Behemoth, Amazon Rising: Power and Seduction in the Age of Amazon

By Robin Gaster

Full of surprises and strategic insights. - Seth Goldman, co-founder of Honest Tea and Chair of Beyond Meat
__________________________

Amazon is the most extraordinary business story of our time. In 25 years, it’s become a $280 billion multi-sector giant, and within 5 years, Amazon may be the biggest company in America. Behemoth: Amazon Rising explains how Amazon built five interlocking rings logistics, Amazon Prime, the Amazon Marketplace, everyday low pricing, and constant innovation. The rings work together to create a moat too deep to scale around Amazon's retail empire.

But Amazon is not just any company. It’s brilliant, agile, cold, efficient, amoral, incredibly innovative, secretive, scary, seductive. It’s entirely customer-obsessed – which is great for customers, but not necessarily for producers. And there is no end to Amazon's ambitions, as it marches into logistics, cloud services, publishing, groceries and much more. As Jeff Bezos hands over to Andy Jaffy, Amazon’s culture and strategy will face enormous strategic challenges.

Amazon's relentless growth will test us profoundly. Behemoth: Amazon Rising describes the challenges posed by Amazon, and shows how to handle them - if we are smart, committed, and prepared to match Amazon's long-term perspective with an equally bold vision of our own.
____________________________

Incredible insights into the secrets of Amazon’s success - Vivek Wadwha, Harvard Law School

Brilliant! Nailed it. - Mike Shatzkin, The Idea Logical Company

Fills your bucket with meaningful takeaways and insights to help you and your team think in new and inventive ways about business, the customer, and society - Neil Ackerman, Johnson and Johnson

This riveting book... is a must-read on the most compelling challenges confronting society - David Audretch, Indiana University

It’s Amazon's 21st century version of manifest Destiny, and Gaster digs deep to show how we are all willingly along for the ride. - Amy Millman, President, Springboard Enterprises

A thoughtful, carefully researched analysis analysis of an amazing and sometimes threatening American behemoth - Charles Wessner, Georgetown University

Fascinating and comprehensive - John Zysman, UC Berkeley

A thoughtful and highly effective study in brand creation, design and execution. - Lou Pugliese, former CEO Moodlerooms

Gaster has his finger on the pulse of what was, is, and will continue to be one of the most disruptive corporations since Standard Oil. - Mark Walsh, former Head of Innovation and Investment, U.S. Small Business Administration

• Language: English
• Publisher: Robin Gaster
• Release date: Feb 28, 2021
• ISBN: 9781736001219

Robin Gaster

Dr. Robin Gaster is president of Incumetrics Inc. (a data and program evaluation consultancy), a visiting scholar at George Washington University, and a nationally recognized expert on innovation and small business. He has a PhD from UC Berkeley, and a BA from Oxford University. He has taught at the University of Virginia, worked in the IMF board room, and completed a Congressional Fellowship at the Office of Technology Assessment. He has since developed a consulting business serving clients in business, government, and the nonprofit sector seeking to understand the intersection of technology, innovation, economics, and politics.Dr. Gaster is currently a visiting researcher at the George Washington University Institute of Public Policy, and a member of the Montgomery County Policing Advisory Board. A full bio is available at www.incumetrics.com.

Book preview

BehemothAmazonRising-FrontCover.jpgTitlePage

Contents

Chapter 1 · Introduction

PART I – BASICS

Chapter 2 · Amazon and the Book Business

Chapter 3 · Prime Time: Amazon’s Customer Superglue

Chapter 4 · Logistics as Leverage

Chapter 5 · Marketplace: Amazon as Platform

Appendix A. Amazon Referral Fees, July 2020

Chapter 6 · Pricing: Table Stakes in Online Retail

Chapter 7 · Amazon Innovates

Appendix A. Amazon Leadership Principle

PART II – IMPACTS

Chapter 8 · Amazon the Anti-Brand

Chapter 9 · How Amazon Makes Its Money (and Loses Some Too)

