Monday, August 18, 2014

The Amazon Slant

As many have said, we live in a very divisive culture. I'm not sure if it's always been this way, but it's certainly that way right now. Left or Right. Black or White. Liberal or Conservative. Right or Wrong. On side or the other with no room for anyone to be in between. No room for ambiguity. No room for indecision. We're just not comfortable with it.

I'm sure that at least some of this attitude has been adopted from the press, since controversy sells. People getting along is not a story; that doesn't happen till they take up sides and start throwing bricks at each other. If you want to sell things, it's a good tactic.

And Amazon knows this.

I don't know how many of you pay attention to the rating systems on the various sites on which you may be rating things, but they are not all the same. In fact, most of them go something like this:
5 -- I loved it!
4 -- I really liked it.
3 -- I liked it.
2 -- I didn't like it.
1 -- I hated it!
Goodreads has a more positive slant on it:
2 -- It was okay.
1 -- I didn't like it.
There's no room for hate there.
Basically, though, most sites offer "like" as the default giving you much more room to rate things positively than negatively. If you're paying attention. What this means is that most sites have a "top heavy" rating system that's geared toward generating positive ratings and reviews.

Looking at Goodreads more closely, what we have is a system that is designed to get ratings of 3, 4, and 5. 2s should be virtually non-existent, leaving 1s as the only real option for an actual negative review or rating.

Why do I say 2s should be non-existent? Because most people most of the time do not have an actual "it was okay" reaction to things. They like things or they don't like things. There are very few "I could take it or leave it"s.

Which is what makes Amazon's rating system so interesting. It has that "it was okay" right in the middle.
5 -- I love it
4 -- I like it
3 -- It's okay
2 -- I don't like it
1 -- I hate it
So, when you look at Amazon reviews, you get high numbers on both ends and almost non-existent numbers in the middle, because the system is designed that way. It wants to pit the 1s against the 5s, because that's what draws attention to products.

As someone who does a fair amount of reviews on Amazon, I have seen a lot of this confrontation first hand. For instance, there is a strong group of Marvel-haters out there. So, if a post a review for a Marvel movie, it is sure to immediately get "unhelpful" votes (my review for Guardians of the Galaxy didn't receive the normal deluge of negative votes when I posted, although, still, the first vote was negative).

What it boils down to is that Amazon doesn't want ambiguous ratings or reviews. Amazon wants "I loved this!" or "I hated this!" When you can shrug your shoulders and say, "Well, it was okay, I guess," no one is going to want to take a look at the product. Whatever the product. I love or a hate, though, will get some attention.

And what's my point with all of this?
The first thing is this: When you're rating things, make sure you pay attention to the rating system you're using. Just to give you an example, I was rating/reviewing some things recently, and I was on Goodreads leaving my stuff there. The particular story I was dealing with got a 3 on Goodreads because "I liked it." When I switched over to Amazon, I was sort of on autopilot, and I gave the story a 3 there, too, which was not accurate. On Amazon, I needed to leave a 4. It was a couple of days before I noticed what I'd done and went back and fixed it. It's just something to be aware of.

The other thing is... well, I'm not sure. I mean, I am sure, but I'm not sure (I'd give that a 3). If you're looking at selling things (like books), it seems that a way to do that is to generate some love/hate around it. That's what Amazon seems to think at any rate. And I've seen that work in actual practice. I don't know, exactly, how I would say to go about doing this, but there probably are ways. For one thing, though, as authors, it may not be in our best interests to be getting all upset about reviews on the negative end of the spectrum. I mean, it's never in our best interest to act out over negative reviews, but it might be even more than that. I think the real key is to learn how to exploit the reviews on either end of the rating scale and make them work in our favor. I'm just not sure how, yet.


  1. I didn't realize the two review systems were different. If I liked a book, I'd certainly give it more than a two. A two is a low rating to me. It's like a D - who wants a D?

  2. Well, I don't want to see any "I hate" this book when mine is finally published. haha

    What I don't like is there are people out there willing to write up a review on something they never even laid their hands on. Those fraudulent reviews need to be removed. I try to only pay attention to the reviews that are verified purchases. Okay, rant over.

  3. I never think of reviewing a book anyway. I have occasionally been asked to post a review and do so, but generally don't bother.

  4. I always read a few one-star reviews before buying a book. So often they are complaints about something completely out of the author's control, like shipping taking too long. I find the whole rating system skewed and corrupted on both ends of the spectrum, really.

