by Brad Stone
The move piqued the interest of hedge fund Jana Partners, a so-called “activist investor,” whose managing partner, Barry Rosenstein, believed Whole Foods was “lost and broken.” Jana bought stock in distressed companies, agitated for change, and usually minted money when it forced a firm to slash costs or found an acquirer to pay a premium for it.
Quietly amassing Whole Foods stock that winter, Jana revealed itself as the company’s second largest shareholder in April 2017. It demanded changes on the management team and board of directors; Rosenstein later said the firm was prepared to take over the company and “fix it ourselves.” But Whole Foods executives worried that Jana’s plan was to merge the organic grocer with another food giant it had a stake in: Albertsons Companies, an amalgamation of traditional supermarket chains like Safeway and Vons. In that scenario, the conglomerate would take Whole Foods’ esteemed brand name and relatively unlevered balance sheet; it would also likely ship out the intractable John Mackey and ship in Coke, Doritos, and other crowd-pleasing fare.
Dismayed, Mackey and his executive team scrambled a defense. They recruited five new independent directors to replace longtime board members who had an average tenure of more than fifteen years. They also sought a white knight, contacting private equity firms as well as billionaire Warren Buffett, according to a former board member. But since earnings were flat and the grocer wasn’t generating enough cash to borrow money, the math on a leveraged buyout didn’t work.
One option remained, which almost everyone at Whole Foods Market considered fanciful. Over the years, they had engaged in several fruitless conversations with Amazon. John Mackey was an admirer though, and the year before had vividly dreamed his grocery chain had been acquired by the e-commerce giant. (“That’s crazy,” his wife, Deborah, had told him.) When Bloomberg News reported that Amazon executives had recently discussed acquiring Whole Foods, Mackey asked one of his advisors to place a phone call and try one last time to save the company.
* * *
Jeff Bezos placed prospective business opportunities into one of two buckets. There were land rushes, when the moment was ripe, rivals were circling, and Amazon had to move quickly or else it would lose out. Then there was everything else, when the company could bide its time and patiently experiment.
Amazon’s attempts at a third-party marketplace and with the Kindle and Alexa were land rushes. Bezos pushed his employees to move with urgency and they had the battle scars to prove it. But for years he took a more passive approach to home delivery of food—that is, until he saw formidable competition emerge and abruptly changed his mind. The resulting shift in strategy would have significant consequences for the massive grocery market and the way customers, competitors, and regulators viewed the e-commerce juggernaut forever after.
The Amazon executive who brooded over the grocery business the longest and who agitated for a more aggressive approach was Doug Herrington, the senior vice president of Amazon’s consumables unit. Herrington often wore plaid shirts and a Patagonia vest in the office and talked in such a low voice during meetings that employees often had to lean forward to hear him. Earlier in his career, he had worked at the first-generation grocery delivery flameout Webvan, which raised close to a billion dollars in private and public financing during the dot-com boom and then went out of business in 2001.
Internet historians would view Webvan as the ultimate symbol of Silicon Valley’s arrogant rush to create a future that people didn’t want. According to Herrington, a Princeton University and Harvard Business School alumnus who ran product development and marketing at Webvan, the real story was more complex. CEO Louis Borders—cofounder of the eponymous book chain—and his team erred by building a network of warehouses that were so costly to operate that the company lost money on every order. Before they could rectify that mistake or even open for business in many cities where they had set up operations, Wall Street stopped funding unprofitable startups during the early 2000s recession. The company’s sales and customer base were growing, but it couldn’t withdraw from its financial commitments fast enough and declared bankruptcy. “I walked away saying, ‘theoretically this model can work,’ ” Herrington said. “We made the wrong choices, we did some inefficient things, but customers loved it.”
