a product or service's quality before spending money on it. That's one reason we've all become so reliant on online reviews. When they drive so much of our behavior, however, we owe it to ourselves to think critically about how we use that information.

According to new research from England's University of York, more than 4 out of 5 of us think checking reviews is an important part of booking a hotel room. Yet nearly one-third of online reviews are actually fake — and we tend to believe reviews without question, especially when they align with our prejudices. If you see a bad review for a luxury hotel, that will seem out of the ordinary, but a bad review at a roadside motel must be true by default. Unsurprisingly, the study found that "good" or "nice" lodgings were less likely to be affected by a poor review than "bad" or "budget" rooms.

Falling back on confirmation bias is a nearly universal human trait: When we taste-test wine, for example, we rate it better if it comes from a famous wine-producing region, even if other wines taste better in a blind test. We also make decisions about other people's skill or competence based on their politics. The York researchers have some good advice to counteract any of these tendencies: Look at overall reviews or personal records, rather than cherry-picking the most positive or most negative elements. In this case, the aggregate is likely to tell you much more than one person's experience.