Estimating aggregate consumer preferences from online product reviews


Nowadays you cannot find any product category on the internet without anyone giving their opinion on a specific product. There will be practically no products available which have not been rated by consumers through the use of product reviews. This data offers an insight in the perceived strengths and weaknesses of a certain product, also the value of the product as a whole can be obtained. With all this data available, what will be the implications for companies? In other words, how can companies use this wide variety of data to determine what to produce next or to improve their current product?

To begin with, the authors acknowledge that “the basic relevance of consumer preferences, e.g., in connection with new product development processes, is widely confirmed in marketing research and practice.”(R. Decker, M. Trusov, 2010) Therefore it will only be logical to attempt estimating the aggregate consumer preferences in order to produce a product which satisfies these preferences. Furthermore, Zhu and Zhang(2010) found out that 24% of online customers make use of online reviews prior to purchasing a product offline. This implies that the preferences can be used to determine future sales in both online and offline environments. It becomes clear that online reviews have become a major information source for consumer prior to making a purchase. With that in mind, it will be stupid to ignore the importance of online reviews in the development or improvement of products.

Online product reviews typically consist of the perceived strengths and weaknesses; an overall product rating; the formless comments and remarks (full text). One of the main strengths of online reviews compared to traditional consumer preference information; is the fact that the reviews have been written voluntarily instead of being requested. Therefore, companies can expect a high level of authenticity with these reviews. In this paper, the authors created an econometric framework which can be used to aggregate the plentitude of individual consumer opinions into aggregate consumer preference data. The suggested methodology proves to be useful in the collection of consumer preferences and also in reputation analysis.

As with every research, the implications are subject to some practical limitations. The authors assume that the reviews are written by real consumers, instead of professionals of the company itself to boost sales. Furthermore, the available reviews can be biased by self-selection. This means that consumers in a certain product category might be more willing to participate in the creation of reviews.

In conclusion, it would be fair to say that the authors make a useful contribution to the scientific body of knowledge. They provide managers with a framework which can be used to aggregate consumer preferences. This will give managers a handle on which they can build further. As the authors also state; “future research should be devoted to the development of powerful filters for detecting fake reviews and to the further automation of the time-consuming data pre-processing and attribute extraction steps.” (R. Decker, M. Trusov, 2010)

Decker, R., & Trusov, M. (2010). Estimating aggregate consumer preferences from online product reviews. International Journal of Research in Marketing27(4), 293-307.

Zhu, F., & Zhang, X. (2010). Impact of online consumer reviews on sales: The moderating role of product and consumer characteristics. Journal of Marketing, 74(2), 133−148.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s