Chapter 10 · Web Services: Amazon Internet Goldmine

Chapter 11 · Amazon@Work: Blue Collar Edition

Chapter 12 · Amazon@Work: White Collar Edition

PART III – FUTURES

Chapter 13 · Amazon in 2031: The Power of the Double-S Curve

Chapter 14 · Prelude to Action: Considering Antitrust and Utility Regulation

Chapter 15 · Balancing the Behemoth

Afterword

Acknowledgments

Copyright 2020 by Robin Gaster

All rights reserved. No part of this book may be reproduced in any form or by any electronic or mechanical means, including information storage and retrieval systems, without permission in writing from the publisher, except by reviewers, who may quote brief passages in a review.

ISBN: 9781736001219

Cover design by Mladen Božić

Typesetting by C’est Beau Designs

Chapter 1

Introduction

Your margin is my opportunity

–Jeff Bezos

Amazon is the most extraordinary and important business story of our time. Facebook has more members and is our social network. Google sits right at the heart of the information tsunami. Apple has by far the prettiest toys. But starting 25 years ago as a tiny online bookstore, Amazon now stands astride the e-powered river of goods that flows through the American economy. It is a retailer, a marketplace, an electronic infrastructure, a publisher, an advertising channel, a distributor. It is increasingly the arbiter of retail, the pacesetter for employment, and a private taxing authority, taking its bite of every transaction on its Marketplace.

And Amazon is just getting started. Just in the last year, its share of books sold in the US tipped to over 50%. It signed hundreds of exclusive brand deals – a new strategic initiative in retail. It grew its dominant and highly profitable Internet infrastructure and services business by 35%. It more than doubled its (even more profitable) advertising business. Its video studios won seven Emmys. Twitch usage by gamers reached 560 billion minutes – up 58%. Including advertising revenues, Amazon now takes 36–40% of sellers’ gross revenues on the Amazon Marketplace. Amazon Prime reached 112 million members. It launched or accelerated a slew of important new initiatives, in groceries, health care, even satellites. Across all its business, Amazon is automating rapidly and leveraging the power of enormous and growing capabilities in AI and machine learning. The sheer range and pace of change is formidable.

Nor is Amazon settling for dominance in online retail and the cloud. Amazon’s self-described mission is to build Earth’s most customer-centric company, a place where customers can come to find and discover anything and everything they might want to buy online.¹ That vision has no natural limits: there are customers everywhere, in every sector and every business, and they could all be served by Amazon, so its expansion has no obvious end point. Other firms are limited by the scope of their industry sector, and believe (correctly) that hopping boundaries is risky. Few firms make the jump, and if they do, it’s usually to a closely aligned business. Amazon doesn’t respect those boundaries at all. As Jeff Bezos also said, your margin is my opportunity, and every functional business operates with a margin, and is therefore a potential target.

Over the past 25 years, Amazon has built five interlocking flywheels for accelerating and defending its existing business interests, and to create platforms from which to expand into new ones. Those flywheels are:

Its distribution and delivery system. For ecommerce, it is often cheaper and better than the US Post Office, FedEx, or UPS. Amazon now delivers more than 60% of its own packages.

Amazon Prime, its membership club, which is customer superglue. Those 112 million members start their search for goods on Amazon, and most end there as well.

Its enormous catalog. Jeff Bezos has always aspired to own the Everything Store – selling every item in the known universe. Thanks to the Marketplace, Amazon has 350 million items in its catalog – more than 20 times the number available through Walmart.com.

Relentless competition on price. Amazon squeezes its suppliers just like Walmart has always done, and it uses its control over access to its Marketplace to enforce low prices there too: independent sellers have little choice but to comply.

Innovation. Everything Amazon does is accelerated and multiplied by the relentless introduction of innovative ideas and technologies. Amazon is a machine for innovation.

