Hey guys,

Music to write by: “Fall to Pieces“ by Velvet Revolver

Okay, I’ll just come out and say it right now: The reviewing system on Amazon.com is broken. So there, I said it, and I stand by it. Here’s why:

Midwest Book Review posts a lot of reviews on Amazon. Thiry-two thousand of them at last count. Yes, I said thirty-two THOUSAND. They must have quite a range of reviews. Except that they don’t. Every single review they’ve ever done gets five stars. Even self published authors universally get five stars reviews. I have nothing against the self-published, but can they ALL have written a work equal to Vinge’s? I think not.

One argument I’ve heard is that they only post positive reviews, to which I ask, why? Readers rely on reviewers to tell them not only when a book is good, but also when it’s bad. By posting only good reviews Midwest is homogenizing the reading landscape. If a reviewer claims all books are equally good, then their reviews are useless. As for the initial assertion, I don’t believe they publish only good reviews. If they’ve given 32000 5 star reviews (the top tier), can you imagine how many reviews they must do overall? How many books are perfect, 1%? So that means they do 3.2 million reviews and toss out all but 32,000? Sure…

Out of the reviews they DO do, many of them show no actual knowledge of the book itself, other than what can be read in the Editorial Book Description (written by the author). This is often a real good sign that the review is fake. In my opinion, if these guys think they’re doing some good by spamming 5 star reviews, they’re daydreaming.

We’re also not fooled by authors that get their friends and families to log on a give them five star reviews. Even a “Real Name” moniker doesn’t help in detecting them. A good rule of thumb is to never trust a reviewer that’s written only one review.

Especially bad are anonymous reviews that are clearly written by the author themselves (like, for instance, if you review your book using your Writing.com handle. You know who you are, and now so do we).

But the very worst kind of review was the subject of a recent post on Amy Casil Sterling’s blog. Sterling’s book was given a poor review by George Randolph Calverhall. Like the fake-looking reviews written by Midwest Book review, Calverhall’s reviews were all spurious, containing very little information that wasn’t available in the Editorial Book Description. Upon further investigation, Sterling discovered that Calverhall was targetting books by minorities for bad reviews. Personally, I was stunned at the pettiness of this tactic, but, looking at the review, it sure seems like Calverhall didn’t read the book. I could have written that review (Sorry Amy, IMAGO’s on the reading list).

What can we do about this? Amazon needs to post a few of each reviewers stats next to their names, not just their location, or if they are using a real name. They should include the number of reviews the reviewer has posted as well as, and this is key, the average rating of their reviews. Amazon could then use this information to “sort” reviews according to relevance. Reviewers like Calverhall (too many negative reviews) and Midwest Book Review (too many positives) would migrate to the bottom of the list . Even if Amazon doesn’t perform this sort, the reader could, themselves, decide which reviews are relevant.

Readers deserve to know when they’re being hoodwinked and minority authors deserve to have their work judged on merit, not the colour of their skin (Or in Ms Sterling’s case: her sex). The system needs to be fixed. Come’on Amazon, put a fire under it.