  5. In general, I just find Goodreads reviews to be more honest. People are less worried about offending, and friends/family members are less likely to leave a slew of BS reviews on Goodreads. Meanwhile, pretty much any book on Amazon has a plentiful amount of 5 star reviews, making it really hard to differentiate between bad books and good books.

    We have a friend who shall remain nameless who released a book. Honestly, it's not good. It needs a lot of work. But it has 19 5 star reviews, all from her friends, and one single 3 star review from a stranger. It's the most honest review on there. But it's buried in negative votes because she had all of her friends downvote it. And yet, it's probably the one review you WANT to read as someone looking to buy it.

    All I'm saying is, awful book on Amazon - 4.9 stars out of 5 from 20 user reviews. That is hardly an accurate representation of the quality of that book.

  6. I was reviewing a NYT bestseller on Goodreads yesterday, a book I did not like at all. Started to leave 2 stars then thought, why not be truthful and give it 1 because I truly did hate that book!

  7. I have "it's okay" reactions to things. Maybe it's more like "I'm conflicted", because I can see good points in it, but also enough bad ones that I don't truly like it.

    Reviews on big websites can be a lot less trustworthy than independent reviews (like on someone's blog). Even on Goodreads I've seen books that I think get way more love than they deserve. Maybe it's something about social pressure.

  8. Alex: Yeah, I get that, which is why the particular systems need to be looked at. I think places like Goodreads take advantage of our sort of general perception about how rating works.

    Elsie: That's why I don't review things I don't finish.
    And verified purchases don't really matter, either. People are just as capable of reviewing those without actually putting in the work.

    Jo: Reviews can be tough.

    Rajiv: Thanks :)

    L.G.: Yeah, which is why it's important to actually look at reviews rather than just looking at the ratings.

    ABftS: I've dealt with some of those people myself. "I went to high school with this guy. He wrote this book, and you should read it! -- 5stars!" Seriously? You expect me to believe you read it after saying that?

    Karen: And a 2 on GR would have meant you thought it was okay. A 1 doesn't even do justice to your response to the book if you really do hate it.

    Jeanne: I don't think ambivalent feelings equate to "it's okay," not necessarily. If you don't truly like it, it's not really "okay," because "okay" implies no strong like or dislike.

  9. It takes a lot for me to "hate" something. I try to recognize the value from a more objective viewpoint and the time and effort someone has put into the creation of a work. I may not particularly like something, but it's rare that I will actually "hate" it and if I use that designation then there is probably some moral or philosophical reason for my so called hate.

    I guess the strong rating categories make for a more interesting discussion. I love reading the one star reviews because they are often incredibly stupid and funny. Often they have nothing really to do with the work being review itself, but instead some extraneous think that shows the reviewer's ignorance.

    Tossing It Out

  10. I'm not so sure I agree with you about "It's okay." But before I work out that, let me say that these gradations are, like temperatures and hundredths of a second, hard if not impossible for humans to sort out, and also somewhat meaningless without some concrete reasons why.

    Take "It's OK." When I first read this the very first books that popped into my mind were "The Sparrow" and "Children Of God," probably because I was talking about them yesterday with a coworker. "Children Of God" to me has an "It was okay" feel, but ONLY because "The Sparrow" was so stellar that "Children" couldn't have really lived up to it. On its own, "Children" is a four -- meaning it was great, not worrying about Amazon's or Goodread's characterizations -- but in comparison with "Sparrow" it feels like a 3. That's the kind of information you could put in a review and rate it a 3 and be honest, or rate it a 4 and be honest, and it's helpful information.

    That's one reason I dislike stars and Rotten Tomatoes and other numbers: it conveys NO information, really. "Andrew Leon says this book gets 4 stars" tells me nothing, and less useful is "Critics give this a 6% on Rotten Tomatoes." And the way Amazon and Goodreads characterize their stars is I think not at all related to how people actually view them. I think most people use them as a simple quantitative measure: the closer to 5 stars, the more they liked it, with 1 being "didn't like it at all" or "hated it." That's a distinction that's subtly different than how they actually use them, as people aren't weighing the books necessarily on their own merit but in relationship to everything else in the world.