Each flywheel is supported by four core accelerants: 1) Amazon Web Services (AWS), the biggest provider of cloud infrastructure in the world, provides the technical support; 2) Amazon’s culture and organization enable the rapid introduction of new services and the improvement of existing ones; 3) its essentially unlimited access to investment finance and its existing assets like its logistics, which Amazon views more like muscles (to be used and grown) than as bank accounts to be carefully husbanded and protected; and 4) the trust of consumers, who are willing to go where Amazon leads: Amazon is the most trusted company in America.

Each one of these flywheels are important and so are the capabilities that underpin them. But it’s their interaction that creates Amazon’s uniquely efficient ecosystem, and its looming threat to competitors. This interaction – ecosystem – gives Amazon nearly unlimited leverage not just in the sectors it currently dominates, but in others that it’s targeting. For example, the key strategic choice to sell items from other sellers on Amazon’s Marketplace platform dramatically increased the number of items available, which both attracts customers and makes possible Amazon’s enormous logistics infrastructure (as well as risk-free commissions for Amazon). Fast free delivery becomes possible at a certain scale, and that attracts more customers, hence more sellers, who bring a larger catalog of items while competition drives down prices, which attracts yet more customers. All the flywheels accelerate. It’s an extraordinary story, and it raises plenty of questions:

Where did Amazon come from? How did it grow so big so fast?

What can we learn from the history? Can we distill key lessons about objectives, strategy, tactics, and especially corporate culture?

Where is Amazon going? What will it look like ten years from now?

What should we – the collective we – do about it? Is Amazon a threat? Should we simply applaud? Are there characteristics to worry about? And if so, what should we do?

Behemoth addresses these questions in three sections. Part One covers the basics: the core elements that underpin Amazon’s success – selection, pricing, convenience, and trust – and adds a deeper explanation, showing how Amazon’s capacity to innovate has fueled its extraordinary growth. Part Two looks at the impacts of Amazon’s success: its relationship to brands, the real story of Amazon’s money (including its big losses in retail), AWS (Amazon’s massive Internet services business), and Amazon’s complicated and ambiguous relationship to its blue- and white-collar workers. Part Three leans into the future: where is Amazon going? It will soon be the biggest employer in the US, bigger than Walmart. Within ten years it could have $2 trillion or more in annual revenues. Will it eventually collapse like previous winners in retail? The final chapters explore the tools available to manage Amazon, and explain how to ensure that the Amazon we get in the future is the Amazon we want.

Amazon has transformed how we shop and how we discover new products. Its customers love it. But… it’s already the Everything Store. Ten years from now, will it stand alone, astride that river of online retail, taking its increasingly heavy toll on every transaction? As traditional levees break and ecommerce floods into new markets, will Amazon own those customers as well? Its ambitions extend far beyond retail, into healthcare, and finance, and logistics, and entertainment, and advertising, and publishing, and groceries, and home security and home automation, and Internet provided via satellites, and of course the innumerable digital services built on the AWS platform.

Amazon offers so many different facets, so many opposing perspectives, that these are difficult questions to answer. Insiders draw parallels with highly mission-driven organizations like the Jesuits and the US Marine Corps. Their characteristic solidarity, mission focus, secrecy, and insularity are just like Amazon. As Bill Vass, an Amazon VP once put it, It’s like a religion about the customer.² As we all know, religions provide comfort, solace, and guidance, but religious obsession has a dark side as well.

Perhaps an analogy can help. Evolution has selected for humans who are good at reading other people. So take a breath, and try to imagine what Amazon would be like if it were in fact a single human being, a person. If Amazon were a human, it would be… brilliant, flexible, cold, efficient, amoral, agile, obsessive, secretive, transactional, innovative, ungenerous, inspirational, ruthless, greedy, effective, cheap, visionary, disruptive, super-confident, messianic. It would beat Deep Blue at strategy games. It would have no friends, and no partners. It would echo Nietzsche’s Ubermensch, a more perfect human altogether, shorn of the emotional layers that connect us and also limit us. Amazon cares only about itself, and its customers. Fully realized, would Amazon’s ideal world contain anything more than… Amazon, its customers, and the robots that serve them?

