According to a recent study by Forrester Research, 30 percent of consumers now research their potential online purchases at Amazon, more than double the number of those who do so with Google. This is considerably reversed from two years ago, when 24 percent of consumers started their research with Google, compared to Amazon's 18 percent.

Obviously people do not just buy from Amazon after doing their research there. However, in terms of books -- if there are more people researching on Amazon because it has customer reviews on almost every book ever published, then more authors and publishers will be tempted to have their books listed there. There will also be more pressure for these authors and publishers to source positive reviews for their books in order to attract customers.

But can we trust Amazon's book reviews? It is well known that Amazon has a group of professional reviewers while allowing anybody to submit a review on almost anything. Interestingly, a comparison between these two groups of reviewers is provided in an academic paper titled "What Makes a Critic Tick? Connected Authors and the Determinants of Book reviews". Published by Harvard Business School in April 2012, the paper finds that professional reviewers tend to be less favorable to first-time authors, while consumers are quicker to identify new and unknown books.

More importantly, the paper suggests that professional reviewers and consumers alike tend to "agree in aggregate about the quality of a book". That is to say, because Amazon allows a large number of reviews to accumulate for a single product, the average opinion of a group of professional reviewers is likely to be very similar to that of a group of consumers.

Hence we can conclude that when doing research on Amazon, it is never wise to trust a single review, be it written by a professional reviewer or a consumer. Furthermore, we should always take into consideration of the subjective nature of reviews, especially in those cases where reviewers are paid or given free books to comment on. As Amazon's director of community Russell Dicker has famously admitted in an interview, "Any one review could be someone's best friend, and it's impossible to tell that in every case... We are continuing to invest in our ability to detect these problems."

Indeed, a recent study by tech entrepreneur Filip Kesler and Cornell University professor Trevor Pinch finds that "more than 80 percent of the reviews on Amazon were positive because 85 percent of prolific reviewers receive free stuff to review". Another possible reason for professional reviewers to tend to provide positive reviews, as suggested by Sue Gee, editor of book reviews for I Programmer, is that Amazon allows a reviewer to receive higher ranking if his or her reviews are found "helpful". In other words, many reviewers avoid writing negative reviews "because such reviews are more likely to be found 'unhelpful' -- something that would adversely affect their ranking" on Amazon's system.

Perhaps the best advice on whether to trust Amazon's reviews is also provided by Gee, "A negative review is on average more trustworthy than one that is superlatively positive, but you still have to be careful." With that said, these days we see an increasing amount of complaints about online book critics being too nice. Perhaps it is time we start embracing negativeness to some degree?