The Shatzkin Files


Subscription services for ebooks progress to becoming a real experiment


My long-held conviction that broad-based subscriptions for ebooks were not likely to work is partly based on facts that are now changing. It is still by no means a slam dunk that ebooks must go where Spotify has taken digital music and Netflix has taken the digital distribution of TV and movies, but it looks more likely today than it did six months ago. Still, looks could be deceiving.

The core of subscription economics is to pay less to the content supplier than they earn other ways to give you some headroom to create a value proposition for consumers. That’s how Spotify and Netflix work. That’s how Book-of-the-Month Club works.

And what happens over time with subscription services is that the power of “brand” passes from the individual titles (and authors) to the subscription service itself. In order to attract customers, a subscription offer depends on recognizable branded product to bring people in. But, over time, the value shifts. Eventually, a subscriber-reader can become used to choosing from what the service offers and will either not know about, skip, or accept purchasing the occasional book s/he wants outside the service if it isn’t offered inside. (A varient of this reality is playing out now in the Amazon-Hachette dispute, where Amazon’s brand power, including people who have a subscription to PRIME free freight, makes any particular publishers’ books subordinate to the seller’s brand with the consumer.)

None of this is particularly startling or insightful. Every agent for a big author knows it. Until very recently, that has meant that big publishers did not put the big books from big authors into these services. When the first shoe — the HarperCollins shoe — dropped and the second biggest trade publisher (and by far the largest of the four majors who trail Penguin Random House) went into Oyster and Scribd several months ago, I should have taken on board that the perception of agents must be changing. Now, with S&S having joined them, and with major authors included in the offerings from both companies, it is clear that agents are withdrawing their objections.

There are three reasons for this.

One is that the incumbents in the book business are circling the wagons against the dominance of book retailing’s most powerful brand: Amazon. As the market share of and customer loyalty to the industry’s biggest player grows, other dangers — such as those posed by subscription services if they mature — look relatively less onerous.

The second is that publishers and agents love the opportunity to establish that if subscription services want to “play” in publishing, they’ll have to pay for each ebook on a purchase deal. That is: the subscription services are establishing their “model”. And the publishers and authors are also establishing theirs!

The last is that the two big current subscription efforts are disdaining the fundamental economics to get their services started. The current model, as outlined by S&S CEO Carolyn Reidy in a letter to agents announcing her house’s participation, is that the service buys a copy of the book at “full price” when a “a certain threshold of reading has been surpassed for a given title”. But her letter also suggests that authors make even more money on these sales than they do on normal sales, which implies that Scribd and Oyster are paying more than 70 percent of the retail price for the privilege of using these books. (I have heard a range of numbers for where the threshhold of use to trigger payment is, from 10% to 40%, but I have no idea what it is and how it might differ among publishers.) Whether they’re paying 70% of retail or more, that means that it would take no more than two full-priced S&S or HarperCollins (assuming they have the same deal) titles a month to cost the service more than the revenue from a full-freight subscriber. And if the subscriber came through iOS, Apple’s 30% cut off the top would mean that even one major publisher ebook being read in a month will likely put the service in a deficit position.

Even when the purchase model is favorable, which this one appears not to be, it has been generally understood that the viability of a subscription model depends on what is called “breakage” or “health club economics” to succeed. They count on the expectation that relatively few subscribers will read and trigger payments on two, three, four books a month compared to many who will read one or less than one, or who will choose from among books (like public domain titles) that cost the services less or nothing.

The first of the subscription services for books — Safari — used a model that is much safer for the services because it assures cost stability, assigning a percentage of the revenue as a pool to compensate publishers rather than guaranteeing a purchase for every read as Scribd and Oyster are doing. I expect the purchase model to be very difficult, if not impossible, to sustain. But persuading the big players to come in depended on getting away from the safe “pool” model and purchasing the ebook anew for each new user.

A huge danger for the subscription services is the likelihood that subscriptions will be shared within families (let alone within dormitories!) That could drive up the average use per subscriber very quickly if it isn’t controlled.

Only now, with two of the Big Five in the game, giving the services about a third of the most commercial backlist titles in publishing, can they really find out whether the price-and-cost model they’ve set up will work to give them a profit. (It is important to note that HarperCollins and Simon & Schuster only put backlist titles into the services, so the most attractive commercial titles, which are new, are not part of the offer. This also means that all shoppers and purchasers of new titles will continue to use the stand-alone purchase model.)

I’m sure Scribd and Oyster have data and analytical skills that I don’t have. But, intuitively, this seems like a tough proposition. Subcription services are attractive to consumers because they’re bargains. If you normally read a single ebook or month or fewer, the $8.99 monthly subscription charge would not seem attractive. But if you read an ebook or two a month or more, the services will likely lose money on you.

