Online product reviews have become ubiquitous on websites aimed at consumers. And while the evidence suggests that on the whole they add to sales, little research has been done at the individual level. What happens when a consumer encounters a product review? Do more reviews equal more profits? And what about the related issue of fake reviews?

Using book reviews, Prof Ying Zhao of the Department of Marketing and co-authors Yi Zhao, Sha Yang and Vishal Narayan delineate some of the factors that drive consumer responses to reviews, and help to fill some of the gaps in knowledge. Book reviews are an ideal subject because they have been shown to increase sales when the reviews are positive.

Reviews are important because they help to reduce consumer uncertainty about a book (or other product) they want to purchase. The consumer’s own experience with the book genre may also help. The authors capture both types of learning in their model, and address reviewer credibility, too.

“More credible reviews for a product are likely to have a greater effect on the consumer’s propensity to buy that product. Since it is well established that the credibility of a source is greater if he or she is more similar to the recipient, we model review credibility as the precision with which product reviews reflect the consumer’s own product evaluation,” they said.

By applying their model to 1,919 book purchases by 243 consumers at a large product review website, they found consumers learned more from product reviews than from their own experiences with a genre, and that they updated their beliefs about the credibility of product reviews over time. Not accounting for this learning process was shown to skew results.

But reviews also were subject to limits on their impacts on sales. Some companies manage word-of-mouth activity by providing incentives to consumers to produce online reviews, but the authors found this was subject to diminishing returns so that increasing the number from zero to 10 had greater impact than going from 90 to 100. And while having more reviews could lead to greater market share, this could backfire if the expense on incentives led to lower profits.

There was in fact an optimum number of product reviews that firms should spend on to maximize profits, for example, if the profit margin was 14 per cent, the optimal number of reviews was 40. Timing was also important.

“From a managerial perspective, although all product reviews have a positive effect on market share, reviews posted earlier have a greater effect than those posted later. Firms spending marketing dollars for incentivizing people to post reviews might wish to consider this dynamic effect of reviews on market share. It would be rational for firms to pay more for eliciting earlier reviews than later,” they said.

Apart from encouraging reviews, firms may be tempted to post fake reviews or allow fake reviews to be posted about their products. This could be even more problematic. Fake reviews are generally more positive than authentic ones, and the authors modeled scenarios that suggested fake reviews would undermine credibility and the purpose of online product reviews, which is to reduce consumer uncertainty.

“In the fully authentic platform, the uncertainty first decreases as a result of learning from one’s own experience and then levels off. In fake platforms, the consumer is exposed to a combination of fake and authentic reviews. We find her uncertainty increases significantly as she purchases more books. Furthermore, the effect becomes larger as the likelihood of getting a fake review increases,” they said.