This page is part of a global project to create a better online reviews system. If you want to know more or give your feedback, write at [email protected] and we’ll grab a beer ;)

Reviews not only impact users' initial consideration (see “Why do we look at online reviews” and “Threshold of consideration”), but they also influence how users interpret a business’s information. When a business has a low average rating (below 4), potential customers often scrutinize all details, looking for flaws. Conversely, for well-rated products (> 4.8), they tend to focus on positive aspects. Online reviews are about trust, and negative reviews foster distrust (see “Expectations, subjectivity, standards & risks”).

This creates a double penalty. A bad score doesn’t just deter potential customers; it also makes those who do consider the business more cautious and critical.

This bias can even affect their in-situ experience with the product or service.

Take this review on Airbnb highlighting noise issues. Even if noise is only a problem for some guests, future guests might be particularly attentive to noise during their stay because of this review.

Review on an Airbnb accommodation.

As a result, potential customers almost always make choices out of caution, since most products and services have at least one bad review. This unavoidable distrust affects their decisions, even on less significant aspects.

The suggestive power is well-documented in sociology:

• In simulated classroom experiments, people judged students' abilities to be lower if they knew the students had performed poorly repeatedly, compared to knowing only a few performance measures $^1$.
• In a consumer context, people’s preferences for goods changed as predicted by Bayesian principles when they received information about how others rated the goods $^2$.

<aside> 💡 Exploration

This issue seems inherent to any review system: seeing others' opinions will likely affect the reader’s perception—that’s the purpose of reviews. We just need to make this influence more “rational.”

• Selection of the reader’s criteria. If users could filter reviews based on what matters most to them, they wouldn’t be unduly influenced by negative reviews on irrelevant aspects. The challenge: people don’t always know their criteria precisely.
• A system relying only on positive recommendations. An effective review system should highlight what users gain rather than what they risk losing by choosing one product over another, thereby reducing frustration. If a business lacks recommendations on aspects important to the purchaser, it signals a potential issue, prompting them to consider other options. </aside>

$^1$ “Judgment biases in a simulated classroom--A cognitive-environmental approach”, Fiedler, Walther, Freytag, & Plessner, 2002.

$^2$ "Social Information Is Integrated into Value and Confidence Judgments According to Its Reliability”, De Martino, Bobadilla-Suarez, Nouguchi, Sharot, & Love, 2017.

➡️ Next up: Too many reviews to read