Amazon shipped more than 5 billion items worldwide in 2017, up from 1 billion items shipped during the 2016 holiday season, the company has said. And while the number of Prime members is still a closely guarded secret, tens of millions of households are likely program participants.
Shipping those billions of items to its many million members is among Amazon's biggest problems. To solve this, the company is reportedly readying the launch of its own shipping service, which will begin with limited pickups and deliveries in Los Angeles, but may eventually roll out to other markets. It will go a long way toward helping Amazon solve its logistics problem, improve profitability and send it miles ahead of its competition, literally.
Getting the last mile right is an imperative for all retailers, big and small.
"The whole industry has come down to an Amazon versus Walmart battle. Walmart has an infrastructure and supply chain that gives them an edge. And Walmart has store locations within 10 miles of 90% of the U.S. population."
Executive Vice President of Cloud Logistics
Even Walmart, Amazon's biggest competitor, struggled with order fulfillment during the holiday 2017 season, even though Walmart's existing warehouses allow it to compete in a way most retailers cannot.
"The whole industry has come down to an Amazon versus Walmart battle," David Landau, executive vice president of transportation management systems provider Cloud Logistics, told Retail Dive. "Walmart has an infrastructure and supply chain that gives them an edge. And Walmart has store locations within 10 miles of 90% of the U.S. population."
As Amazon begins ramping up its own proprietary delivery fleet, it's not just retailers facing competition from the e-commerce behemoth — FedEx and UPS are also under pressure. Shares of both tanked after Amazon's announcement that it planned to launch a delivery service.
But it will take a long time before Amazon is able to disassociate from existing delivery companies. At the moment, UPS, which shipped 4.9 billion packages and documents in 2016 and FedEx, which said it expected to ship between 380 million and 400 million packages during the 2017 holiday season, are still safely in the shipping lead, with DHL also picking up a fair amount of share. Yet as online sales increase, the shipping industry is looking for new ways to deliver.
The long haul for the last mile
"The world is changing so fast," Burt White, vice president at supply chain management consulting firm Chainalytics, told Retail Dive. But not all parts of the equation are equally complex. "The last mile is the hardest to crack," he said. While long haul companies such as UPS can efficiently move product from warehouse to warehouse, door-to-door delivery is more of a challenge.
"The legacy shipping companies are extremely good at moving packages a significant distance," Nicholas Hodson, principal at global strategy consultancy Strategy& (part of the PwC network), told Retail Dive. "And the thousand-mile leg that the package travels in trucks is incredibly cheap. But they're not good at moving packages 15 miles, because they were never designed to do that. The pieces at either end of the long haul are rather expensive, and the cost per mile on short deliveries is very, very high. And it's a problem to solve that."
Yet according to shipping industry specialists, this is where retailers are most likely to see significant change. "Final mile is becoming so hot right now because we’re trying to keep long hauls along a main line," Terese Kerrigan, director of marketing communications for FreightCenter, a third party shipping provider that handles logistics, told Retail Dive. "You're going to see more last-mile moving in smaller vehicles."
Some legacy companies are actively trying to engineer change. FedEx sponsors a logistics accelerator in Memphis, TN that emphasizes first mile and last mile solutions. And several shipping lines and cargo companies, plus Ingram Micro, sponsor another accelerator seeking technology-based solutions.
"Some of the big players such as UPS and FedEx are in some ways hoping disruption occurs," Sean Monahan, lead partner in the operations and performance transformation practice of A.T. Kearney, told Retail Dive. "UPS and FedEx infrastructure is designed for B2B and large hub, but not for localized deliveries, like showing up at doors every day. That's why they've already collaborated with services like Deliv, UberRUSH, and even Task Rabbit for deliveries."
In the short term, last-mile is the nut to crack, Chainalytics' White said. "They'll do it with crowdsourcing or Uber. And maybe we just dummy down our expectations, and we have drop boxes in our yards or something," he said. "But longer term, we're going to have more individual customization, and that's going to effect inventory and warehousing."
For example, when a company moves from a small number of SKUs to 700, the entire dynamic of shipping changes, White said. "The next phase is going to be mass customization, and then you're not even going to need big warehouses. Our client SKUs are going through the roof. They need so much stuff now."
The answer lies in tech
Any solutions to last-mile delivery problems need to solve issues of speed, security and most importantly, surety, according to A.T. Kearney's Monahan.
