Jack Messenger | Feed the Monkey

Reading | Reviewing | BLEATING INTO A GUNNY SACK | Writing | Publishing

Do Readers Really Read, Even When They Buy?

If You Sell the Book, Will They Read It?

Originally published by Digital Book World

Andrew Rhomberg, September 16, 2015

At the IDPF/BE conference in May of this year, Kobo disclosed data that revealed only 60 per cent of books purchased are ever opened. And that says nothing about whether they are even finished.

Interestingly, the more expensive the book was, the more likely it was the reader would at least start it, though data wasn’t shared on whether the likelihood of the book being finished went up. It most likely didn’t, though, as completion is probably entirely dependent on the quality of the book, not the price of it.

At Jellybooks, we have been conducting reader analytics surveys of readers using iBooks, Apple’s reading app, by using modified ebooks. We measured that 50 to 55 percent of ebooks get opened, though it should be noted that these are Advance Reading Copies (ARCs) that users receive for free.

More interesting, though, is how many readers actually finish a book. Here the percentages diverge significantly: some books glue readers to the page with completion rates at 70–90 per cent – well above the norm – whereas, for other books, it might be 20–40 per cent. Readers are generally more likely to finish a plot-driven genre novel than they are a literary one.

This will not come as much of a surprise to anyone, but within each genre or category there exist significant variations that can indicate if a book is capturing the attention of readers or not. More revealing is how this data may sometimes fluctuate among different kinds of audiences: women and men, young and old, fast and slow readers, those who read during their daily commute and those who leisurely read during the weekend.

One of the most fascinating points we are starting to notice is that the decline in reader attention almost always starts early – that is, within the first 10–40 percent of the book. Strong beginnings matter: users are fickle and easily distracted by other forms of entertainment, or will give up on a book to start the next one instead. However, once a reader reaches the midpoint of a book, they are usually committed.

At Jellybooks, we collect this data weeks – even months – prior to publication, and it is starting to influence how publishers position a book: how they promote and market it. A book appealing to young commuters, for example, is marketed differently than one that captures the attention of older weekend readers.

Now some might ask, what does it matter if people don’t read the book as long as they buy it? It may sound like a shallow statement, but publishing is a business after all. Well, it does matter if buyers read the book, if word of mouth is the driving force of sales. Will a reader recommend a book to others if they gave up on it halfway through? Unlikely (though in all fairness, we are still in the process of collecting the data to prove this point empirically).

In my discovery and discoverability workshop at Digital Book World, I have been using this illustration on the virtuous word-of-mouth cycle:


The graphic is a modification of the traditional marketing funnel. It highlights one way for a book to become a blockbuster, and that is viral growth fuelled by word of mouth. If people who buy the book read it, finish it and share their views with others, then a virtuous feedback loop is created. Exponential growth will be the result, if the number they inspire to check out and buy the book is larger than the number who recommend it (some will not recommend it, but as long as they are outweighed by those who inspire two or more people, the ratio will still be greater than 1; it is averages that matter).

Now, it is important to acknowledge that word of mouth is not the only driver of what can make a bestseller. At Jellybooks, we have measured reading engagement for books that sold well, but where few people actually completed them. In such cases, it is often very powerful and influential reviewers (a kind of ‘celebrity endorsement’) that might be the driver of sales. Also, in non-fiction, people may recommend a book even if they haven’t finished it. And sometimes a book like Capital in the 21st  Century becomes such a hot topic that many people buy it even though few actually even start it.

There is also the phenomenon of the debut author who sells well, but whose second book flops. Reader analytics can show if completion rates were low despite good sales, which means the author did not build up a loyal audience who will buy his or her second book.

There are many paths to becoming a bestseller, but word of mouth is a powerful one. For that to get going, buyers reading the book is often a basic requirement, and now we can actually measure if a book is creating that level of engagement before it is published. Where a book shows great promise, we might wish to up the marketing spent on getting the viral feedback loop going on social networks and in reading clubs across the world.

In short, keep your eye on book sales, but also start paying attention to whether the book is actually getting read. Data allows for smarter publishing decisions!



  1. That is a very interesting article on whether a book really gets read. A writer does want to be read as well as to sell.

    • Thanks for your comment Laura. I agree. I want to be read rather than just sell units. As the article indicates, word of mouth is the most important element determining whether or not a book is read by more and more people. As for strong opening pages that engage a reader, I suppose it also depends on connecting with the right kind of audience for our book. So we need to know our potential readers. As readers ourselves, we know that if a book is not the kind we enjoy, a strong opening will make little difference.

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: