by Brad Stone
The story was discussed and celebrated inside Amazon for years—not as an example of volatile leadership, but as an illustration of the high standards of company leaders, and the grit and resilience of employees, which were needed to execute complex services like FBA.
Bezos believed that FBA had a chance both to succeed and to have an outsized impact on the company. “I need you to deliver so that we can fund the portfolio of other businesses” that also have enormous potential, he said, according to Tom Taylor, instructing the FBA team that they should be moving three times faster than normal Amazon teams.
Other “Jeffisms,” recorded by the FBA team in their annual OP1 planning session with Bezos, further shaped their perspective. Among his greatest hits, recorded in a memo that was later passed to me:
“Focus on lowering cost structure. It is better to have low costs and then charge to maximize your value versus charging to cover costs.”
“Having a stupid rate card equals stupid things happen. Rate cards must be equal to the value.”
“We do not charge more because we can’t figure out how to make it cost less. We invent to make it cost less.”
“We should be able to fulfill 100% of the 3P business. I do not know what the debate is, yes, we must fulfill low priced selection, it is crucial.”
“Averages are bad measures. I want to see actuals, highs, lows and why—not an average. An average is just lazy.”
By 2014, after more than a decade of such blunt guidance, the service became profitable for the first time and the number of sellers using FBA was growing briskly. “Don’t pretend that anyone else on the project was some kind of genius, because that is just not true,” said Neil Ackerman, a former FBA executive. “Jeff was the one who challenged everyone to lower fees and not focus on income, but instead to put our attention to adding sellers and growing selection. He knew that was how we could get the business to scale and become profitable. Jeff always said that when you focus on the business inputs, then the outputs such as revenue and income will take care of themselves.”
At the same time Bezos was shaping FBA, he was nurturing its conjoined twin, the Amazon Marketplace, which allowed third-party sellers to register on Amazon’s website and display their new or used products. In 2007, the marketplace was already a few years old but was basically just a dusty repository for used books, accounting for a meager 13 percent of all units sold on the site. Bezos was frustrated by its lack of progress and tore up the team’s documents in OP1 meetings and demanded more ambitious rewrites. “How would you get a million sellers into this marketplace?” he asked the stream of executives whom he interviewed to take over the group.
Eventually, he found the right person. In early 2009, Peter Faricy, the head of Amazon’s thriving music and movies business in the years before Prime Video, invited the actor Tom Cruise to speak at an all-hands meeting at Benaroya Hall in downtown Seattle. Backstage, Bezos and Cruise fell so deep into conversation about airplanes and space travel that Faricy couldn’t get the CEO out on stage on time for his Q&A with employees. Afterward, in appreciation for his department’s success, Bezos invited Faricy to lunch with the S-team. A month later, he asked him to run what he described as one of the most underachieving teams at the company.
Faricy understood that there was only one answer to Bezos’s interview question—you couldn’t possibly reach out to sellers and recruit a million of them one by one. You would have to build a machine that would have to be self-service, and sellers would have to come to Amazon instead of the other way around.
Over the next few years, Faricy and his team rebuilt Seller Central, the website for third-party merchants, giving merchants the ability to easily list their products on Amazon.com, set prices, and run promotions—all with a minimum of oversight by Amazon employees. As he had with FBA, Bezos supervised the project closely at first. “I think in my first two weeks, I got seven question mark emails from Jeff,” Faricy recalled. “Right out of the gate, it was like throwing me into the fire and a once-in-a-lifetime learning experience.”
What helped Amazon to recruit third-party merchants was its rival eBay, which was alienating its unruly seller community by raising fees and giving favorable deals to large retailers. At the inaugural Amazon sellers conference in 2010 in the Marriott hotel near the Seattle-Tacoma International Airport, Faricy addressed “all of you in the audience who mostly sell on eBay.” He said that Amazon was committed to a fair marketplace and level playing field for all sellers and invited them to double down on their business with the company. They gave him a standing ovation. But that reservoir of goodwill wouldn’t last for long.