Surety is where some of the more interesting technology ideas, such as Latch and Amazon Key, have begun to emerge. "Amazon Key is trying to solve the surety problem by delivering into your home," Monahan said. "Lockers solve that too, and also allow Amazon or their shipper to streamline their operations and not have to worry about operators coming back to redeliver."
Drones, robotics and other automated tech are also being tested. "We're seeing huge advances in robotics in warehouses," David Landau, executive vice president of Cloud Logistics, a transportation management systems provider, told Retail Dive. "Disruptions are happening every day, and because they're happening so frequently, we're almost losing sight of the fact that they're happening. So now imagine a warehouse with robots flying around. We don't need pickers anymore. What if those robots load orders onto self-driving trucks with drones that get released into a neighborhood? A person-free delivery system, [which] operates without labor and people."
"In China, it's mostly all robotics, and that's partly out of need because most people don't want those jobs now anymore. The new warehouses that are being built are more automated."
Vice President, Chainalytics
Kerrigan, whose company FreightCenter is developing a point-and-click shipping app set to launch in 2019, agrees that self-driving vehicles are on the horizon, but notes some problems haven't yet been resolved. "I see some drones, and the Tesla semi is popular," she said. "And DHL just bought a fleet of 20 self-driving trucks. But more solutions mean more problems, whether that's danger from self-driving trucks or self-driving drones."
Certainly mass automation is already happening in overseas warehouses. "In the U.S., we have trouble implementing robotics [for shipping] when product sizes are different, like mattresses versus iPhones," White said. "But in China, it's mostly all robotics, and that's partly out of need because most people don't want those jobs now anymore. The new warehouses that are being built are more automated."
Yet Monahan said it's unlikely that most of these warehouse solutions will work for last-mile. "Think about swarms of drones in Manhattan, or even in remote places where they might be used for target practice," he said. "People sometimes fail to grasp how many drones it takes." He said that while autonomous vehicles could be viable in 25 years, urban environments make even trucks may be problematic.
"When will [New York City Mayor Bill] DeBlasio say, 'Enough is enough. I'm tired of all these trucks clogging the streets. It's detracting from the efficiency of the city.'" Monahan said, adding that if e-commerce sales rise to 50% of total retail sales, the industry may have to address other constraints on same- and next-day delivery.
The ship-from-store dilemma
In addition to exploring innovative ideas in tech, retailers are also looking at new ways to use existing personnel. In particular, ship-from-store has become more intriguing to many companies over the past ten years. However, it comes with its own problems.
"Typical in-store inventory accuracy is 60%," Cloud Logistics' Landau said. "And if it's not in the store, then you may or may not be able to same-day ship. Even if it's in store, you may not know where it is in the store. And then on the transportation side, in terms of doing this as local delivery, it's challenging and expensive and fraught with risk. It's subject to whims of traffic and weather. And so most companies depend on local couriers in local markets and the you have to manage all those local relationships."
One solution to the inventory issue might come from using a technique familiar to online retailers: demand shaping, where a company massages a consumer's demand for a product by offering different incentives. "When you buy on Amazon, you see you can get free delivery, but if you wait five days you can get $5 in credits for something," Landau said. "And there's a pretty good chance you'll never use that credit, but Amazon is weeding out people who may not need two-day shipping. Demand shaping allows customers the illusion of control, or even actual control."
This kind of consumer manipulation might allow stores to offer customers what they actually have in stock. "Most consumers don't know what they want," Monahan said. "If you look at consumer expectations, they lag Amazon's promise by about six months. Amazon is shaping the expectation. Amazon lays out this whole menu of possibilities, but they use things like discounts to back you away from that expectation."
Walmart is already trying a version of this out. "Walmart is trying to tell you what's available on the shelf, in store, in your neighborhood," Monahan said. "You can pick it up yourself, or associates who live in your neighborhood can deliver it on their way home" — a program Walmart began testing last year.
However, the idea isn't entirely cost-effective for any retailer. "The math for ship-from-store is higher because there's more labor than in a large facility," Hodson said. "The store isn't designed for that, and consumers have a nasty habit of moving things around the store and stopping the picker to ask questions. However, the distance is less, so a retailer might think that the savings are more. It turns out you do save shipping costs, but not a great deal."
Looking for its Uber
Instead of ship-from-store, crowdsourcing might offer a solution that is both cost-effective and locally driven, according to Strategy&'s Hodson. But it only works in certain areas.
