The sessions I scheduled for myself at Tools of Change focused on libraries, readers, and data, particularly data about readers. I still get the sense that many of the biggest publishers are still shifting in their recognition of the reader (all of us contentious folks who don't agree with each other but want to be heard) as their main customer base. So I wanted to hear the data from Bowker about readership and how folks are reading, and I was determined to go to the session from Goodreads CEO and Founder Otis Chandler.
Unfortunately, there were sessions that were misleadingly titled, which meant that in my perspective they pretended to be about something they were clearly not. The two panels I attended Tuesday morning were titled in a manner that made me think I was going to receive data about the topics explored, but instead heard product promotion, to which I say, “Feh.”
For example, What Should I Read? A Brief History of Recommendations, started out with a cursory summary of recommendations from bookstores, and moved on to Amazon, Pandora, and Netflix algorithms, and the difference between searching and browsing to discover new material to read. Then, about 20 minutes in, it turned into a promotional session about how Zite collates user clicks to recommend related content.
To be honest and fair, Zite is very cool. When Angela James, who was sitting next to me, fired up her iPad to sign up, adding her Twitter feed and her Google Reader created sections on the NFL, chocolate, dessert, shopping, publishing, and shoes. Pretty accurate given the topics she reads – but I didn't need half a session with a cursory exploration of discoverability and then a promotion of how great Zite is. I wanted more of the history and possibility of recommendations, not how one company was exploring recommendations for their product.
That said, I have put Zite on my iPad and am fascinated by it. So while I didn't want to be the recipient of promotion, clearly that promotion worked!
I would request from Tools of Change, please, stronger marking of sessions that are going to be product promotion. Otherwise, I have to assume one or two people presenting from the same company will likely be promoting themselves and I'm going to start avoiding those sessions. Perhaps I also should have been more aware of the likelihood of promotion before, but I consider myself more educated now.
Now, on to the stuff that blew my mind some more.
Consumer Attitudes Toward E-Book Reading was a presentation of data from the Bowker and Book Industry Study Group survey of, you guessed it, reader attitudes toward reading, specifically print (referred to as P) and digital (E) reading.
Highlights: Readers are switching from P to E but the number of people migrating has leveled off a bit, while those with e-readers still buy more books.
There may be a seasonal shift in e book purchase. e.g. Person receives digital reader as a gift at the end of the year, they buy a boatload of books in January, then their purchases level off. The data for the past year or so seems to support that hypothesis.
The other factor in digital reading shift is the use of tablets, and how many people are using them for Things Other Than Reading. While this strayed into Anecdata (™ Brett Sandusky), and wasn't part of the fielding, the presentation suggested that those with tablets are more likely to consume other media or play games – Kelly Gallagher said that several years ago, if you would have told him he'd be spending a lot of time playing Scrabble with total strangers, he'd have laughed at you. Now, several people in the room admitted to having 2 or more Words with Friends games going at once.
During the section about buying habits and readers switching from purchasing print to ebooks, E_Bookpushers asked via Twitter if there was data about which of those readers stopped buying print simply because they no longer had a local bookstore. I asked that question of the presenters on her behalf, and they said that they had not asked whether absence of a local bookstore was a factor in any purchase of digital books. Future surveys may include geographic data and that would make it perhaps easier to pinpoint bookstore locations as well.
Then came the Goodreads session with Otis Chandler, How Consumers Discover Books Online, which was my favorite – gee, I wonder why. All that data about readers! Eventually the slides may become available online, in which case they are worth looking at just for the graphical breakdown of reader/book interaction. An interesting point from the end of the session: Chandler revealed that he has always loved reading, and when he began to build different social networks professionally, he realized that there was an opportunity to combine social networks and reading, and hence, GoodReads. When he was in college, he was asked to design and build something out of foam to present to the class. He built an ereader – and this was in 1996. So it's cool to listen to someone who is personally enthusiastic about reading talking about reader data.
Chandler's presentation explained how Goodreads users register and add books to their shelves, and that he was looking most specifically at the 'to-read' shelf to get an understanding of reader intention. Looking at what readers want to read revealed an assortment of influences, and allowed me to see in graphical form the numbers and spikes in how readers discover and interact with books.
Highlights in BULLET form ahoy!
Goodreads has a community of 7 million registered readers, and is largest reader site in the world.