After joining Amazon in 2005 to run consumables—goods that are used up relatively quickly, like laundry detergent and food—Herrington assembled a team of employees to meet at night on an ambitious project, which would be outside the bounds of their daily responsibilities. The mission was to develop a plan for Amazon to launch a national grocery service. He wanted to solve the problem of home food delivery and finally vanquish the acrid taste of the Webvan bankruptcy. A year or so after, in late 2006, back when Amazon was in the old Pacific Medical Center, Jeff Bezos reviewed their plan, which called for a $60 million up-front investment, and rejected it. “The feedback was ‘love the vision, hate the numbers,’ ” Herrington recalled. Instead, he got $7 million to open a limited service beta test in Seattle. Then-CFO Tom Szkutak asked him to try not to let it distract the rest of the company.
Amazon Fresh launched in August 2007. In Bellevue, east of Seattle, Herrington’s team leased an old Safeway distribution center, which was sitting fallow amid the collapse of the local real estate market and was “horror movie scary,” according to Ian Clarkson, Fresh’s first general manager. Everything about the new service had to be different than other parts of Amazon: rooms in the warehouse had old walk-in refrigerators, while the website displayed multiple products on each page instead of just one and gave customers the option of choosing a specific window of time in the day for delivery. Bezos frequently reviewed Fresh’s progress and at one point, scrutinizing the ballooning delivery costs, suggested Amazon offer predawn drop-offs. Like garbage collection trucks, they could take advantage of the dearth of traffic in the early morning.
Seattle customers appreciated waking up to groceries on their doorstep. But Fresh’s other challenges were more formidable. Unlike other Amazon services, which instantly had national or even global reach, Fresh’s potential was bounded by the zip codes where its fleet of drivers could make deliveries. The Fresh team also had to solve a nest of thorny problems, like what to do with expired food, how to manage the complex banana ripening process, and how to respond when customers complained they had found something unseemly in their dinner. Still, over the course of six years, they made slow but steady progress that crept toward profitability.
During that time, Herrington regularly pitched plans to the S-team to replicate Fresh in other cities. But Amazon was smaller back then, and the land-rush opportunities, such as the doomed expansion in China and the Fire Phone, took precedence. Consumer adoption of home grocery delivery, Bezos believed, was going to be a more gradual process. Forced to stay local, Herrington was frustrated by the perpetual delay.
Then in April 2012, Bezos convened the S-team at Willows Lodge in Woodinville, Washington, about a half hour northeast of Seattle, for their annual off-site retreat. Each executive was asked to bring a one- to two-page memo that contemplated a significant new opportunity for Amazon. Herrington had joined the vaunted leadership council a year before, and his blunt memo would resonate inside the S-team for years. Even its title was provocative: “Amazon’s Future is CRaP.”
In company parlance, CRaP stood for “can’t realize a profit” and had several meanings. CRaP included items like stepladders and whiteboards, which couldn’t be put in boxes or efficiently shipped to customers. But in his memo, Herrington was talking largely about the inexpensive, bulky items stocked by supermarkets, such as bottled water, Diet Coke, or even a bag of apples. In the wake of the Webvan fiasco, most online retailers at the time considered these types of products to be economic quicksand. To the extent it sold them at all, Amazon had developed an “add-on” program to minimize their harmful financial impact. Customers could only include CRaP in their orders when they were making a broader assortment of purchases, such as books or electro
nics at the same time.
Herrington’s memo pointed out that Walmart, Carrefour, Tesco, Metro AG, and Kroger were the world’s five largest retailers at the time. “All of them anchor their customer relationship in groceries,” he wrote. If Amazon’s retail business was going to grow to $400 billion in gross merchandise sales, it needed to transform a model based on infrequent shopping for relatively high-priced goods to more regular shopping for low-priced essentials. In other words, if the company was going to join the ranks of the biggest retailers, the S-team had to figure out a way to profitably sell supermarket items. If they didn’t, Amazon was going to be vulnerable to rivals who already enjoyed the shopping frequency and cost advantages of the grocery model.
