There’s quite a bit of publishing about publishing going on in the next few weeks.
British academic John B. Thompson has written a solid scholarly history of book publishing in the past quarter century or so called “Book Wars” that will publish next month, focused on the arrival of ebooks and how publishers and the distribution system have been changed by that opportunity.
Robin Gaster has done a deep dive on Amazon called “Behemoth, Amazon Rising” which he is self-publishing. Gaster’s book and thinking go way beyond Amazon’s book business.
And on April 22, West Margin Books in Berkeley will release “The Family Business” by Keel Hunt, a history on approximately their 50th anniversary telling the story of the Ingram Content Companies. I was involved in the development of that project over the past few years.
All three of these books have an exciting story to tell because the book business has changed dramatically since the 1990s. As Robert Paris Riger and I wrote in “The Book Business: What Everyone Needs to Know” a couple of years ago, Amazon is a big part of that story. Ingram is a big part of Amazon’s story. And Ingram had a quarter century of innovating the book business before Amazon arrived.
I’ll do a post about Ingram and “The Family Business” nearer to its publication. Today’s post taps a review of some Amazon history that Gaster encouraged me to recall for a marketing session for his book that I joined a week or so ago. It turns out that I remembered a few things that escaped other documentation.
Jeff Bezos made a couple of absolutely critical early decisions that have not been appreciated as much as they should be. They have to do with what database he chose to use as a foundation for customers looking for a book on Amazon and about the delivery expectations he set for his customers from the very beginning. This required adroit use of the capabilities offered by the two national book wholesalers, Ingram and Baker & Taylor, two companies whose set of services for the book business really made Amazon possible.
Both companies had databases of published books. They, along with Bowker, who compile and deliver “Books in Print”, had the corpus of information Amazon needed to power their business. But Ingram, particularly, and Books in Print, were pretty rigorous about cleaning their databases. So books that were once available and no longer were didn’t appear. Baker & Taylor hadn’t been so tidy.
But Bezos saw opportunity in that deficiency. He recognized that a customer coming to Amazon and looking for a book would not know that not finding it in a search meant nobody would have it. In the retail world, that signal usually meant “go shop for it in another store.” Bezos saw the benefit of having his searchers find the book that was no longer available and being provided the information that it was “out of print”. That would encourage the customer to find a replacement at Amazon rather than search other retailers for the unavailable book.
On the other hand, Ingram’s greater efficiency made them a much better primary supplier than B&T. Books that were on hand at the Ingram warehouse closest to him (in northern Oregon when Amazon organized itself in Seattle), and he would know which those were, would be in his hands two days after he ordered them. All other books, whether they came from Baker & Taylor or other wholesalers, or from another Ingram warehouse, and especially if they had to come from the publisher, would take somewhat longer. Sometimes a lot longer.
Based on all this, Bezos invented the “promise date”: the date by which Amazon would deliver a book to you that you were ordering. How Amazon sourced it was invisible to you, but drove this information. If they could get the book from Ingram’s warehouse in Oregon, they knew they’d have it to ship to you in two days. They built in different amounts of time for other availability points. The most problematic were books they had to get directly from publishers.
Of course, communicating the expected delivery date was a boon to customers. But it was also a boon to Amazon. The savings on customer communication must have been cumulatively massive, with expectations managed that way starting literally before the order was placed. And, of course, the customer looking for a book they needed to get quickly could often shift their purchase if they really needed it; they would just look for a title with an acceptable promise date. Ingram’s nearby warehouse had all of the most popular books and were the most frequent source of supply, by far. So, usually, the promise date was a pleasant piece of news to the customer.
These two innovations, making it much more likely a customer would find the book they were looking for and making clear whether and when they would get it, were a secret sauce fueling Amazon’s early growth.
But there was another hidden magic trick in their set-up. It made all their sales cash-flow positive. They would get a customer order and cash today on a book Ingram would deliver day after tomorrow and expect payment on a month or two from now. Physical book retailers were constrained by their need to invest cash to stock their shelves. Amazon turned that upside down.
Gaster is the Amazon historian, and he traces their deep discounting philosophy to 2001. That is probably true in the broadest strategic sense, but heavy discounting of books began earlier than that unless my memory is playing tricks on me. And I have reason to remember.
Because it was before 2000 that Ingram recognized how substantially their own capabilities had enabled Amazon to grow quickly. After all, Ingram was the primary source for a big percentage of the books Amazon was selling on a cash-flow-positive basis.
But Ingram would never compete with its customers by selling directly in competition. So the response instead was to enable many Amazons through an initiative they called I-squared, S-squared: Ingram Internet Support Services. The percentage of books sold online as opposed to in stores at that point was paltry, but it was obviously growing. I2S2 would enable any bookstore to compete with Amazon for online sales because Ingram had added “third party distribution” to their suite of services: they’d ship the book to your customer for you with your bookstore’s name as the return address.