Some 250 million books are shelved at Goodreads
Over half the books added to user shelves were added in 2011. The To-Read shelf has 63 million books – as many as the Library of Congress, NYPL and Boston libraries combined.
In January 2012, over 5 million books were added to the to-read shelf. (Sarah notes: come talk to this guy when someone grouses about how no one reads books any more. Good gravy!)
The top 40 most popular books (the titles everyone has heard of, basically) represent only 5% of the to-read shelf. Everything else is books that aren't super popular.
- People add books to their shelf during registration (19% of books added), and after a search (19%). Searches are folks looking for a specific title. Search is the stronger mechanism for adding books to the to-read shelf. Search also includes the Goodreads recommendation engine, which launched in September 2011. There are over 20 billion data points involved in recommending books for readers.
There was a 60% increase in shelving after the recommendation engine launched, and the recommendations engine is meant to hit the “mid-list sweet spot.” Angela James asked during the Q&A what the minimum threshold of ratings a book needed to have in order to be added to the recommendation engine. Chandler didn't know the exact figure, but guessed it was maybe about 100 ratings.
There is a minimum level of user star ratings needed to be included in the recommendation engine.
A smaller percentage of to-read shelving comes from a friend's update (9%). The average Goodreads user has 19 friends. Readers also discover books from user-created book lists (7%). Readers also add books to their to-read shelf using the goodreads app (6%). There have been over 800,000 installations of the app.
Giveaways are also a source of reader discovery. 2% of to-read shelving occurs during a giveaway. In January 2012 they ran 1065 giveaways, and shipped over 8000 copies of books.
- Goodreads also did a survey of readers, asking how they found books. They had over 3000 responses. 96% of the respondents said they found books because they knew of the author already. Chandler pointed out that this data indicated the importance of marketing an author to that author's known audience.
79% indicated they discover books from friends who are offline, and 64% from Goodreads friends.But surprise: only 14% said they found a book via Facebook, and only 6% said they found a book via Twitter.
- Breaking the data by genre revealed that Science Fiction, Fantasy and YA readers discover books most commonly by looking for known authors, using Goodreads lists, or browsing those genres on the site. Romance readers use Goodreads recommendation engine quite a bit and are more likely to discover books from reader lists. Chandler supposed this was because romance readers may not talk to other people offline about reading romance.
When Chandler began looking at the effect of media exposure on books, the presentation got really interesting. This is the one picture I took of his slides: check the higher spike of NPR exposure vs. New York Times exposure.
Seeing friends add a book also means more activity for that book after that initial media spike. In other words, there's a longer-term effect of media exposure for books when they are located and tracked on a social network like Goodreads.
My theory as to why NPR has a greater spike in reader attention than the New York Times (and Candy and I saw this happen with the Bosoms after we were featured on NPR's All Things Considered Weekend edition): NPR listeners may think of themselves as part of a conversation, even though they may not call in to talk to the reporter they're listening to. So they're likely to find and read the books being discussed (especially since NPR has several book programs, bless them) (Hey NPR, More Romance Please!) because they want to be knowledgeable regarding any continued conversation or programming about that book. The New York Times readers are already reading, and there's slightly less of an opportunity or expectation for interaction. I'm curious about the overlap of the NPR and the NYT audiences, though. (This is totally all my own theory and conjecture in this paragraph – not part of the presentation at all.)
Anyway, back to the presentation: the biggest point made in the presentation is that there is no one effective strategy, no “magic bullet” (silver, of course!) to create discovery for readers. It happens in a number of different ways, and Goodreads can amplify the effect of any attention a book receives because of the social connections between readers.
Other sessions that received positive comments in the ToC Twitter feed were the session featuring Small Demons' Valla Vakili, who spoke on Exaggerations and Perversions as he demoed Small Demons.
While Tools of Change in Publishing is not a conference about readers, it is about the changes happening in reading material, both in terms of corporate workflow and of the technology being used to produce, market, and actually read books. I loved the sessions that examined reader habits and interactions, but that's particularly appealing to me for a whole mess of reasons. It's a rare thing to be able to accurately determine and guide how users will apply technology to their lives, especially with a fundamental activity like reading, which uses a universal posture but is performed mentally and physically in a multitude of ways. While I didn't leave this year's ToC as excited and brainful as I have in the past, I did leave with a great deal to think about, and lots of data to ponder.
ETA: Goodreads put most of the slides online on their blog – have a look!