He concluded the memo by subtly needling his colleagues, including Bezos, who considered himself implacably bold. “We should be less timid in investing in this future,” Herrington wrote. “We have the capacity to put a much more significant bet on the table, if we have the will.”
Bezos typically responded well to this type of critical introspection, especially when it was coupled with a proposal for aggressive expansion. Such thinking reflected his own mindset. After the S-team sat in silence for several hours reading one another’s papers, the CEO picked up Herrington’s and said, “This one really made me think.” A few months later, Herrington got a green light for a limited expansion of Fresh into Los Angeles and San Francisco.
He had won the battle. The problem was that he hadn’t yet figured out how to wage the war. The introduction of Amazon’s grocery service into California in June 2013 was initially greeted with fanfare by the press. But it didn’t quite work, at least not at the scale that Herrington had hoped. To defray shipping costs, Amazon charged customers a hefty $299 annual Fresh subscription fee. To fashion a new supply chain for perishable groceries, it created chilled rooms inside existing Amazon fulfillment centers an hour or so to the east of each city, in San Bernardino and Tracy, California. Then Amazon routed trailer trucks twice a day into staging areas in each region, where orders were moved to bright green Fresh vans for the last-mile delivery to customers’ homes.
Logistics employees who worked on the California service said this hub-and-spoke model ended up being inefficient and unreliable; one said that Amazon was “basically stapling a $10 or $20 bill to every order.” The Fresh team also tracked a metric called “perfect deliveries”—when an order was promptly delivered and included every item. They found they were hitting that target less than 70 percent of the time.
Grocery industry veterans belittled the effort from afar. “Amazon Fresh is their Waterloo,” John Mackey told me during our chat in 2014. “What’s the one thing people want? Convenience. You can’t do that with distribution centers and trucks.”
Aside from its quiet introduction into parts of Brooklyn a year later, the expansion of Fresh into new markets slowed down considerably after the California expansion. Success in delivering online groceries relied on getting the logistics exactly right and amassing enough demand to make it profitable to send drivers into residential neighborhoods. Amazon had set up warehouses too far from customers, made it too expensive for them to sign up, and saddled them with bulky tote bags and sacks of dry ice after each delivery. Bezos had finally agreed with Doug Herrington that Amazon needed to reinvent its retail business, but they were going to have to find a different way to do it.
* * *
Then, as so often happened in Amazon’s history, the arrival of competitors onto a shifting landscape injected some resolve into the calculations of Bezos and the S-team. The race for online ordering and rapid delivery of groceries was about to turn into the one thing that reliably captured their attention and investment dollars: a land rush.
Two rivals introduced same-day delivery services, using similar business models. The San Francisco startup Instacart was founded by Apoorva Mehta, a former Level 5 senior engineer in Amazon’s logistics division (in other words, a relatively low-level worker on a hierarchy that stretched from Level 1 warehouse recruits to Bezos at Level 12).
Instacart raised millions from venture capital firms including Sequoia, the original backer of Webvan, and struck partnerships with grocery chains like Whole Foods Market, Costco, and Safeway. It then contracted with smartphone-toting pickers to select orders off the shelves of retail stores and with drivers to transport them in their own cars to customers’ homes. There was no inventory risk or costly employment contracts, since the workers were all independent contractors. With few of the fixed costs that sank Webvan, the startup had tremendous leverage.
After Instacart burst onto the scene in 2012 and started furiously expanding into new cities, Amazon’s M&A team tried to reach out to learn more about the company. Wise to Amazon’s methods, Apoorva Mehta didn’t return the call.
The second challenger seemed even more dangerous at the time. Amazon’s archrival, Google, introduced a service called Google Shopping Express—later Google Express—offering customers unlimited same-day delivery from retailers like Costco, Target, and Smart & Final for a $95-a-year annual subscription. In 2014, the service expanded to Chicago, Boston, Washington, D.C., and soon after would land in Bezos’s backyard in Seattle. In case there was any confusion about the search giant’s intentions, Google chairman Eric Schmidt cleared it up at a speech that fall in Berlin. “Many people think our main competition is Bing or Yahoo,” he said. “Really, our biggest search competitor is Amazon. People don’t think of Amazon as search, but if you are looking for something to buy, you are more often than not looking for it on Amazon.”