Are other Big Tech companies like that? To some degree, some of the time, but the pictures that finally emerge are different. There’s IBM’s buttoned-down organization man, Apple’s product-obsessed control-freak hipster, Google’s slightly academic math whiz, Facebook’s careless bro programmer. But none is remotely like Amazon; and none radiates so powerfully the triggers for that lizard-brain message flashing danger.

Amazon as human is a metaphor, a heuristic device to help us integrate the data and the evidence to get a feel for Amazon. As we explore Amazon’s ecosystem, and the coming Age of Amazon, echoes of Amazon as human are everywhere – in logistics, in its extraordinary Marketplace, in sectors like books and electronics, in Amazon’s media empire.

Maybe those echoes will help us answer the central question: is Amazon leading us to a promised land, a consumer-led utopia, where prices are lower, selection is better, convenience is extraordinary – where we just voice our wishes to the air, and goods magically show up on our porch next day (or sooner)? Or are we heading, perhaps irreversibly, into a dystopia of fungible and easily discarded workers controlled through relentless surveillance and second-by-second metrics, operating as best they can within Amazon’s obsessive culture, in service to customers who themselves are ever more closely managed by Amazon as it relentlessly expands into every aspect of their lives?

The final chapter in Behemoth offers a pathway for retaining the benefits of Amazon without simply ceding control, without just allowing Amazon to continue single-mindedly, obsessively, endlessly, toward cheaper and faster delivery, more selection, more services, more everything, without regard for collateral damage. In 1964, Barry Goldwater famously said that extremism in defense of liberty is no vice. Voters didn’t agree. In the end, consumers too will turn against an Amazon that is obsessively expanding to serve every imaginable need of every customer, at any cost. Balancing the Behemoth is what we can and must do.

Endnotes

Jeff Bezos, Amazon letter to shareholders, 1999

Jon Swartz, How Amazon Created AWS and Changed Technology Forever, MarketWatch, December 7, 2019.

Part I – Basics

Chapter 2

Amazon and the Book Business

No trader has ever been so successful in its concentration on consumer pricing – all this impervious, of course, to the broader considerations of the overall welfare of the industries in which it is operating. It’s all so simple. Make and build your brand on a reputation for absolutely rock-bottom pricing. Do this single-mindedly and ruthlessly.

–Tim Waterstone, bookseller¹

Books were Amazon’s first successful market. A combination of strategy, leverage, vision, and ruthlessness has brushed aside the entrenched competition, transforming the book business. Twenty-five years ago, Jeff Bezos, Marianne Bezos, and a handful of staff were shipping books out of the Bezos garage. By 2020, Amazon sold just over half of all the printed books sold in the US, as well as 83% of all ebooks, and a large majority of audiobooks. And Amazon is far from done with books: its self-publishing platform is a direct attack on the traditional business model for publishers.

Amazon’s journey from a microscopic online bookstore to industry dominance rested on its systematic and long-term investment in four core building blocks: Logistics, Amazon Prime, Pricing, and the Amazon Marketplace, along with persistent bets on innovation that its competitors would not make. Dominance required multiple iterations on multiple fronts, so the path is not sequential; it’s not as though Amazon figured out logistics and then figured out customers and then added Marketplace and pricing. Amazon was working all of them out more or less simultaneously.

Thus, rather than a sequence, these components fit together like stones in a pyramid, each a solid mass in itself, but also providing the base foundation stone for the next step. Together, they drove the irresistible growth of Amazon’s reach and capabilities.

Amazon Begins: The Online Bookstore

We start at the point of inception. Jeff Bezos picked books as Amazon’s first market segment when he wanted to start an online business. It wasn’t an accident, and it wasn’t because Bezos loved books. Books are standardized – every published book is exactly the same wherever it is sold. No need to get into personalization like apparel or jewelry. Books weigh little and come in regular sizes, so they are easy to ship – no odd-shaped machinery or heavy exercise benches, and the US Postal Service (USPS) even offers a discount for media mail. As Shel Kaphen (Amazon’s first CTO) said, the choice of books was totally based on the property of books as a product.²

Helpfully, Amazon found it easy to acquire inventory. There were only two big book wholesalers – Ingram and Baker&Taylor. Both happily sold books at a standard discount to the upstart online bookstore. Through just two distributors, Amazon got universal access to all books in print in the US. Toys and electronics, sectors that Amazon entered a bit later, were completely different and much more difficult.