  11. ... part 2, as it was too long for one form:

    Now: the gradations. Can most people really distinguish between "I liked it" and "It was okay"? I know there's a difference; if Sweetie cooked a dinner and I said "I liked it" she'd take it as a compliment and whereas "Ok" she'd probably take as an insult, but there's nothing inherently contradictory about those two statements. OK is just a slightly lesser liking. And just like people can't tell you if it's 75 degrees or 70 degrees, it's a distinction that because it's so hard to make is relatively meaningless but takes on incredible importance because of the arbitrary divisions. We place a huge amount of significance on tiny divisions that are meaningless. Books are $9.99 because $10 is too much, that one cent leap meaning a lot. If the high yesterday was 79 that would mean something different than if the high was 80 -- people would say "It hit 80 degrees yesterday!" but the difference between 79 and 80 degrees is infinitesimal.

    And I submit that the difference between 5 and 4 and 3 stars is like that and hence even less helpful. So Amazon's gradations are meaningless on 2 levels: most people don't know what they mean and hence don't interpret them that way, and the difference between 2 and 3 stars isn't really that great, but we will ascribe all sort of meaning too it.

    As for whether negative reviews can help, they possibly can if they generate enough controversy. I was relatively uninterested in "The Goldfinch" until I started reading the backlash against it, after which I went and read about it and read some reviews and then added it to my list.

    I wasn't aware that Anne Rice was writing at all until I heard she'd been attacking her critics on Amazon.

    Tony LaPlume and PT get into pitched battles about negative reviews.

    My own response to negative reviews was to ignore them or try to be polite and comment on them, but I'm thinking that perhaps a total flame war might help -- especially if you've got a few positives, because if it generates some interest people will read about the fight and then a few might go read about the book. "No publicity is bad publicity."

    So there's a thought: should authors openly attack and fight with bad reviewers? We should do a round robin on that on IWM.

  12. I know Pat unleashes his bulldog every time he gets a bad review. I want to reiterate for the record that I didn't pitch a battle over a negative review, but a negative as well as dismissive one, and from someone who by all accounts is generally generously supportive of me. But enough of that.

    Last year I had Goodreads reviews that covered the whole star spectrum. "It was okay" was an acceptable response to some of what I read. It still looks bad, and at a star level only certainly is, but I generally try to find the positive, which is what I especially tried to do with my three-stars (word to the wise: authors still don't like that).

    I find one-star reviews to be hilarious, especially when accompanied by detailed if brief explanations. Most often these boil down to experiences that shouldn't have happened in the first place, and these reviewers know it, as mentioned by others here.

    I find five-star reviews most suspect, as they often accompany explanations that aren't even very critical but rather fostered by the reviewer's relationship to the author. They generally lack any real perspective, and are simply there to prove that the relationship exists. Therefore I dismiss them out of hand.

    This is to say I don't base the quality of the review on how many stars are included, but the critical analysis that accompanies it.

  13. Lee: I really -hate- reviews that aren't about the work. "This gets 1-star because it arrived cracked." Seriously? How is that about the product?

    Briane: I agree with you about one individual rating; it has no real meaning, especially without a review (which is why I say reviews are so important). But on something like Rotten Tomatoes, I think the rating does tell you something. I mean, 6% means a lot of people really didn't like it. On Amazon, it may not be so clear cut, because a rating of 1.5 stars may be based on just a handful of ratings. So there is a thing with ratings where quantity actually matters. Not that quantity can't be faked.

    I'm up for that discussion on IWM.

    Tony: Right, it's the quality of the review, not the rating, that really matters. Unfortunately, most people don't bother with more than "I loved this!" or "What a waste of time!" If you didn't like it, I want to know why you didn't, and, if you did like it, I want even more to know that.

  14. This was a very informative article.
    I know someone who sells antiques on Amazon. She prefers Amazon over Ebay. She has told me that there can be conflict when you step on someones competitive toes there. I've read reviews there and it reads like a giant clique there.

  15. None of this every really occurred to me before, but it does make sense.... granted, I don't know how accurate the lack of 2's on GR is... I see a lot of them, with very negative reviews attached to the rating.... but yeah, an interesting post.

    It also never crossed my mind to leave reviews of movies on Amazon. Huh.

  16. Alex H: The negative reviews are there because people don't pay attention to what the rating means. That thing in particular is an issue because of the way goodreads handles ratings and their whole thing about how many people liked a book.

  17. Interesting! I didn't realize the ratings systems between the two were different. And when choosing to buy a book or not (especially if I don't know the author), I always take a look at their negative reviews just to see what people didn't like. Great post!

  18. Sarah: Most people don't know that any ratings systems are different, because we already have it in our heads what "the system" is. I think, sometimes, these places bank on that to skew the results.