In early 2017, a memo circulated inside Amazon that imagined an ambitious new grocery chain. The document was written like a news release, a common practice for ideas being weighed inside the company, with the title “Grocery Shopping for Everyone.”
The new stores, the document envisioned, would have robust sections for produce, fresh food and prepared meals. Nonperishable products, like paper towels or canned beans, would be stored on a separate floor, away from customers. Shoppers could order those items with an app, and while they shopped for fresh food, the other products would be brought down in time for check out. There would also be an area to pick up groceries ordered online and to manage packages for delivery drivers.
The faux news release, which has not previously been reported, cited a fictional grocery expert named Hal Apenyo, as in the chili pepper, declaring success in just six months. “The conversion from offline grocery shopping to mixed format shopping has been massive,” the character was quoted as saying.
A few months later, in June 2017, Amazon barged into the grocery business in a different way, by announcing a blockbuster deal to buy Whole Foods for $13.4 billion. The purchase catapulted Amazon near the top of the $700 billion grocery industry, and sank stocks of traditional grocers on fears that they would be outmaneuvered into oblivion. The memo and other big grocery proposals stopped circulating inside Amazon, as Whole Foods demanded everyone’s attention.
But two years later, instead of Whole Foods being the answer to Amazon’s grocery ambitions, it seems to have only whetted executives’ appetites.
The marriage has made clear the difficulties of selling fresh food inexpensively, either in a physical store or through delivery. Bananas are not the same as books.
But the combination has also shown glimmers of success, particularly in delivery. And that has provided some fuel to Amazon executives pushing to add another food-selling option — one built from the ground up that would change how people buy groceries.
The company is now quietly exploring an ambitious new chain, probably separate from Whole Foods, that is not far removed from the one outlined in the old memo. It would be built for in-store shopping as well as pickup and delivery. As the discussions heated up this year, employees passed around a slightly updated version of the memo.
The details of Amazon’s challenges and ambitions in the grocery business are based on interviews with more than 15 people who have worked at or with the company. Most spoke on the condition of anonymity because they have nondisclosure agreements or were not authorized to speak publicly.
“People really need to understand — Whole Foods is the beginning, it’s not the end,” said Brittain Ladd, who worked on Amazon’s grocery operations until 2017. “It’s not everything.”
An Amazon spokeswoman, Rachel Hass, said the company “doesn’t comment on rumors or speculation.”
Before it bought Whole Foods, Amazon was an afterthought as a grocer, well behind chains like Publix and ShopRite. The food it sold was limited to mostly canned and dry goods, and its decade-long effort to sell perishables through a pickup and delivery program called AmazonFresh never caught on.
Whole Foods had struggles of its own. The company was fending off activist investors and had stopped expanding. While its base remained loyal, grocers like Kroger and Walmart had started selling many of the products that once set Whole Foods apart, like organic kale or kombucha.
“Whole Foods was broken — we shouldn’t forget that, which is why they could buy it,” said Phil Lempert, a food marketing analyst.
It was clear from the start that the two companies differed culturally. John Mackey, Whole Foods’ co-founder and longtime chief executive, had written a best seller about how companies should have a social conscience and consider all stakeholders in their decisions. Amazon corporate principles say good leaders “do not compromise for the sake of social cohesion.” But Amazon pushed ahead with some changes that were once held up as points of pride for the grocer.
In an effort to shed Whole Foods’ “whole paycheck” reputation, Amazon bought more from national food distributors and cut back on the local farms. United Natural Foods, a leading organic distributor, has increased its sales to Whole Foods by 38 percent over the past two years. And inside stores, employees stopped making signs on chalkboards by hand. Now, Whole Foods prints signs with black ink on paper in a font that resembles handwriting but requires less labor.
Other price-cutting efforts failed. The former head of a major produce company said Amazon told him it wanted to sell marquee fresh items at low prices every day. The executive said he had to explain that certain products, like berries or lettuce, may be available all year thanks to global supply chains, but that they cost more in the off-season. Forcing flat, low prices would put too much risk on growers.
Amazon executives, the person said, were caught off guard by the response. It didn’t seem as if they had fully appreciated how seasonality made predictable pricing far harder than selling cereal or paper towels.
The mixed results are reflected in prices at Whole Foods today. A standard basket of goods has fallen about 2.5 percent since the acquisition, according to Gordon Haskett Research Advisors. Amazon has said its Prime members, who get charged $119 for an annual subscription, have saved hundreds of millions of dollars in discounts at Whole Foods. But over all, Whole Foods is still more expensive than other major grocers, particularly for items like meat.
Amazon has also run into some trouble integrating Whole Foods into its delivery machine.
Amazon never saw delivering cold milk and fragile fruit to doorsteps as something for the masses, according to former employees. Instead, executives thought of it as an option for people who wanted high-quality foods and could afford a premium price to have fragile and fresh items arrive at their doorstep.
In theory, that was a good fit for Whole Foods and its affluent shoppers. Within six months, Amazon began making two-hour deliveries from Whole Foods in four cities for Prime members. Six months later, that had expanded to more than two dozen cities. It’s now available in 90.
But Whole Foods stores are not like Amazon’s delivery warehouses. Because Whole Foods sells so many fresh items, its stores have smaller back-of-house areas than a standard supermarket. That means employees who pick products for online orders must gather more items from the same shelves as shoppers. They roam aisles with scanners in hand, asking associates on the floor when they can’t find something.
In addition, items in grocery stores are grouped together. Walk into a Whole Foods, and a picker for an online order might be standing there trying to see if the identical tubs of Parmesan she’s grabbed are grated or shaved. In a warehouse, similar items are kept far apart to avoid confusion.
Still, deliveries have shown big potential, making up almost all of Whole Foods’ growth.
The promise of serving customers, but doing so more efficiently, has Amazon thinking again about aggressive investment in groceries.
Rather than dramatically substantially expand Whole Foods, several former employees said, Amazon is considering designing stores specifically with pickup and delivery in mind, and with a smaller area dedicated to fresh shopping — as the old memo imagined.
While it is unclear what hybrid design Amazon has in the works, a recent job posting for a store designer on “an exciting new team” was looking for someone interested in “creating multiple customer experiences under one roof.”
And Amazon has been looking for spaces close to Whole Foods locations, indicating a hub-and-spoke approach where one store serves as the warehouse and commissary for others. Experts say it could take more than a decade to build a new chain from the ground up.
To be a major grocery player, Amazon would need a little more than 2,000 stores, the old memo estimated. That’s far fewer than the 5,000 run by Walmart, the country’s top grocery seller, but more than the roughly 1,200 operated by Publix. Whole Foods got Amazon about a quarter of the way there.
A store designed with different shopping options, “Mr. Apenyo” predicted in the old memo, would be “highly scalable.”