The Shatzkin Files

What Oyster going down demonstrates is not mostly about the viability of ebook subscriptions

Tweet about this on TwitterShare on LinkedInShare on FacebookShare on Google+Share on RedditShare on StumbleUponEmail this to someone

The news that the general ebook subscription offering Oyster is throwing in the towel was not really a surprise. The business model they were forced to adopt for the biggest publishers — paying full price for each use of a book with a threshold trigger at considerably less than a complete read while, at the same time, offering consumers a monthly subscription price that barely covered the sale of one book, let alone two — was inevitably unprofitable. Their only hope was that they’d build a large enough audience fast enough that publishers would become in some way dependent on it (if not the revenue it produced) and agree to different terms.

It would be a mistake to interpret Oyster’s demise as clear evidence that “subscriptions for ebooks don’t work”. Obviously, they can. Safari has been a successful and profitable business for nearly two decades. The Spain-based 24Symbols has been operating an ebook subscription business, mostly outside the US and mostly not in English, for too many years to be running exclusively on spec VC money. Scribd has very publicly (and a bit clumsily, in my opinion) adjusted their subscription business model to accommodate what were unprofitable segments in romance ebooks and audiobooks, but the inference would be that for other segments the business model is working just fine. And then there’s Amazon’s Kindle Unlimited, which is sui generis because they control so many of the parts, including deciding more or less unilaterally how much they’ll pay for much of the content.

What seemed obvious to many of us from the beginning, though, was that a stand-alone subscription offer for general trade books could not possibly work in the current commercial environment. The Big Five publishers control the lion’s share of the commercial books that any general service would need. All of those publishers operate on “agency” terms, which makes it extremely difficult, if not impossible, for a subscription service to pull those books in unless the publisher allows it. The terms that the publishers would participate in the subscriptions required, which were, apparently, full payment for the book after a token amount was “read” by a subscriber, combined with a limited number of titles offered (no frontlist), made the subscription offer inherently unprofitable.

The publishers see the general subscription offers as risky business for books that are currently selling well a la carte. Not only would they threaten those sales, they threaten to convert readers from a la carte buying to going through the subscription service. To publishers, this just looked like another potential Amazon: an intermediary that would control reader eyeballs and have increasing clout to rewrite the terms of sale.

So they only participated in a limited way. Penguin Random House (the biggest, and in shouting distance of half of the most commercial books all by themselves) and Hachette Book Group did not even experiment with the non-Amazon subscriptions. HarperCollins and Simon & Schuster, and to a lesser extent Macmillan, participate in a limited way. Multiple motivations drove the participation that did take place. The primary goad, probably, was to simply oppose Amazon. Having customers nested anyplace except the behemoth in Seattle can look like a good idea to most publishers. But another was to collect at least some of that VC money poured into an unlikely-to-work business model before it was exhausted. And because the publishers got to decide which books to include, they could choose backlist titles that weren’t generating much revenue anyway and which might benefit from “discovery” within the subscription service.

(Carolyn Reidy, the CEO at Simon & Schuster, tipped to this in her talk last week at the BISG Annual Meeting where she specifically mentioned the value of the discovery S&S has seen take place in the subscription platforms.)

But not all the subscription services were equal. The established Safari was in a market niche, serving mostly B2B customers in technology companies. (They have recently gone to an expanded offering because Boeing and Microsoft techies don’t just need books about programming; they’re also parents and cooks and gardeners so general-interest non-fiction can appeal to them. But that’s not the foundation of Safari’s business and they’re not trying to push fiction.) Scribd had a foundation business as a sort-of “YouTube for documents” that the ebook subscription business both built on and enhanced. For Amazon, Kindle Unlimited just gave them another way to transact with the ebook customer and it gave them another outlet for their exclusive Kindle content.

Only Oyster and another pretty-much simultaneous startup, Entitle (which had a proposition more like a book club than a straight subscription service), were trying to make the alternative ebook revenue stream into a stand-alone business. Entitle went down before Oyster. Librify, another variation on the theme, was acquired by Scribd.

So the failure of Oyster is actually another demonstration of a “new” reality about book publishing, except it is not so new. Book publishing — and book retailing — are no longer stand-alone businesses. Publishing and bookselling are functions, and they can be quite complementary to other businesses. And as adjuncts to other businesses, they don’t actually have to be profitable to be valuable. What that means is that entities trying to make them profitable — or, worse, requiring them to be profitable to survive — are at a stark competitive disadvantage.