Like other Amazon execs, Faricy appeared to adopt a few of the severe management tactics Bezos had popularized in the company. Originally from Detroit, he had spent time at McKinsey & Company and the Borders book chain before Amazon, but he quickly acclimated to his new workplace. If deputies were late to deliver their weekly metrics reports, for example, Faricy would casually suggest they might be getting paid too much or weren’t right for the job. He also battled with his counterparts on the retail team, whose priority was a premium selection of merchandise to guarantee a good customer experience, versus the anything-goes anarchy that accompanied the seller platform, where anyone could sign up and start selling cheap, low-quality products.
The perennial debate inside Amazon was pitting the quality of products versus the quantity. The fights often had to be arbitrated by Faricy’s boss, senior vice president Sebastian Gunningham, or by Bezos himself. Both leaned heavily toward expanding the breadth of selection as fast as possible. “Jeff and Sebastian’s view was that all selection is good selection,” said Adrian Agostini, a longtime marketplace executive. “They wanted rules in place: don’t offend, don’t kill, don’t poison. Other than that, you take what you get and let customers decide.”
In the first few years of the 2010s, attentive entrepreneurs in Amazon’s primary markets in the U.S. and Europe recognized a lucrative new opportunity. They could develop a unique product, find a manufacturer, often in China, and sell it to millions of online shoppers. Faricy and his team considered sellers to be their customers and cultivated them with programs like Amazon Exclusives, to highlight unique products, and Amazon Lending, to help finance sellers’ growth, using the inventory they stored with FBA as collateral. They regularly convened focus groups, asking sellers to identify issues that needed to be fixed and new tools that should be built.
“Amazon at the time actually cared about helping brands thrive,” said Stephan Aarstol, whose firm sold a popular line of stand-up paddleboards on Amazon and was featured on an episode of Shark Tank. By 2015, Aarstol employed ten people in San Diego and was bringing in more than $4 million a year—one of countless entrepreneurs who minted small fortunes on Amazon’s flourishing platform. But his opinion about the marketplace would gradually evolve.
Bezos was delighted by the progress. That year, for the first time, the value of goods sold through marketplace surpassed sales from Amazon’s retail side. Best of all, since the business was largely self-service, revenues were growing much faster than headcount. “Finally, a business that is able to get some leverage after it becomes successful,” Bezos crowed at the team’s OP1 meeting that year, holding the six-page narrative to his chest. “I’m going to take this document home and sleep with it.”
Bezos then told Faricy that he no longer needed to individually review the marketplace during the annual OP1 session and only wanted to spend time on new initiatives like Amazon Lending. The CEO was devoting more of his attention to new products at the company while the details of Faricy’s business were getting so complex that he felt like he could provide only limited guidance anyway.
He told the FBA team the same thing. He would still continue to audit the businesses, adjudicating disputes and sending question mark emails when he learned of problems. But he no longer needed to be so intimately involved in the planning stages.
Which would put Bezos at som
e distance from the coming chaos.
* * *
As the Amazon Marketplace and Fulfillment by Amazon grew, the executives in charge were eyeing a potentially disruptive competitor. In 2010, Peter Szulczewski, a Polish-Canadian former Google employee, cofounded an online advertising startup called ContextLogic. When it didn’t gain traction, he pivoted the company toward e-commerce with an ingenious twist—a kind of geographic arbitrage. Most internet sellers sourced their products from manufacturers in China, shipped them in bulk to the West, and marked them up for rapid delivery to relatively affluent online urban shoppers. Why not instead allow merchants in China to sell inexpensive, unbranded goods directly to customers in the West who might not care if it takes several weeks for products to get to their doorstep?
Szulczewski renamed his company Wish.com and in late 2012 started hiring Chinese staff to recruit sellers and handle customer service. His timing was perfect. Alibaba had helped to spawn a vibrant community of resourceful Chinese internet merchants who were looking for new customers outside their home country. In fact, Alibaba had the same idea and was developing a cross-border commerce site called AliExpress, which was enjoying early traction in Mexico and Europe.
