On Book Ratings and Reviews

When it comes to book ratings I am not fan. One reason I have been reluctant to use a system like Goodreads is because it seems to be centered around a 5-star rating system. (Another reason is the mishmash user interface that is overly busy and confusing.)

What’s wrong with a 5-star rating system? As a consumer of such a system, I find it difficult to know what each level means? It seems to me that some books that get 5-stars are books that I know I wouldn’t like, and others that get 3-stars are book I know I’ve loved. That isn’t particularly helpful. Ah, but there’s a density to the system as well! If 50,000 people give a book 5-stars that’s got to mean something, right? Sure! It means 50,000 people might like a book–or perhaps that 100 people really enjoyed it and 49,900 ranked it highly so as to seem not out-of-step with the rest of society. Either way, it still doesn’t tell me if I will like the book. The only way for me to know is to read the book. And that is a judgement call.

Early on, when using a tool like Goodreads, I would rank books. Over time, however, I have stopped because I don’t find it helpful. In my own list, I don’t rank books. Instead, I answer two simple questions:

  1. Is this a book I would consider reading again someday?
  2. Would I recommend this book to someone asking for a book on a particularly subject?

If the answer to both these questions is “yes”, then I mark the book as one I would recommend, and move on. To me, that seems much more useful than a 5-star rating. Part of my problem with the 5-star rating is that there is no consistency of measurement. Even I don’t know what I mean when I rank one book 4-stars and another book 5-stars. What is the difference? What pushed that book up one star? It is much easier for me to say, sure, I’d absolutely read the book again, or recommend it if someone asked me if I knew of a good book on, says, boxing in the 1950s.

Book reviews are almost as useless to me as ratings. Almost, but not quite. Too many reviews I’ve read outside professional forums (i.e. review columns in newspapers and magazines) are critical of things completely unrelated to the context of the book under scrutiny. “This book is way over priced–1-star!” “When my book was delivered it was damaged.” Too many reviews seem to focus on the author and not the content. There are also those reviews that give too much away, or try to be too erudite. I wrote a review column for a magazine for a year or so, and in my reviews, I tried (a) to keep them short, (b) to focus on what I liked about the book in question, and (c) relate it to other things that happened to be on my mind when reading it. Those are the kinds of reviews I would find useful.

Still, reading, like writing, can be a lonely business, and in an effort to be more social about it, I have started to update Goodreads again. You can find what I am currently reading on my Goodreads page. You can also find what I have read in the past there, although I still consider this list to be my authoritative source. There are a few caveats:

  1. I’m not rating the books.
  2. I am writing short reviews, but I wouldn’t necessary call them reviews. When I finish reading a book, I generally scribble some comments about it in my journal. I am writing for my review a cleaned up version of what goes into my journal, which may or may not be useful as a review. You’ll have to skim through a few of them to find out.

If these caveats don’t scare you away, feel free to friend me on Goodreads.


  1. Followed you on Goodreads and I would agree – to my mind, the only rating I need is 5 stars or none.

    This opaqueness allows me to say – it was interesting enough for me to finish it, or not interesting enough and so I dropped it.

    This way, even if I stick with a book through thick and thin and reach the last page, it means I enjoyed it and would recommend it to others.


    1. @Nitin, I think it would be great if Goodreads added a simple, “Would you recommend this book to others?” field, either Yes or No. The data they collect from that could be used for similar books and it would be on a percentage basis. Also, I think recommendations are more valuable than ratings because they have a tendency to eliminate the things out of the author’s control. The book might be too expensive, but I still really like it and would recommend it, vs. giving it 1-star rating because of the cost of the book, even though I would recommend it to others.

  2. Commenting without reading the whole article…

    I picked up the tip that ratings indentify whether the author has found their target audience. Assuming they have a well-written book, then a five star indicates the target audience. Four stars are near the audience. Below that, then thhey have missed their audience. So, I read a few of the 4- and 5- stars to see if I am in the audience, and a few of the lower ones to confirm that I am not not in the target.

    Thus, I have found high rated books that I won’t read because I am not it’s intended audience.

    1. Ben, that’s interesting and I never really thought of it that way. Still, that makes it seem like the ratings are author-centric rather than reader-centric. I keep coming back to my notion that art is so subjective that it defies ratings. I’d much prefer a system that showed a pie chart of how many people recommend the book vs. how many people don’t. If I got a defective book but ultimately, I liked the book enough to recommend it, that eliminates the silly reviews and ratings that are completely outside the author’s control.

      Of course, thanks to you, I am now picturing the rating system as a bullseye, a series of concentric rings, where the rings in the center represent the targeted audiences…

  3. There’s even a certain amount of discipline missing from reviewers too. Reviews of cost and quality of paper should go to Amazon/where they bought it from. Reviews of the content should go to Goodreads. Reviews of the author/their opinion of them should go into the void.

    I agree that Goodreads should really have simpler controls. There’s a lack of progress there that really irks me!


This site uses Akismet to reduce spam. Learn how your comment data is processed.