Goodreads Gives Good Recommendations

Many of us are avid readers. We’ve all gotten tips from friends on what the next most amazing, imaginative, exciting, fresh, brain-imploding new book is and we add it to our to-read list. Well, friends… that list is going to get a lot longer and a lot better.

You may remember a few months ago when we told you that Goodreads bought a book recommendation algorithm? Well, it’s here! On its blog last month, Goodreads announced Goodreads Recommendations. It works by combining “multiple proprietary algorithms which analyze 20 billion data points to better predict which books people will want to read next”. Basically, the recommendation engine looks at a user’s well-rated books and compares them to each other to gauge the commonality. Then the engine examines where they appear on other users’ bookshelves, alongside what other titles, and how often. Finally, going beyond just the numbers, Goodreads Recommendations also bases its recommendations on users’ bookshelves. Users can organize their books any way they like and with custom names: favourites, romantic, reminds-me-of-home, anything. Based on how books are organized by a user, Goodreads Recommendations will analyze what other books have been placed in the same category and make recommendations based on what kinds of titles are considered to be, for example, ‘reminds-me-of-home’.

Already, this sounds awesome. In fact, it is awesome. To demonstrate, here is a list of some books on a bookshelf I’ve named Resilience:

  • Infidel by Ayaan Hirsi Ali
  • A Long Way Gone by Ishmael Beah
  • Wild Swans: Three Daughters of China by Jung Chang
  • Room by Emma Donaghue
  • Jubilee by Margaret Walker
  • Please Stop Laughing at Me by Jodee Blanco; and
  • The Diary of a Young Girl by Anne Frank

The list has cultural variation and spans history, but most of them are memoirs and tell the story of resilience and strength. Considering most of these books were ranked on average 3.5 stars, Goodreads has made fifty recommendations. The recommended titles are culturally varied, tell of historic and current affairs, and seem to be a mix of male and female protagonists, though mostly female. The books actually look really interesting! I haven’t heard of a lot of them, which is a huge bonus for the reader because one of the main problems of recommendations with real book-addicts is that they’ve read all of them! And it’s a plus for publishers because it brings a receptive audience to new books regardless of whether the author is famous or established.

The more books you rate, the better the recommendations become because Goodreads will have more information to work with. Maybe my 26 books isn’t enough for the recommendation engine to do its best work, but so far things are looking very interesting and I’m eager to read a few titles listed. I will admit, I spent a solid chunk of my workday reading the synopses of many recommendations, then perusing the user reviews and have already added a few to my ‘to-read’ list. Though, for the purposes of work, I think this is acceptable! Plus, now I can rave about the wonders of Goodreads Recommendations to all my friends.

Finally, recommendations that actually take into account preferred reading based on time periods, mood, geography, or voice. Gone are the days of relying on recommended reads that suggest books based on your search history. No, I don’t want a travel mug; that was a gift for my sister-in-law.

Hey, Goodreads: THANK YOU!

If you’re somehow still not familiar with Goodreads, watch their presentation from Technology Forum 2011 by Patrick Brown.