The Wish.com and AliExpress websites were crowded with products and difficult for novice online shoppers to navigate. But customers seemed to relish the treasure hunt–like pursuit of disposable fashion, such as $12 faux leather sneakers. In 2014, Wish raised $69 million from venture capitalists and was featured in the Wall Street Journal. After one conversation about the startup, Bezos looked at Sebastian Gunningham and said, “You’re on this, right?” Gunningham later noted that “Wish inspired us. They hit a nerve.”
Amazon’s strategy toward such disruptive startups is usually to develop a relationship with them and learn what it can—often by dangling the possibility of an acquisition. That year, Szulczewski and his cofounder Danny Zhang were invited to Seattle, where they spent the day talking to a dozen marketplace executives. They came away with the impression that the Amazon execs were skeptical of their business model.
But over the next two years, Wish continued to raise capital and grow. In 2016, Amazon reached out again, inviting Szulczewski to meet with Bezos. By then the Wish CEO was dubious of Amazon’s intentions and said he would only meet with Bezos one-on-one. When he later noticed other Amazon executives on the calendar invite, he said he canceled and never went.
By now Amazon was hurriedly adapting to a changing e-commerce landscape. The flood of Chinese sellers onto the internet represented a potential Cambrian explosion of new, low-priced selections. Such generically branded goods might not appeal to everyone, but they could draw younger or lower-income buyers to online shopping, and they later might graduate to more expensive products and even sign up for Amazon Prime.
Amazon had conspicuously failed to develop an online marketplace for Chinese sellers within China. This represented another opportunity to do business in the world’s most populous country. As part of a new initiative dubbed Marco Polo, after the thirteenth-century Italian explorer, Amazon hired teams in Beijing to sign up local sellers, translate Seller Central into Mandarin, and provide live customer support for merchants. To lower freight costs and streamline the process of shipping overseas, the company also developed an initiative called Dragon Boat. The service consolidated merchandise in coastal hubs like Shanghai and Shenzhen, moved them in bulk through customs, and then shipped them in containers that Amazon reserved at wholesale rates from shipping companies like Maersk.
New employees who joined the new global selling teams were pushed to move quickly—to get big fast. Internal documents, shared with me by a former employee, describe the team’s goals from that time, such as quickly growing headcount in China to recruit local sellers and teach them how to use FBA. Though AliExpress hadn’t gained much traction in the U.S., execs noted that Alibaba’s fees were lower than Amazon’s and worried that its hypercompetitive CEO, Jack Ma, might suspend the fees altogether to secure a foothold in Western countries. “Are we doing enough to capture the China-based seller opportunity?” asked one Amazon document. “Should we reduce the friction in our China-based seller onboarding by relaxing our listing standards to accelerate selection growth?”
While it’s not clear whether Amazon actually relaxed those standards, what’s evident is that they weren’t very high to begin with. Throughout 2015 and 2016, thousands of Chinese sellers registered on Amazon’s marketplace each day. “The numbers were astronomical. No one had seen volume like that,” said Sebastian Gunningham. Predictably, quality varied dramatically. “People would see a bestselling coat in the U.S. and literally it would appear on the site from a Chinese seller within hours,” Gunningham said. “Then a customer would pay for that coat and they’d leave a review about the sleeves falling off in the first minute.”
Gunningham, a member of the S-team, was in charge of global selling as well as FBA and the marketplace. He’d grown up on a ranch in Argentina, earned a mathematical sciences degree at Stanford University, and worked with Larry Ellison at Oracle and Steve Jobs at Apple before scoring a kind of high-tech hat trick by joining Jeff Bezos at Amazon. Colleagues said he was creative and empathetic—and a big thinker, in Amazon’s lexicon.