Meanwhile, there are two players currently sitting on the sidelines that could really disrupt the subscription incumbents (which also include Spain-based 24Symbols, which has been around much longer than Scribd and Oyster but which hasn’t succeeded so far at bringing in the big publishers and the big books.)

There are rumors that Amazon is already canvassing for participants to deliver a subscription service of their own. Of course, they really already have one. Their PRIME subscription offer, for which the headline attraction is free shipping of hard goods, also includes access to the “Kindle Owners Lending Library”, which is effectively a broad-based ebook subscription service with some limitations and a far less robust title selection than Scribd and Oyster. Amazon could find ways to expand that. Will they match the implied compensation from Scribd and Oyster and pay more than the 70 percent which is the current standard for sales by agency publishers (which, therefore, becomes the basis for royalties to big authors)? One would suspect they would want something in return for that: exclusives, perhaps, or earlier access to the titles than Scribd and Oyster have.

Of course, Amazon (or Google or Kobo or Nook or Apple) would have an automatic advantage over the subscription incumbents if they decided to compete with them. Because they already sell all the books, they could sell you the books you wanted that weren’t in the service as part of a single offer.

The other future player of consequence is Penguin Random House, which by itself has well-known commercial titles that exceed in number what the services would have even if they signed up one more of the remaining big publishers. Hachette’s chief marketing and sales officer, Evan Schnittman, is quoted by the Wall Street Journal saying that this model is “not for us”. That leaves Macmillan, but even if Scribd and Oyster get them, PRH could have the most attractive title base on offer all by itself.

When I speculated some time ago about the opportunity PRH had to do this, one of their executives set me straight about why they wouldn’t. What I was told was that PRH was not thrilled by the idea of turning $500 and $1000 a year book customers into $100 a year book customers. Of course, that calculus changes for them if others are succeeding at doing that, and those new $100/year customers are then one step further removed from buying PRH books.

If PRH did this, they’d have one big decision to make: do they attempt to include the biggest titles from the rest of publishing in their offering or not. They’d already be starting with the most attractive title selection, but the Scribd and Oyster assortments would be competitive. If they went for some of the rest — even if only the top 10 percent of the rest — PRH could present a noticeably more attractive selection than Scribd or Oyster.

Would other publishers go in with them? I’d say, “probably”, because they can’t afford not to have their biggest books exposed to all possible substantial audiences, and PRH would almost certainly have the biggest subscription audience.

Would Penguin Random House want them? I’d say, “probably” again. It would stamp their offering as by far the best, and they’d still be advantaged dealing with authors because they’d be the only publisher not paying a third party to get the subscription revenue.

If “fear of Amazon” is the factor that made big agents relent in their opposition to subscription, would they also support joining an Amazon subscription service? That’s a trickier call, but as noted above, Amazon would have the capability to sweeten their offer to make it more compelling if that’s what they had to do.

But the main thing that works in favor of participation, now that the dam may have broken, is the psychology of trade publishing. Every big trade publisher has grown to be what they are today by selling their publications through intermediaries. Bookstores and then Amazon became the “gatekeepers”, owners of the customers. There was a symbiotic relationship: the retailers depended on publishers to deliver products to please their consumers and the publishers depended on the retailers to merchandise their offerings and manage the transactions. Access to a retailer’s customers first depended on getting your offerings into their store and then on having them be seen by the largest possible number of the store’s customers. That meant front tables and face-out display in the physical world; it means the right screen real estate, recommendations, and response to search terms in the virtual one.

That’s why the current hegemonies of Barnes & Noble and Amazon are so disconcerting to publishers. And that’s why the potential control of customer access by Scribd or Oyster might now look more like counterweight than threat.

Of course, it is also possible that the price-and-payment models Scribd and Oyster have begun with will prove unsustainable and that HarperCollins and Simon & Schuster — and their authors — will simply be the beneficiaries of a short-term bonanza financed by money that took a flyer that didn’t pay off. (And they’re not done taking those flyers.) That seems to me at least as likely as an outcome as these broad subscription offers becoming a permanent part of the bookselling landscape.

A lot going on around our place, so we haven’t had the time to switch away from what has become the horrendous service from Feedburner distributing The Shatzkin Files to its email subscribers. This one from last week on Amazon and Hachette (which is also linked to above) never was sent. (Of course, as I write this, who knows if this one will be or not?) It was written before the latest escalation where Amazon has removed pre-order buttons from Hachette book, a nasty blow that makes getting books on the bestseller list the week they come out very much harder. A lot has been written on this subject, but I think it still delivers some consideration of what it all means that hasn’t been picked up anywhere else.

  Back to blog