Having worked in the transportation and logistics industry for nearly 30 years, the question that friends, family, and colleagues ask most often these days is some variation of, “Is Amazon going to start competing head-to-head with other parcel services?” Now I can just refer them here for my answer.
Before beginning, I have to emphasize that these are my own observations and opinions, based solely upon publicly available sources and industry experience. Sorry, no inside or confidential information. Furthermore:
Crystal balls are notoriously cloudy.
Odds are I’ve got it wrong. So don’t make investing (or any other kinds of) decisions based on what you read here. OK? Let’s dive in.
For more than fifty years, the leaders in the parcel delivery business in the United States have been FedEx, UPS, the Postal Service, and DHL, and before consolidation RPS and Airborne. Conventional wisdom held that this list would not grow because the capital barriers to entry were too high. Newcomers would have to build out a package movement infrastructure with hubs, ramps, sort facilities, and stations. To move the packages they would need trucks, vans, and airplanes. And lots and lots of people. And billions and billions of dollars.
Amazon has been turning that conventional wisdom on its head.
Mail order shopping has been around for more than a century, pioneered by Sears in the early 1900s, with direct-to-customer deliveries fulfilled by the postal service or a parcel delivery service. Amazon initially followed this same model, injecting millions of packages each day into third-party carrier networks.
Through the years, Amazon has become really good at figuring out how to stage inventory as close to the customer as possible, both geographically and temporally. Amazon’s distribution network has grown from a pair of fulfillment centers in 1997 to more than two thousand facilities worldwide in 2024. In 2014 they received a patent for “anticipatory shipping,” which is the application of predictive analytics to customer shopping and browsing history to start moving product even before an order is placed. Today, 40% of customer demand is expected to be fulfilled locally, and 80% fulfilled regionally.
So, what does that have to do with competing against the pickup and delivery giants?
It is important to view any expansion by Amazon into consumer logistics through the prism of network topology.
The mathematical foundations for the algorithms that optimize transportation networks go back centuries, and their application to logistics was formalized in the 1940s to support the war effort. Amazon operates what is referred to as a one-sided network. Shipments are created within a relatively small number of warehouses and fulfillment centers, and delivered to a much larger number of customers and destination locations.
In two-sided networks, shipments can be created anywhere to be delivered anywhere. While pickup routes usually include a large number of regular customers, the full set of locations changes daily. Two-sided networks must also balance conflicting demands. Customers want their deliveries as early in the day as possible and their pickups as late as possible. This may require multiple visits to the same address at considerable additional cost. Interestingly, this is less of an issue for the Postal Service because they are required by law to provide services to all communities, and pickups are made at the same time as delivery, whenever that happens to be.
Now, let’s get back to Amazon.
Over time, shipping costs increased, but so too did sales as well as the number of deliveries made within a certain geographic area (a.k.a. delivery density). “Fulfilled by Amazon” with its blue vans was introduced in 2014, and by 2020 Amazon had invested $39 billion to build out a U.S. delivery network that delivered about half of its own packages.
A comparatively small number of returns flow from designated drop-off points, not individual customer residences, to an expanding set of locations which now includes Kohl’s, the UPS Store, Whole Foods, and Staples.
In 2018, Amazon launched a ground shipping service called Amazon Shipping on a limited basis to select third-party sellers. It was discontinued in 2020 in the wake of volume surges during the COVID pandemic, then relaunched in 2023 with expanded service areas, more sellers, and seven day a week pickup and delivery. Significantly for our discussion here, shipments can carry products sold through Amazon, the producer’s website, or other sales channels. Not just through the Amazon platform.
Notice, though, that despite all of these service expansions, the network remains essentially one-sided. The number of origin locations may have increased, but it is still fixed and still overwhelmingly vendor to consumer.
So, back the original question, “Is Amazon going to start competing head-to-head with other parcel services?” My guess:
Possibly, but probably not for a while.
Although fulfilling shipments from sellers’ locations, regardless of the platform on which the sale was made, is a step in that direction, it appears that shipping is being used as an incentive to increase platform adoption and e-commerce market share rather than the other way around. Today, Amazon delivers about two-thirds of their packages themselves. And those packages are likely strategically selected. Which brings us to the first hurdle:
Until 100% of Amazon’s volume can be accommodated within their own delivery network, they will have to maintain relationships with other carriers.
To compete in the consumer shipping market, Amazon would almost certainly have to go it alone. Will current delivery partners still carry the other third of their package volume if they are competing directly? Will the UPS Store still accept Amazon Returns or Amazon Shipments? Maybe. I’ve seen other examples of “coopetition,” where competitors partner on a contract or service agreement that is profitable for both parties.
Assuming that Amazon intends to compete in the consumer shipping market, person-to-person shipping would likely be introduced as a new benefit for Prime members. Individual shippers typically pay the higher list prices and maybe the revenue and yields would justify the expansion. It would probably work like Amazon Returns where shippers enter details online and then take their packages to a specified drop-off location. This would increase the volume on the ingest side of the network and in the hub sort, perhaps significantly, but the pickup locations would still remain constant and known. Still the same one-sided network.
Perhaps they’ll go all the way and offer package pick-up at residences, but the pickups would likely be integrated into the delivery routes (dynamically), and the pickup would be done when the driver happened to be there. Like the Postal Service. There would also probably be an additional charge for the home pickup service. The network is starting to look two-sided, but the pickups are incidental to deliveries.
The biggest hurdle is the potential for its success.
The most significant challenges appear to be in the areas of sortation and network capacity expansion.
Space to hold the dropped-off packages may be an issue in locations that have only a shelf or two for returns. Accommodations would be required to handle the overflow. The number of packages accepted may need to be throttled to avoid overwhelming the network. Additional drivers and sorters will have to be hired.
Amazon will have to balance the costs of expanding end-to-end network capacity and fulfilling the shipments currently handled by other carriers, against the revenue generated through this new service as well as the expected gains from new sellers. Only the Marketing and Finance folks know if it is worth it.
The rollout of any new large capability is invariably accompanied by some hiccups followed just as reliably by reports questioning the viability and prudence of the new venture. Maybe those reports will be prescient.
At first.
Amazon learns fast and adjusts fast. They’ve demonstrated their willingness to make large investments and to scale operations. They’ve already tried one pickup side expansion, discontinued it, and are now trying again. Consumer-to-consumer shipping through Amazon might not be imminent, but the capabilities that they are developing and experience that they are gaining are moving them in that direction. And it seems to be a logical step at some point.
Ultimately, the answer to the question, “Is Amazon going to start competing head-to-head with other parcel services?”
Only if and when they want to.
