The recent news that digital revenues have reached approximately 20% at some of the Big Six houses makes me believe that we are on the verge of a tectonic shift in the industry. It provoked me to think through the logical extension of well-established trends and comment recently that I see an 80% ebook world for straight narrative text coming in two to five years. (By “straight narrative text” I mean books of just words.)
One of my most respected sometimes colleagues privately told me the 80% number was “absurd”. But considering the logic and evidence that was offered up to refute my projection only made me more convinced that I’m right. So it’s time to expose the thought process to one of the most acute groups of critical thinkers I know: the readers of this blog.
When the Kindle came on the scene in November 2007, almost exactly four years ago, ebook sales were in the neighborhood of 1% of major publishers’ sales revenue. Since then it has risen, by my calculations, between 2 and 2.6 times per year. So we’re at 20% of revenue now in October 2011; we were a bit under 10% a year ago, around 4% this time in 2009, and about 2% at this time in 2008, when Kindle was in its infancy and its only device competition was the Sony Reader. (When you look at “annual” numbers for each of those years they’re lower, but that’s because share was gained throughout those years, fueled by new device releases; I’m talking about where things stood at about this point of the year.)
I interpret 20% of Big Six sales revenue to mean something closer to 25% of units sold, because ebooks bring in substantially less revenue per copy sold than print on major hardcover books (although they can bring in a bit more on paperbacks). But many books were not yet ebookable: most juveniles and illustrated books are not represented in that figure. So I don’t think it is a stretch to figure that ebooks are constituting 30% of the units sold for straight narrative text.
(Working backwards, that means I think ebooks were about 15% of the straight narrative text units a year ago, about 6% of the units in 2009, 2-3% of the units in 2008, and probably fewer than 1% of the units in 2007, before Kindle.)
Although the following analysis was widely misunderstood when I offered it on a prior post (the misunderstandings are evident in the comment string), the way I’m defining the measurement of this — what percentage of a straight text book’s total sale will be ebooks? — the number cannot exceed 100%. So, clearly, it is impossible for the rate of share growth which has been sustained for four years, since the introduction of the Kindle, to continue for more than about another 18 months.
In fact, at some point the switchover from print to ebooks will slow to a crawl. Sales of straight text books won’t reach 98% digital for many years, perhaps decades. What I’d expect is that we’ll reach a point of print resistance and adoption will slow down dramatically. We can argue about where that point will be. I think it is 80%. A major executive was reported to have said in Frankfurt that he thinks ebook sales will “plateau” at 40%. (Maybe he meant 40% of revenue, which, depending on how much of the house’s output was straight text, would probably be nearer to 60% of straight text units.) Everybody’s entitled to their opinion and only time will prove us right or wrong.
There are a lot of reasons to expect a continuation of the recent trend of share doubling every year, at least for a while longer. Ebook readers and tablet computers are getting cheaper and more widely distributed, by which I mean that more and more places are selling them. (One hears widespread speculation that next year we’ll see offers for devices to be free with the purchase of a number of ebooks.) The number of titles available in multiple languages continues to grow. The price of new books in digital editions is established at about half the publisher’s suggested hardback price for the hottest new releases (and also much less than most stores would sell the print book for). Everybody who hasn’t yet switched to a digital device yet knows people who have successfully and comfortably done so. More and more libraries have ebook offerings (although they can’t obtain a lot of the bestsellers at this time.)
Cheaper books, more to choose from, and more plentiful and cheaper devices would not imply any slowdown in adoption in the short term, except that those most receptive to switching have already done so. But I don’t find that a persuasive argument for an imminent slowdown; some of the late adoptees, particularly the young, just couldn’t afford the devices until the prices came down.
In fact, one thing Amazon established very early in the life of the Kindle is that the heavier book purchasers tend to move to the readers faster. It makes intuitive sense that the price of a reader is amortized more quickly by somebody who buys more books. So, in fact, we could reach 80% of the units being purchased digitally if a much smaller number, say 40% of the people who buy books, make the transition.
Among those reading this post who would fervently hope I’m wrong would be anybody with an interest in a brick bookstore, whose survival challenge is only made more difficult if the trend to ebook reading accelerates. What this says to me is bookstores would be wise to specialize in books that make great gifts and children’s books (and there is some anecdotal evidence that the stores doing well have done exactly that; the most often cited being Books and Books in Coral Gables, FL).
So except that we know the adoption rate must (at some point) slow down as we approach saturation, I find little reason to assume that it will do so anytime soon.
If the trend that has been unbroken for four years continued for another year, ebooks would constitute 40% of big publisher sales volume and 60% of units for all straight text books by a year from now. At that rate, we’d reach 80% units on straight text in the quarter after Christmas 2012.
