Question: can data about how books are read in turn influence the way in which books are written? The Guardian (“The new platform luring readers into short fiction“), leads off by observing how short fiction has remained a relatively unpopular e-reading genre while others have taken off. Now, we see more and more self-publishing options such as MacGuffin focusing upon short fiction, poetry, public domain works, and “#fiveminutereads” to try and catch our collective e-reading attention spans. However, the part that caught my attention was MacGuffin’s focus upon reading analytics
“The subtle joke in MacGuffin’s name – tricking busy online readers into spending time with fiction through interaction and ease-of-use – might also point to its most interesting feature. Alongside every story published are its open analytics, visible to both author and readers. Mercilessly, these detail the exact number of people who have opened a story, and the number of people who actually finish it. They even display a chart of exactly when each reader stopped reading: which, while painful, does give writers the chance to test their narrative structure. Whether this will prove a digital innovation too far for more sensitive writers remains to be seen, but if MacGuffin does take off, mining this data for insights into human attention might be one of the smartest things any publisher has done in some time.”
Ok, I might have contributed towards the dropout out rate on a couple of stories while I was poking around their website. The combining of reader analytics and audio soundtracks is kind of a neat experiment (while we’re on the subject, check out this post for more on books and soundtracks).
I’m curious to see if having more data and information about readers’ habits will have an influence upon the way in which books are produced. And another thing to think about, from the authors’ and publishers’ perspective: how much does dropout rate really matter? Are there other metrics that could determine a books’ success? (This reminds me of an earlier post on the Hawking Index, too).
The Bookseller (“Jim Hinks: Reader analytics as a self-editing tool“) has a great interview with useful insights about what MacGuffin is all about. He makes some good observations about how devices might affect our reading consumption habits, now and in the future:
“Literature in audio form is certainly on the ascent. Audiobook sales are rising; 4G coverage is improving and getting cheaper for consumers. I suspect that one of the reasons publishers, at least, are so keen on audio is it seems relatively future-proof. Consumption of digital literature is largely device-led, after all. As Amazon added more functionality to Kindles, transforming them from vanilla readers into tablets, ebook sales started to level off — why read capital-L literature when you have social media, YouTube and Netflix? But whatever wearable devices come on-stream in the next few years, it’s difficult to foresee a time when we won’t want listen to stuff while keeping our eyes free to do other stuff (even in our driverless cars).”
The drop out graphs are pretty rudimentary data at this point, but it’s a pretty cool idea. The future of book reading might be full of these kinds of insights which simply didn’t exist five or ten years ago:
“During the beta test, we found that most stories and poems typically have a lot of drop-outs right at the start of the text or audio (sometimes over 50% during the first 10% of a story or poem). It seemed readers were scanning the first few lines of text, or listening to a few seconds of audio, then deciding it wasn’t for them. … It seems this simply reflects the way we browse literature, whether digitally, or picking up a book in a bookstore: scanning the first few lines, then deciding it’s not for us, and moving on to something else. The really useful drop-out stats for authors seem to come after that initial bedding-in phase, where you’ve won the reader’s confidence, only to loose it again with a misstep in plotting or pacing or tone.”