Information was also standardized. Books in Print maintained a listing and ID system which was already available electronically and used everywhere. So Amazon didn’t have to invent a catalog and persuade customers to use it – Books in Print was the universal reference. In the end, Amazon chose to use a lower quality version of the catalog from Baker&Tylor, precisely because it included a lot of out of print books. Even at this stage, Amazon had recognized that it was going to dominate the long tail: older, low-volume, and out-of-print books.³ Specific customer demographics were low-hanging fruit: clusters of customers not well served by bookstores, such as people who weren’t near a bookstore or couldn’t get around easily, and those who wanted one of the millions of titles not carried in bookstores.

And of course, Amazon had a huge and immediate cost advantage. The standard wholesale discount at the time was about 50% – bookstores actually paid publishers only about 50% of the list price of a book. That discount covered the cost of running the bookstore and the bookstore’s profit. Because Amazon had no bricks-and-mortar expenses, it could pass much of that discount on to customers. And even though bookstores were willing to order books for customers, they usually charged full retail price, often plus service charges. Amazon typically discounted its books by 40% off list price for best sellers and 10% off for everything else. Customers paid for the shipping, and typically waited a week or two for delivery.

So Amazon’s initial business model was simple. It was a traditional bookstore, except it sold only online – like old-fashioned mail-order, except the orders came through its web site. It acted as middleman, taking orders one by one from customers. Those orders were sent to one of the two distributors, which shipped the book to Amazon, which in turn delivered it to customers via the USPS. But Amazon was already searching for ways to connect better to customers. It quickly hired an editorial group to add content to the home page and elsewhere on the web site, and to suggest books to readers, aiming to replicate the feel of a bookstore, but online.⁴

This first model was highly successful. Revenues grew rapidly, 30–40% per month in early 1996,⁵ and the initial business model carried Amazon through its IPO and into the public financial markets in 1997. At this stage, Amazon was still a price-taker. It accepted the terms imposed by the wholesalers, and just used the standard shipping arrangements that were available to any business through USPS and UPS. Its own internal ordering and tracking processes could (charitably) be described as primitive. What transformed Amazon (and the book business) were its three core initiatives of the early 2000’s: logistics, Amazon Prime, and the Amazon Marketplace.

Logistics as Leverage

For years, Amazon struggled with distribution. Long after Amazon moved out of that garage and into offices in Seattle, and even after it purchased its first warehouse in a run-down district south of downtown Seattle, distribution seemed likely to kill Amazon altogether: every succeeding Christmas rush was bigger than the last, and required ever more heroic efforts to ship out orders. Amazon’s distribution system wasn’t good enough to predict the day on which a package would reach the customer, or to track failures effectively, or to deal with the massive pre-Christmas rush. In many ways, solving distribution was the decision that set the course for the company; disappointed customers probably wouldn’t be back. But building an entire distribution system seemed impossibly difficult.

Thus by 2001, Amazon faced a blunt choice: it could outsource its logistics to the capable partners it was already working with (FedEx, UPS, and USPS); or it could build its own logistics network – including software, warehouses, systems, and eventually last mile delivery. This was a huge and risky decision, with long-term strategic implications. To build its own network, Amazon would have to spend big money on warehouses and infrastructure; certainly enough to swallow any profits for the next few years. But that’s what Amazon did. It rejected conventional wisdom in favor of a different vision, where logistics became a linch-pin of Amazon’s growth strategy.