Amazon is the past master at making this reality obvious. Remember that they started as a “book retailer” and nothing else. They leaned on Ingram’s Oregon warehouse to enable their business model, which was to take an order for a book and accept payment, then procure the book from Ingram and send it to the customer, and then a little later pay Ingram’s bill. This positive cash-flow model was so brilliant that Ingram could have readily enabled lots of copycats, and they formed a division called Ingram Internet Support Services to do just that. So Amazon killed that idea by cutting their prices to no-margin levels and discouraged anybody else from getting into the game. That was in the late 1990s.

They could do that because the financial community had already accepted Amazon’s strategy of using books to build a customer base and to measure future business prospects by LCV — the “lifetime customer value” of the people they did business with. And it became clear pretty rapidly that they could sell book readers other things so no- or low-margin sales were simply customer acquisition tactics. This was a game Barnes & Noble and Borders couldn’t play.

Now book and ebook sales are almost certainly no more than a single-digit percentage of Amazon’s total revenue. Kindle Unlimited, like their publishing enterprises and self-publishing offerings, are small parts of a powerful organization that has many ways to win with every customer they recruit.

Scribd is not as powerful as Amazon, but they began with a network of content creators and content consumers. That gave them a marketing advantage over Oyster — not every customer had to be acquired at high cost since many potential customers were already “in the tent”. But it also gave them some stability. Eyebrows were raised recently when Scribd put the brakes on the lending of romance books and audiobooks. But tweaking the business model for those verticals simultaneously leaves open that the model is actually working in other niches.

We can see this playing out in a much more limited way in Barnes & Noble stores, where books are being replaced on shelves by toys and games. But that’s not likely to be enough diversification to matter in the long run. It is certainly not going to get B&N where Amazon is, where far more than nine out of every ten dollars comes from something other than books. And Barnes & Noble is nowhere near a point Amazon has reached: where the profit from book sales is incidental if they keep bringing in new customers and also keeps them loyal.

The story on Oyster, still incomplete as of now, is that a lot of their management team is on its way to Google, which, in effect, “bought” the company to get them. Google seems to be trying hard to make sure we don’t think they bought Oyster’s business, they just bought Oyster’s staff. Obviously, Google fits the description of a company with many other interests in which books can play a part. In the beginning, that was all about search. Now it is also about the Android ecosystem and media sales in general. An ebook subscription business, or even a content subscription business, could make sense in Google’s world. But it would be a relatively small play for them. My hunch, and it is only a hunch, is that they have something other than a mere “book subscription service” in mind for that Oyster staff to work on. Smarter observers than I seem to believe that the personnel Google recruited give them knowledge about Oyster’s mobile reading and discovery technology. Of course, that’s core information for Google.

Similarly, Apple, which now has subscription service for music, might also consider doing one for books — or for all media — at iOS at some point. They don’t have one of Amazon’s advantages — a big stable of intellectual property they control — but they are all about creating an ecosystem that people stay in and don’t leave. Book subscriptions could enhance that.

But the central point I’d take away from this is not that subscription failed, but that a pure book business play failed. One obvious question that provokes is when we will see some signs of synergy between Kobo and their owners at Rakuten, who presumably have Amazon-type ambitions but haven’t seemed to use their ebook business to help pursue them.

And what is true of book retail is also true of book publishing, as we observed in this space quite some time ago. Both publishing and book retailing will increasingly become complements to larger enterprises and decreasingly be stand-alone activities that business can dedicate themselves to for profit.

The New York Times this morning has a front-page article essentially reporting that the ebook surge is over, at least for now, and the print business appears stable. This is great news for publishers if the trend is real. Unfortunately, there were a few important points either elided or ignored that might have undercut the narrative.

One is that, while publishers report ebook sales as a percentage of total book sales steady or slightly declining, Amazon says (and Russell Grandinetti was quoted in the article) their ebook sales are going up. Assuming all this is true, is the difference perhaps sales migrating away from publishers (which sales would be reported by the AAP stats they rely on) and moving to cheaper indie titles available only through Amazon (which sales would not)?

Another is that publishers are raising prices on ebooks and making the price rises stick because of Agency. Is all the sales resistance created by higher prices resulting in print sales, or is some of it causing the book to be rejected for something cheaper? In other words, might total sales for many titles be less than publishers would have looked for before? (At least one agent tells me this is the case.)

And another is that the indie bookstore resurgence has occurred in the years following Borders’s demise and the shifting of the product mix in Barnes & Noble. It is worth asking whether the indies are temporary beneficiaries of a sudden shelf space deficiency or whether we’re really seeing not only an increase in print reading, but a renewed interest by book readers to go to stores to buy the print. That question isn’t posed in this piece.

Tweet about this on TwitterShare on LinkedInShare on FacebookShare on Google+Share on RedditShare on StumbleUponEmail this to someone

  Back to blog

  • Anthony Pero

    My thoughts exactly on theTimes article. But the ‘bubble’ for indie bookstores might last 20 years. Totally worth pursuing as a small business owner

    • We agree, although I’d feel more comfortable agreeing with 10 years than 20!