Amazon execs and employees expressed varying opinions about which rival posed a bigger threat. Jeff Wilke acknowledged that Google Express, which offered a broad assortment of merchandise from both regular stores and supermarkets, was “demonstrating that customers preferred a faster delivery option.” But both were dangerous. In the past, Amazon simply acquired competitors like Zappos and Quidsi (owner of Diapers.com) that offered superior selection and delivery speed in a focused product category. With Instacart and Google, it faced a pair of challengers that couldn’t be bought and weren’t going anywhere.
With Bezos spending his time on newer initiatives like Alexa, he was leaving the daily maneuverings in Amazon’s consumer business to his retail chief. That September, in the midst of the Google Express expansion, Wilke was conducting a quarterly business review with the Amazon Prime team and asked his top lieutenants to propose a response to the threat. His deputies pitched expanding the selection of items that were available to Prime members for same-day delivery at an extra charge. Wilke didn’t think that would be enough to match the new offerings, rejected the idea, and, as he put it later, “blew up the meeting.”
Wilke announced that he wanted to attack this problem from a totally different angle. They were going to form an independent team to build a service that was separate from the Amazon website and singularly devoted to ultra-fast delivery. The goal, he declared, was to launch it within a hundred days. Dave Clark, Amazon’s head of operations, would oversee the effort in conjunction with Herrington.
Sitting next to Wilke in the meeting and taking notes was his technical advisor, a ten-year Amazon veteran named Stephenie Landry. While she was dutifully typing, a chat window popped up on her screen, from Clark, who was sitting across the room. Did she want to spearhead the new project? In fact, she did.
Landry was a fast-rising star who exhibited some of the management qualities—detail-oriented, dogged determination, a merciless driver of underlings—of the Bezosian leadership template. She was originally from New York City and had attended Wellesley College, where she majored in women’s studies. After graduation, she received a grant from the school to spend a year building, of all things, wooden boats. After that, she joined a struggling internet company during the dot-com bust, earned an MBA from the University of Michigan, and then joined Amazon in 2003 in operations, working on how to transform its warehouses to accommodate products other than books, DV
Ds, and small consumer electronics. In her first employee badge photo, she sported a quasi Mohawk.
Landry joined the original Fresh team in Seattle and then managed another of Herrington’s beleaguered efforts to crack the CRaP problem, Prime Pantry. That service, which was perpetually unprofitable, allowed customers to load up heavy boxes of household staples like cereals, pasta, and bottled water at steep discounts.
In her new job, Landry’s first task was writing the PR FAQ, the preliminary press release that conjured the kind of service that Wilke wanted. The paper and its subsequent revisions described a smartphone app-based service that Landry first dubbed Amazon Magic, and then Amazon ASAP. She proposed forming three separate teams, each with a magic-themed name, which would all take different approaches toward the same goal of ultra-fast delivery.
One group would develop a retail service, code-named Houdini, which would store and sell a limited selection of the most popular products on Amazon from strategically located urban warehouses. That would allow Amazon to deliver frequently purchased items to customers within a few hours.
Another group would take a third-party marketplace approach, dubbed Copperfield. That team would seek to form partnerships with retail and grocery stores and list the products they had on their shelves on a new Amazon smartphone app, just like Google Express and Instacart.
Finally, a third group was formed to pursue an idea called Presto. It called for assembling an even smaller selection of top-selling products and driving them around in a truck or van to deliver items in less than ten minutes to surrounding neighborhoods. This approach proved complex and risked overlapping with the others, causing confusion, so it was quickly shelved.