"In Dallas, Amazon has two fulfillment centers, out by the airport, doing 50 million units per year," Hodson said. "Some portion of that volume is local. So by lunch, enough orders have built up [going] to Plano, Texas, which is seven miles away and the Amazon Flex app has offered up a bid asking for a driver to go to Plano. There's a pile of boxes and you have an $18 per hour person."
So in some cases, Amazon may make a same-day delivery, not because the customer wanted it, but because it was more cost-effective. "That's freaking disruptive," Hodson said.
But Monahan is not certain that crowdsourcing is currently capable of solving the surety or security problems. "Crowdsourcing or sharing can work," Monahan said. "But what someone has to solve is standards about how packages are delivered. There needs to be a background test. There needs some kind of certification."
Of course, the best version crowd-sourced delivery likely isn't here yet. "The procedures and the technology for crowdsourcing may not be there now, but they may develop," White said, who believes a future version of crowd-sourced delivery might even become a social event.
Still, carrying one package at a time isn't productive. "The Delivs of the world are quite good at this short distance, but they're only carrying one or two or three packages," Hodson said. "They use a crowd-sourced resource to drive a vehicle like Uber, which drives down the per hour and per mile cost a lot. But the productivity is horrible. It's point to point. That person is driving empty to Best Buy and has one package going to your house. So that move is more expensive. It's a market, but it's a niche. If you really freaking need it now and you will not wait and you're willing to pay for it, then okay. But as soon as you ask people to pay for shipping, the number of people who want it right now goes down a lot."
"Rather than commit to purchasing a fleet of trucks that must be maintained, retailers are Uberizing their fleet by hiring local couriers to handle final mile and same day deliveries."
Director of Marketing Communications, FreightCenter
This means the model simply isn't efficient if it scales larger. "If you're Deliv and looking at economics, you realize that your drop density sucks because you're doing point to point," Hodson said. "You can bet they're trying to figure out how to get that up."
Even Google has tried to disrupt the delivery puzzle with Google Express, with limited results, according to Landau. "Other companies such as Deliv are trying to become the Uber of same-day delivery," he said. "But they are essentially using Uber drivers to do that."
So what might real disruption look like in last-mile, same-day delivery? Until it happens successfully on a scalable level, there are as many ideas as there are challenges. "Is it ride sharing?" White asked. "Group deliveries? Is Amazon becoming the next FedEx? If a company can get the demand consolidated, it can start making money. It can't make any money if it can only drop off one or two packages at a time."
Crowdsourcing could even take over some of the broker business. "Freight brokers match loads with truckers," White said. "Uber might take that over in time."
Some of that is already happening, Kerrigan said. "Blockchain technology in transportation will make the market more competitive, allowing for smaller carriers and independent truckers to bid on shipments," she said, something that is already taking place at a third-party logistics level. But looking past that, she sees smaller shippers coming into play. "Expanding our carrier network to include Uberized service options allows shippers to find same-day service on demand," she said. "Carriers are adopting the same philosophy, and leveraging smaller asset-based companies or independent drivers to transport any surplus their fleet cannot handle." Kerrigan is already seeing a surge in for-hire couriers.
"Rather than commit to purchasing a fleet of trucks that must be maintained, retailers are Uberizing their fleet by hiring local couriers to handle final mile and same day deliveries," Kerrigan said. "As consumers, we’re already seeing Amazon and Walmart resort to hiring local couriers to meet delivery deadlines when package volumes go beyond what their supply chain can handle."
Even as legacy shipping companies race with Google, Amazon, Walmart and new entrepreneurial ventures to solve some of the biggest delivery conundrums, it's hard for experts to predict what’s going to work in the long run.
"We're still waiting to see who has the best idea," Monahan said. "There will be a lot of regional players, and then someone will go ahead and buy the winners in each market. They'll find a way to get together with these localized solutions, but until someone integrates them you'll always have some variance of quality."
Then there's the possibility that the real disruption will be not in how we ship, but in whether we ship at all. "At some point we might end up using less over time, and we might become more experiential," White said. "Or you could scan something and print it and make a 3D suit. You have to create a supply chain that's adaptable to what's coming."
It's also possible that the smartest idea hasn’t even been considered yet. "We're at the beginning of this, and usually at the beginning, the beginning players don't last," White said. "Look at all the industries. There's a disruptor that's going to come in and invent or reinvent some of this, but you'll still have to have the right product at the right time and get it to the right person at the right time."