      • Anthony Pero

        Well, what if I said my 20 years started in 2011, would you raise to 15? 😉

      • As I know you know, there’s no way to be that precise. 10 or 15 are the same thing. 15 or 20 are the same thing. But 10 and 20 seem far enough apart NOT to be the same thing. Futurist thinking requires accepting something other than precise math.

  • xist publishing

    You’re right to point out the importance of niches in subscription services. As a digital-first children’s publisher, these companies have been instrumental in our growth, partly because parents feel comfortable paying around $5 a month for a walled garden of content, and partly because these apps mimic existing customer behavior around picture books. Kids go through a lot more books than their parents buy, and they go through them quickly, so the Spotify-type approach works much better than for trade content.

    I wonder if there are other verticals where the subscription services can improve existing experiences rather than create new models.Or if it will take, as you suggest, non pure book business plays, to see these types of revenue streams succeed.

    –Calee Lee, Xist Publishing

    • Curated content is a great use case for subscriptions and managing what kids can see is one of the best reasons of all to be restricted to curated content.

    • jinglebellz

      Bite sized chunks, so serialised novels etc should be able to work as sub models.

  • Anthony Pero

    I, personally, would find value in a subscription service that didn’t offer frontlist titles but had a large catalog of backlist titles from major publishers. While anecdotal, this is how I use the library–to discover writers. I burn through old titles, and then buy new ones from authors I like when they come out. That’s also how I used second-hand bookstores back in the day. A subscription service that allowed me to replace this behavior seems like a really good idea for titles that aren’t selling anymore. Assuming publishers are willing to adopt a payment model that is profitable for such a service.

    • The problem with models that are profitable for subscription services is the threat that they’ll cannibalize a la carte sales that are much more profitable for the publishers. But what you’re asking for very much seems to fit with what Carolyn Reidy observes as the benefit to S&S from their participation in the subscriptions.

      • Anthony Pero

        And of course, it only provides marketing value to the publishers if a) they still have the author in-house writing new books and b) the author is still living and writing new books. So I can see the fear of cannibalizing the books sales numbers, however small they might be. However, I would try to argue that for fiction, once people have bought in to a subscription model and are invested in it, they may be significantly less likely to buy a random backlist title at full price anyway, and funnel everything but their must-have essential frontlist reading through the subscription service.

        As I said in the past, I personally don’t find KU to be of any value, but for those of my friends who have it and pay for it, that’s all they read now, other than five or six frontlist titles a year. Again, all anecdotal, so YMMV.

      • Steven Zacharius

        There’s definitely some value in having backlist titles in the program depending on what the publisher and author is getting paid for them. It certainly does help bring readers to try an author.

  • GalleryP

    What I find interesting is that Oyster’s “certain to fail” business model, as correctly pointed out in this post, got funded.

    • Don’t be too surprised. VCs look for 10-to-1 shots; there are very few sure things. Were the odds less than 10 percent of success? I think so, but our tools for prediction aren’t that finely calibrated.

      • GalleryP

        Yes, but Amazon’s subscription service and Safari’s (as well as Knovel, Lynda, etc.) – though different models – are built for sustainability.

      • Steven Zacharius

        KU is going to end up being only Amazon published titles or indie published books. So I don’t think you can claim that it’s built on sustainability. I know that many of our most popular books in KU are being removed because they were too expensive for them to be kept in KU when they had to pay publishers a full royalty. I would never be able to explain to an author or be able to calculate royalties due to an author if they weren’t receiving the full amount normally due. It’s a shame because I do think that the program did allow readers to experiment with many authors that they may have not read before. I also didn’t see those sales cannibalizing the traditional purchase of the same title. Both were doing quite well. Romance is a very difficult category to have in a subscription service as I explained to Amazon, Scribd and Oyster when they all started.

      • Steve, as always, your informed comments are very much appreciated. It is useful info to learn that Amazon is throwing the publishers’ books out of KU.

      • Steven Zacharius

        Not all books yet, but a lot of the more popular titles that were being downloaded heavily. Can’t say I blame them one bit.

      • GalleryP

        I certainly don’t know the details of your arrangement with Amazon, but any deal that pays the full wholesale price based on some print equivalent for accessing a title on a subscription service is not sustainable, though Amazon (and maybe others with robust financials) could afford it as a loss leader. I could have it wrong, but I don’t think the majority of titles in KU get paid in that fashion, but as a share of a pool of funds as determined solely by Amazon, now based on page views. Others (Safari, Knovel, etc.) use different methods.