When I say I think we’ll hit it in two to five years, I’m being consciously restrained. To get there in two years would require that consumers switch from print to digital at about 60-70 percent of the speed they have for the last four years over the next two. Were it to take five years, it would mean the conversion rate would have slowed to a crawl compared to where it has been.
So the outer edge of the prediction I stated (five years) is, to my way of thinking, unlikely because it is too slow. Predicting the current rate for 18 months is probably too aggressive, but 2-3 years is not. Having it take longer than that would surprise me and I’d love it if anybody predicting that would explain what they think will slow things down so drastically in the months to come compared to the recent past.
The colleague who thought I had taken leave of reality offered some logic. First of all, it was observed that ebook sales rose most rapidly in 2011 right after Christmas, particularly as a percentage of total sales, rather than steadily throughout the year. That didn’t surprise me. It is due to an effect I have written about previously which last year was not softened by new device releases midyear and previously had been.
Ebook readers make great Christmas gifts (better every year than the year before because there are more to choose from and they get cheaper). This has turned Christmas Day into a great sales day for ebooks but the process of new device owners “loading up” apparently continues for a couple of months after Christmas. So ebook sales in the first quarter are artificially inflated and will continue to be until we reach saturation on readers, which will probably be at least two more years. When just about everybody who reads many books already has an ereader, the post-Christmas bump will diminish markedly.
But, at the same time, the print sales reported are depressed in the first quarter. Returns come in from what have too often recently been disappointing print sales at Christmas and, at the same time, the purchase of new titles in the first quarter is dampened because some stores give up the ghost after a failed Christmas season and others are jolted into greater conservatism in their stocking by declining sales.
Since print book sales net of returns are depressed and ebook sales are stimulated by gift devices, the percentage of sales that are digital reaches a dramatic new height in the first quarter. This has happened in recent years and will happen again in 2012 and maybe in 2013.
Ebook sales in dollars were also reduced in the past two years by switches to agency pricing. Five of the Big Six went agency on April 1, 2010, when the iBookstore opened. Random House went to agency in early March, 2011. When publishers switch from the wholesale model to agency, the amount they get from each ebook they sell goes down. So even if the book continues to sell at exactly the same velocity (and it might not, since agency also raises prices to the consumer), the publisher’s revenue will decline.
These changes, which have raised the price of major publisher ebooks, have not prevented the year-on-year growth described in this piece, but the timing of the agency switches did tend to make the increases look like they are grouped in one big step increment at the beginning of the year.
To an extent they are but not as much as they look. And we’ll be taking that step again when the calendar turns over to 2012.
(The dampening impact on revenues of the switch to agency by the five publishers in April 2010 was mitigated, indeed overwhelmed, by the impact of all those iPad devices creating new purchasers for ebooks. And there were new devices that year from Nook as well.)
Another piece of evidence I was asked to consider that would apparently contravene the 80% prediction is that music sales are still split 50-50 between digital downlads and shrinkwrapped CDs. I love knocking down comparisons with the music business (I started doing it in the very early days of the blog) but this one is almost too easy.
While sales of music may still be split 50-50 between downloads and CDs, consumption is almost certainly not. People can acquire their music on CDs and still consume it through digital devices. By doing that, they get additional value in metadata (those little books that come with the CDs) and they get a copy of the music that they can readily give away as a gift.
But when somebody switches over to consume their books digitally, purchasing the hard copy version is not an option. So it isn’t helpful or indicative to look at how music sales divide; we’d have to look at how music consumption divides. And I’ll bet anybody who wants the wager that it is not 50-50! When was the last time you saw somebody playing a CD?
It was also offered up to me that Bowker polling of book consumers has found consistently this year that only 15% of the people report having bought an ebook each month. I’d say that is entirely consistent with my hunch that 30% of straight text units are digital. We’ve observed throughout the digital transition that ebook purchasers are heavy purchasers. In fact, I’d have been surprised (and felt I had some explaining to do) if the number of ebook purchasers were higher than 15% at the moment.
Until the leap this year, the switch from print consumption to ebooks was deceptively easy for a publisher to absorb without making drastic changes to its organizational structure. That time has passed. The book business we see today — how titles are acquired, developed, marketed, and distributed — is still built on the basic industry that was constructed over the past 100 years. Unless there is something wildly wrong with my logic (and I’m counting on my readers to make me see it if there is), we’ll see more fundamental change in the way straight text books are published over the next 36 months than we have over the past 36 years.
The implications of this shift require a lot more thought than I’ve been able to give it so far. But one thing I think it will mean is that trade publishing will trifurcate in the next few years. With bookstores as the primary distribution channel, it was no problem for one publisher to do straight text narrative, children’s books, and illustrated books. They shipped to the same customers in the same box. If bookstores aren’t the primary channel, these different kinds of “books” will not have a lot of commonality: in sourcing, creation, marketing, or distribution channels. I wonder how many publishers are thinking about their publishing programs with that in mind.