That story is fully described in Chapter 4, Logistics as Leverage. Briefly, Amazon went through four logistics models in twenty years. By 2020, Amazon owned the best ecommerce delivery system in the world, a complex mix of specialized and increasingly automated warehouses linked to each other by Amazon’s own freight and air operations, and connected to customers through Amazon’s increasing last mile delivery capacity. That transformation cut delivery times, improved efficiency, and reduced costs. In books, the key was convenience. Lower costs were helpful, but what mattered was that customers received their books quickly; often before they could even plan a trip to the bookstore.

Prime and Beyond: Amazon’s Customer Superglue

Amazon’s mission is to be the most customer-centric company on Earth. Before Amazon, big bookstores knew little about their actual customers. They might have a loyalty card scheme, but paid almost no attention to their customers as individuals. When you walked into a chain bookstore, staff didn’t leap forward to suggest that because you had enjoyed Harry Potter you might also enjoy The Hunger Games. There might be staff selections, but those selections were about them, not you. They didn’t know that you had even bought Harry Potter, let alone that you rated it highly. Small bookstores did know their regular customers, but they were never Amazon’s competitors; more like collateral damage in its war with the big chains.

In 2013, when Amazon bought Goodreads, the leading web site for book reviews, it had 20 million members, which Amazon seemed to believe could be a great market for Amazon. But Goodreads was soon supplanted by Amazon Prime. Prime is the buyers’ club for Amazon’s customers. It now costs $119 a year in the US, and in exchange offers fast free shipping, plenty of discounts and deals, and free high-quality video and other entertainment. Prime started in 2004, reached 10 million members in 2010, and had 112 million as of June 2019 (see Chapter 3, Prime, for details).⁶ Amazon Prime is almost everywhere, and because Prime members pay upfront, the more they buy, the better the free shipping deal is for them. So Prime glues customers to Amazon, making them much more likely to search and shop there; they also spend much more on Amazon items.

Prime has been a powerful weapon in Amazon’s war with the big book chains, as well as a key lever over publishers. The trifecta of free shipping, free entertainment, and endless deals on pricing makes Prime irresistible. So itbecame superglue for Amazon customers, and having all the online customers is Amazon’s true superpower.

Pricing and Leverage: Squeezing the Publishers

Amazon adopted Walmart’s everyday low prices in 2001. Since then, while customers cannot be sure that Amazon has absolutely the lowest price, they know that Amazon will have a low enough price, so searching further (especially for relatively inexpensive items like books) may not be worth the effort. Amazon makes sure customers know this: search results and product pages prominently display a discount off list prices. For example, the first Harry Potter book is discounted in hardback by 75%; a more serious book like Thomas Piketty’s Capital is discounted by 44%.⁷

Starting around 2004, Amazon began using its growing size as a bookseller to squeeze the publishers. It let publishers avoid many of the costs imposed by bookstores – paying for prominent positioning in the store, the high cost of returns, other marketing costs. Amazon however believed that those cost reductions should be passed directly on to customers, via the extraction of larger discounts from publishers. Instead of the standard wholesale discount of 50%, perhaps Amazon should receive 55% or 58%. It could then cut prices, which would attract yet more customers.

To squeeze publishers, Amazon started by picking off selected victims. Its Gazelle program targeted those small and medium-sized publishers most dependent on Amazon for sales of their backlist. They were squeezed ruthlessly for bigger discounts and longer payment terms. Brad Stone reports that Jeff Bezos himself suggested that Amazon should approach these publishers the way a cheetah approaches a gazelle.⁸ Picking off publishers one at a time made a collective response impossible (and probably illegal anyway, under antitrust law).

"By his own admission, [Randy] Miller⁹ took an almost sadistic delight in pressuring book publishers to give Amazon more favorable financial terms. He ranked all of the European publishers by their sales and by Amazon’s profit margins on their books. Then he and his colleagues persuaded the lagging publishers to alter their deals and give Amazon better terms, once again with the threat of decreased promotion on the site. Miller says he and his colleagues called the program Pay to Play."