It was, briefly, very exciting to contemplate many bookstores — dozens, maybe hundreds — all enabled to deliver to the growing pool of internet-ordering customers. But it was “briefly” because shortly thereafter, or so memory says, Amazon started discounting books deeply. This was the beginning of a strategy I always called “using books as a customer acquisition tool rather than a margin generator”. So before I2S2 really got off the ground, the bookstores they intended to serve saw a no-margin business not worth the effort and the initiative ended.
Years later, I recalled this history in conversation with John Ingram and observed that “it must have been painful”. His answer said a lot. “Not really. Third party distribution is a key to our business, and I needed a deadline to get the capability built. I2S2 provided it and we have benefited greatly for many years since.”
Ingram, like Amazon, takes the long view.
There is one more tidbit of Amazon history which is worth recalling here because it is so underappreciated even with all the attention the company has received. And it has to do with the next real revolution Amazon spawned when they sharply accelerated ebook uptake with their release of the Kindle in November 2007.
Books had been read on hand-held devices since the late 1990s. Rocketbook and Softbook were two dedicated ebook readers that launched before 2000. Personal digital assistants, the Palm Pilot and others, also had the capability to display ebooks. Microsoft created MS Reader as the format for most of them; Palm had a proprietary Palm Digital format. And then Sony got into the game with the Sony Reader format, their proprietary device.
All of these hand-helds were “fed” the content through a PC. So the customer downloaded the content through their computer’s Internet connection and then transferred (“synched”) the content to the smaller device. Palm was the most successful format in those early days. But Palm wouldn’t allow any other retailer — not Amazon and not B&N — to deliver their content. That was the province of their subsidiary, Palm Digital. So the other ebook stores, such as they were, could only sell PDF ebooks (not very useful or popular) or MS Reader ebooks, which never got much market penetration.
This meant that print book retail couldn’t be used as a springboard to ebook retail. And that was among the things that Kindle changed.
One breakthrough for Kindle took place before the device was launched. Amazon set out to persuade the publishers to make a much bigger variety of titles available as ebooks. Sales to that point of ebooks had been meager and there was little incentive for publishers to bear the cost of creating ebook files even for their new titles, let alone for a lot of their backlist. Amazon was an important enough customer that their entreaties, which I believe also included some “help” with digitization, resulted in Kindle having a much bigger available catalog on launch than any ebook platform had ever had before.
But the critical step Amazon took was to enable loading the title directly into the device. Doing that required allowing the Kindle to connect directly with an Amazon server that could manage the transaction and deliver the file. But these were the days (shortly) before wifi. The Internet was reached by “dial-up”. And the dial-up charges from a Kindle would be paid by Amazon.
That constituted “risk”. The Kindle had modest and klunky Internet access. And it could be used to shop ebooks and even download free samples without making a purchase. And Amazon would be paying for those connections. If too many of the dial-ups didn’t include an ebook purchase, this could be expensive for Amazon.
That is part of the reason that the Kindle device was launched with a $400 retail price. Amazon sweetened the deal by making the ebooks available for no more than $9.99, even when they had actually paid the publisher more than that for the wholesale purchase of the file. That is, they were sometimes willing to lose money on the specific transaction. (Amazon always claimed that cumulatively their ebook sales always delivered positive margin.)
The result of these policies was that Kindle could become a pretty good deal if you read a lot of books. A Kindle user could well be saving ten dollars or more per purchase for the ebook compared to what the print book would cost them. That ate away the purchase price pretty quickly if you read three or four books a month. But it was an expensive way to read if you only read three or four books a year.
That meant there were big financial incentives for heavy readers to use Kindles. And, sure enough, it was evident pretty quickly that the small and growing number of Kindle owners starting in November 2007 were buying ebooks in per capita numbers far higher than most customers.
Kindle’s success was evident pretty quickly. Shortly thereafter, wifi became much more ubiquitous and was available to Barnes & Noble to use for their Nook reader when it launched in 2009 and, of course, was key to Apple’s iPad and iPhone ebook strategies another year or so later.
But Amazon’s jump on the market, a jump that really proved there was an ebook market, created a sustained advantage that persists to this day. Estimates vary, but the consensus is that Kindle captures three-fourths or so of ebook sales against the array of competitors that expanded to include Google as well.
Amazon’s unique success has been built on many specific decisions and insights. So many brilliant things have been done that some apparently can get forgotten. A mere quarter century after they began, those of us with long involvement in this business have to do our share to recognize and remember what has made all this happen.