‘Worthless company! Communication is truly rubbish’, ‘Wish I had checked Trustpilot before purchasing’, ‘I'm never going back there again’. While many of us feel happier and more relaxed during weekends, online reviews tell a different story. A large-scale study finds that online reviews written on weekends tend to be significantly more negative than those during the week, a pattern known as the weekend effect.
The weekend effect refers to systematic differences in behaviour during the week and weekends. This is not an uncommon pattern, the phenomenon is observed in many industries, such as finance, healthcare, and psychology. A new research by Andreas Bayerl, Assistant Professor at Erasmus School of Economics, Verena Schoenmueller (ESADE), Jacob Goldenberg (Reichman University & Columbia Business School) and Florian Stahl (University of Mannheim), called “The Weekend Effect in Online Reviews”, shows that these temporal patterns are also visible in online reviewing. One might expect people to be in a lighter mood during the weekends, which results in more positive reviews. However, the opposite seems to be true, as weekend reviews consistently contain lower ratings.
Online reviews can shape costumer decisions, since many people consult online reviews before making a purchase. This study is a reminder that when certain products are prone to the weekend effect, it can give a distorted view of the true quality of a product or service.
Measurement of the weekend effect
The researchers combined and analysed the ratings of over 400 million reviews and the corresponding personal profiles from 33 different platforms, covering products, services, and employers across multiple countries. Reviews that were sent out during the week were compared to reviews that were sent out during the weekend. Furthermore, reviews on public holidays were examined to see if similar patterns appeared outside of the typical seven-day week.
The results show that reviews that were written on the weekends are on average 0.04 stars lower (out of a 1-5 star grading) than those written during the week. This is mainly driven by fewer five-star reviews and more one-to three-star reviews. Even when limiting the sample to experienced reviewers or well-reviewed products, this rating pattern holds. Public holidays tend to show the same trend, suggesting that behaviour is linked to time and context, rather than specifically the weekend days.
Temporal self-selection as the main driver
It may seem that negative reviews arise from poorer customer service, as they get more busy during the weekend. However, the researchers found that while crowdedness slightly amplifies the effect, it cannot explain the ratings entirely. Even businesses that are equally busy throughout the week receive lower ratings during the weekend.
The strongest evidence seems to point towards the idea that a different group of users tends to post reviews on the weekend than those who post during the week. The researchers call this phenomenon temporal self-selection.The authors looked at words that are used in the online reviews of weekend and week reviewers. They noticed that weekend reviewers use less words that show they are socially connected, e.g. ‘family’, ‘friends’, ‘colleagues’ etc. The perception is thus that these reviewers might be more socially isolated. And so the effect cannot be explained by people becoming more negative in the weekend, but rather that a different set of people are reviewing at that time.
Implications for businesses
This research shows that the known phenomenon ‘the weekend effect’ is also present in online reviews. This result seems mostly driven by who chooses to write reviews in the temporal self-selection. While the difference of 0.04 stars might seem small, for (specifically small) businesses this can have real implications. A low-rated review can heavily influence average review scores and future sales. Encouraging customers to leave reviews during the week could reduce the likelihood of receiving negative feedback. Consumers and producers rely on user feedback, understanding such temporal biases thus becomes crucial in understanding people’s behavioural patterns.
- Assistant professor