      • Steven Zacharius

        GalleryP, the payment pool that you are mentioning is only for self-published writers and not books from traditional publishers. Traditional publishers were paid full royalties from the subscription services otherwise they wouldn’t agree to make their books available. It was the marquee authors that were initially the draw to get people to join these services. As a publisher that publishes about 70% of books geared toward women’s fiction readers, I told all of the subscription services that this was not a viable business model because most women who read these types of books are voracious readers. This of course turned out to be the case. The model will probably continue to work for other books although I doubt mysteries will work very well either.

  • Ed Stackler

    I’d like to see the “surveys” cited by NYT indicating “young readers who are digital natives still prefer reading on paper.” Seems highly unlikely. I mean, during the school year they may read more on paper than on e-book, given school libraries, classrooms, passed-around books, assignments, and so on. But given their choice…? Gotta be digital. Then it’s up to the parents, I suppose, to support such a preference. Or not.

    • Ed, I don’t have the data at hand but I’ve seen surveys that show that same thing. I know it is counterintuitive, but there have been multiple reports indicating that kids still read paper. I don’t doubt the Times on that one, although CITING something would have been a good idea!

    • xist publishing

      I think that a child’s choice or preference is going to coincide with what they are allowed to do most frequently. Even though my 8-year-old does most of her summer reading on an iPad, she still isn’t allowed to bring it to school and so has read on paper exclusively since school started. She’s said that she doesn’t want to start a book she can’t read at school, so if she was surveyed at that time, I think she’d indicate a paper preference, even though she loves our Scribd and Epic subscriptions.

  • jinglebellz

    Very good points, purely as a reading service it wasn’t sustainable. But Google makes a smart move here, the discoverability and other tech are vital, and I wouldn’t be surprised if Google is aiming towards a content subscription via their MOBILE services. Given mobile is the future, I am fairly certain this would be the exact path they are taken – they aren’t very active in a pure retail sense, but this path has potential.

    ‘One is that, while publishers report ebook sales as a percentage of total book sales steady or slightly declining, Amazon says (and Russell Grandinetti was quoted in the article) their ebook sales are going up. Assuming all this is true, is the difference perhaps sales migrating away from publishers (which sales would be reported by the AAP stats they rely on) and moving to cheaper indie titles available only through Amazon (which sales would not)?’

    The new Author Earnings report goes into this and it’s pretty obvious that this is happening (btw that NYT article is very average, the comments worse).

    ‘Another is that publishers are raising prices on ebooks and making the price rises stick because of Agency. Is all the sales resistance created by higher prices resulting in print sales, or is some of it causing the book to be rejected for something cheaper? In other words, might total sales for many titles be less than publishers would have looked for before? (At least one agent tells me this is the case.)’

    This is my issue. I think, in my smaller than yours market, that the high prices push people away from buying the book altogether. The benefit of Agency is purely in my opinion that all prices are the same NOT that you can jack the price up. I would prefer lower prices for sure. But on the larger scale of reorganising books from container to content stream, I don’t know how viable that is for us to do…unless Google or someone does it. Would be interesting to work out the terms, since the content stream comes from the author and not the publisher….

    When you’re getting low star Amazon reviews because the ebook is 20+ bucks, there’s something going wrong.

    • The point that the value of agency is to level prices across retailers is one not to be lost. It certainly would appear that the publishers are pushing the envelope on price increases to what might be a breaking point. It will be interesting to see which publisher leads the trend back in the other direction.

  • EricWelch

    Some very interesting points. I have been a Scribd subscriber for quite a while now, and I think you are correct that having another core business is an advantage. A second reason is simply their software is better than Oyster’s. I tried both and dropped Oyster. I hope Scribd continues and would pay up to $20 per month (perhaps more) to maintain a subscription. I like them. With regard to ebook sales, I think agency and higher ebook prices, as you suggest, will hurt the publishers and drive people to Amazon where they can find many good titles in genres they like at far lower prices. I for one will just not pay more for an ebook than a print book. When that occurs, and I really want the book, I go to Amazon and purchase a used print copy, often for ridiculously low prices. A total loss for the publisher and author. I would prefer the ebook, but not at $12+ prices. (I’ve seen some non-fiction at $20+ for an ebook.) Unless their goal is indeed to eliminate the ebook market or to cede that business to Amazon, the “legacy” publishers are being very short-sighted.

    • I also had a Scribd subscription. I was fine with it, but I found I kept buying and reading Kindle books, so I didn’t use it enough to keep it. I thought their interface was just fine.

      Publishers are experimenting with book pricing. Things will change. They will learn.

  • Fujifan

    I suspect that the notion that Oyster “failed” is not quite the right idea – it seems that three guys, the founders and the CEO, got a better offer and jumped ship 🙂 Or is that too cynical even for big business?