Gunningham quickly recognized that the flood of Chinese merchandise would stir controversy among sellers in the West who would not be able to match the low prices. One solution, at first, was to publicly and somewhat disingenuously downplay the shift. “The risky downside to this is that U.S.- and EU-based sellers do not find this avalanche of China-based sellers very amusing,” he wrote to fellow S-team members, in an email that was later submitted as testimony and made public in congressional antitrust hearings. “I have coached the team to be aggressive marketing in China to sell globally, but [to] take a low-key approach in the import countries.”
The low-cost merchandise was also divisive inside Amazon, so Gunningham devised symbolic ways of illustrating both the benefits and the challenges. One day, he started wearing a gaudy 80-cent stainless steel necklace with a dangling owl pendant. Amazon was selling tens of thousands of them per month, with Chinese sellers recouping a tiny profit margin on shipping charges. His point was that Amazon should not dismiss such low-priced items. “Everybody thought that lots of trash was coming onto the site, but trash is in the eye of the beholder,” Gunningham said. “Lots of it was very fashionable to many.”
Gunningham also bought dozens of black cocktail dresses of various sizes and styles on the marketplace and displayed them on a rack in his conference room. Amazon was selling thousands of such garments, no-name brands of variable quality made in China. Some were long, some short; some cost a few hundred dollars and others were as cheap as $20; some seemed durable, while the zippers on others seemed to spontaneously self-destruct on the first use. Colleagues recall that the rack of dresses sat there for months. His point was that his team needed to do a better job of distinguishing the various dresses from one another, giving customers the ability to evaluate them individually, so that the well-proven reviews system could penalize the low-quality sellers. The dresses “illustrated the broad set of dilemmas that mostly stemmed from the stuff coming out of China,” he observed.
Despite these demonstrations, the influx of Chinese products onto the marketplace remained contentious within Amazon and between the company and its partners. It was an accelerant, sprayed onto the already combustible frictions between sellers in the U.S. and China, and between Amazon’s first-party (or 1P) retail division and its third-party (or 3P) marketplace group. Quality was once again being pitted against quantity. Did Amazon want a calm, orderly store with only well-known and trustworthy brands? Or did it favor a more chaotic marketplace with a more extensive variety of products and prices?
Execs didn’t have to guess which customers preferred: their choice was clear in numerous trials and experiments. Amazon’s German website, for example, allowed third-party merchants to list and sell a wide variety
of branded and generic shoes, while Amazon’s UK site featured a curated shoe store with only more expensive, brand-name footwear. The German site performed markedly better, because of the greater selection and cheaper options.
This finding was significant, because Amazon’s corporate compass only pointed one way: toward what customers wanted. And it turned out that plenty of people will buy dirt cheap sneakers on the internet, even if they suspect the shoes are not destined to last that long.
Nevertheless, Amazon’s retail execs continued to object to the inundation of low-quality Chinese merchandise, and the debate regularly made its way to the S-team. In one meeting, Jeff Bezos was asked to resolve a version of it: What should Amazon’s broader strategy be in apparel? Should the company prioritize the sale of high-end clothing, usually from premium Western brands, on carefully curated, dedicated websites? Or should it favor low-end generic apparel and private-label products across Amazon.com and via the Amazon Marketplace?
The room went silent while everyone waited for Bezos’s decisive answer.
“I think we should target everybody who wears clothes; I haven’t seen that many people naked over the last few days,” he finally said, laughing uproariously. “I believe people are going to be wearing clothes for a long time.”
* * *
It was one of those questions that Bezos believed should not be answered. He wanted Amazon to do it all. But by not answering, he was also casting a vote for uninhibited, low-priced selection on the marketplace—one that would have enormous ramifications.
Chinese startups, some quite formidable, arose almost out of thin air to sell on Amazon.com. In 2011, a software engineer named Steven Yang had quit his plum Silicon Valley job at Google and moved to Shenzhen to start an electronics company called Anker to sell accessories like replacement laptop batteries. Over the next few years, his product line expanded to include just about every kind of cable, charger, and battery imaginable, many of which soared to the coveted top spot of Amazon’s bestseller lists.