–Brad Stone (2014)

Amazon didn’t have to entirely block publishers from its platform. Just excluding them from its recommendation and review engines – thus cratering sales – was usually enough to break their resistance (mostly within 30 days, according to Chris Smith, a senior Amazon book buyer at the time). A shut-off could cost the publisher 40% of its sales on Amazon, and those sales had quickly become a key source of publisher revenues.¹⁰ As the Gazelle program gathered steam, Amazon moved up the ladder and went after bigger publishers, using the same tactics. Again, shutting off website features was often enough to force concessions. According to industry insiders quoted by George Packer, only a few years later even huge publishing houses like Random House were offering Amazon a standard discount of 53%.¹¹

These savings were largely passed on to customers, who benefited directly: book prices, especially for hardbacks – the flagship products of traditional publishers – fell rapidly. That was tough on the publishers, who had to eat the discounts. But lower prices also made Amazon even more competitive against big chains like Borders (as well as the small independent bookstores), and increased its price advantage over competing online stores as well.¹²

In the longer run, Amazon’s demands eroded the relationship between publishers and authors. While the public was paying high prices for books, there was money for publishers to try new authors, and to offer substantial advances (as well as those famous three-martini lunches). As Amazon demanded higher discounts, that money dried up; publishers offered smaller advances, and tried to focus more tightly on books that were likely to be successful. Amazon saw itself as very different from traditional publishers and bookstores. As George Packer described, Amazon executives considered publishers to be antediluvian losers with rotary phones and inventory systems designed in 1968 and warehouses full of crap…. the New York publishing business was just this cloistered, Gilded Age antique just barely getting by in a sort of Colonial Williamsburg of commerce.¹³

At the time, Amazon controlled the final price of books on its site; it bought at a discount and could then charge what it liked. That was the traditional wholesale model. Eventually, the big publishers pushed back, focusing on the new category of ebooks (see Kindle section, below). In 2010, five big publishers moved to an agency model, through a deal to sell ebooks through the Apple iTunes store, which they hoped could become a full-scale competitor to Amazon. Apple simply charged a 30% commission, so publishers controlled the price of their books. This was the agency model.

Amazon retaliated, removing the buy button from all titles published by one of the big five, Macmillan. That made Macmillan books hard to find and hence buy, which in turn generated customer complaints and negative press. The final result was a draw favoring Amazon. The agency model stuck with four of the big five publishers (although the biggest, Penguin Random House, still uses the wholesale model) but the publishers and Apple were eventually caught up in a Department of Justice antitrust suit – which claimed that they had conspired to raise prices – and paid heavily to settle it.¹⁴ Amazon showed that it could live with the agency model if necessary. And that was the last serious pushback from the publishers.

Over time, Amazon has extracted deeper discounts and higher commissions, and now dominates US sales of printed books, eBooks, and audio books. It is truly the behemoth of the book business.

Innovation

Amazon is known for its big innovations, like Amazon Web Services or of course the Kindle (discussed below). But Amazon’s earlier innovations in books were perhaps even more important. It introduced customer reviews very early on, and automated personalized recommendations soon after. These two key innovations changed the way people bought books and of course other goods as well.

Reviews and recommendations

Amazon introduced its first recommendations engine in 1999, based on identifying customer purchasing patterns and finding other similar customers and observing their buying patterns. That let Amazon recommend purchases that similar customers had also bought. This worked very well – similar customers do have similar tastes and needs. For the first time, Amazon as book distributor could at scale successfully recommend products differently to every individual customer. Those automated recommendations work: in one experiment, they accounted for an increase of 8–20% in sales revenues.¹⁵ So Amazon doesn’t have to segment its customers into the usual groups like soccer moms and Millennials and romance readers. It segments its customers into an audience of one: that specific individual customer, with their unique buying and browsing patterns.

Automated recommendations transformed book distribution. Bookstores can offer only a handful of recommendations – staff have only so much time, and space is limited. As Amazon’s customer base grew, its algorithms became better at predicting what items customers might want. For Amazon, more customers meant more data, which generated better recommendations, so scale became a blessing, not a curse. Today, it’s hard to remember just how revolutionary this was: before Amazon, no large retailer had ever offered