Authors:
- Eric Markowitz, Managing Partner & Director of Research
- Daniel Crowley, CFA, Managing Partner & Portfolio Manager
Amazon’s “Buy Now” button isn’t magic. It’s just meticulously engineered to feel that way. Tap the button on a bottle of vitamins at lunchtime, and by mid-afternoon, an Amazon package could be on your doorstep.
Near instant delivery of virtually any product you could imagine would have seemed implausible just a few years ago. But today, this is not just the reality of Amazon’s delivery system—it’s increasingly becoming the standard against which all other retailers are being measured.
And we believe it’s only the beginning.
We recently toured an Amazon Robotics fulfillment center just outside Denver, Colorado—a nearly one-million-square-foot facility topped with 17 acres of solar panels on its roof. The sheer scale of the operation was staggering, but what truly stood out was the sophisticated robotics infrastructure powering it all.
From autonomous mobile robots swiftly transporting inventory to robotic arms sorting and moving goods with precision, the entire system was a masterclass in operational efficiency. It was a vivid look at how Amazon continues to push the boundaries of automation to drive speed, accuracy, and scale across its logistics network.
Beyond our boots-on-the-ground research, we’ve also leveraged our partnership with InPractise to conduct a series of in-depth interviews with former Amazon engineers and executives. These conversations provided critical insight into the scale, ambition, and long-term roadmap of Amazon’s logistics and robotics strategy. This report has been several months in the making, involving hundreds of hours of combined research, analysis, and firsthand investigation.
Our goal was to move past surface-level narratives and build a deep, nuanced understanding of how Amazon’s automation infrastructure is evolving—and what it means for the company’s long-term economics (and why we’re so bullish on AMZN over the long-term).
The world of retail is undergoing a radical transformation in logistics, as giants from Walmart to Target race to catch up with Amazon’s logistics and delivery network. For the past decade, Nightview Capital has invested in Amazon, and in that time, we have become increasingly convinced that the market underappreciated the true innovation behind Amazon’s supply chain. Amazon’s logistics prowess isn’t just a cool trick; it’s a disruptive force setting new customer expectations. If Amazon can routinely deliver in hours, competitors must adapt or fall behind.
How did Amazon clinch this prime (yes, pun intended) position and what does it really take to get an item from a warehouse to your door in two hours flat? Through interviews with former employees, firsthand tours of Amazon fulfillment centers, and our own internal research, we traced the vast, interconnected web of forecasting models, warehouse robotics, delivery drones, and routing technology that makes it all possible.
We remain highly bullish on Amazon, particularly because we believe the company’s core retail business—long viewed by investors as low-margin and capital intensive—is now exiting its heavy investment phase and entering a “harvest cycle” of profitability.
Years of infrastructure buildout, automation, and logistics optimization are beginning to bear fruit, and we expect operating margins to expand meaningfully over the next 2–3 years.
In our view, Wall Street continues to underestimate this margin inflection, which we believe could serve as a powerful catalyst for stock appreciation as the true earnings power of Amazon’s retail engine becomes increasingly evident.
To tell the story of how Amazon makes it happen and why we believe it gives Amazon a competitive edge, we’ll start with what happens the moment you place an order on that bottle of vitamins. We’ll then follow the high-speed journey of a single package, as it travels through fulfillment centers, regional hubs, and finally into the hands of delivery drivers racing the clock.
We collaborate with InPractice for primary research to ensure a comprehensive, on-the-ground perspective and ensure our process is entirely compliant. This report is purely for informational purposes and is not a recommendation to buy or sell Amazon stock. All in Practise interviews are for informational purposes only and should not be relied upon as a basis for investment decisions. InPractise is an independent publisher and all opinions expressed by guests are solely their own opinions and do not reflect the opinion of InPractise or Nightview.
Part 1: Placing an order
Our journey begins at the click of the “Place Order” button.
Instantly, Amazon’s backend brain, known internally as its Supply Chain Optimization Technologies (SCOT) system, springs into action, running complex algorithms to decide how to fulfill your order in the promised time. One of the first tasks is determining which fulfillment center is best positioned to ship your item. Amazon has hundreds of warehouses across the country, so picking the optimal one is crucial. The system considers your location, where the item is in stock, and which facility can deliver fastest.
Amazon doesn’t leave this to chance. Long before you clicked “Buy,” Amazon’s AI-driven forecasting models predicted that someone in your city would likely order those vitamins about now. Browsing trends, past purchases, seasonal patterns, even regional preferences all help the SCOT forecasting system anticipate demand for every product and route inventory accordingly. During big shopping events, this predictive power operates at an incredible scale. On Cyber Monday 2023, Amazon’s AI systems forecasted daily demand for over 400 million products and even predicted where orders would come from.
Once your order is in, Amazon’s system quickly selects a warehouse that has the item and can meet the two-hour delivery window. In the past, any fulfillment center in the country could ship to any customer for two-day delivery. But for same-day deliveries, Amazon realized it needed to pivot from a national network to a local one – positioning inventory close to customers. This led to the creation of smaller, city-centric fulfillment hubs, sometimes called Prime Now hubs or sub-same-day sites, which hold the most frequently ordered items. These mini fulfillment centers are strategically located near major urban areas, so that a driver can reach you within hours.
If the item is in stock at the closest same-day facility, that warehouse fulfills the order. If not, the system might look at a regional fulfillment center to do the job. The goal is to minimize distance and handoffs, because every extra mile or transfer could add hours and cost.
By the time you’ve seen the “order confirmed” screen, the item has likely been claimed by a specific warehouse a series of automated actions kick off in the warehouse: An order ticket is generated, and a robot or human picker is directed to grab your item from the shelf. Compared to a regular two-day order that might sit in a queue for a bit, a same-day order is flagged as urgent, often bumping it to the front of the line for picking and packing.
Part 2: Inside the Fulfillment Center
Behind the scenes, AI is coordinating these workflows, assigning tasks to warehouse staff and robots in optimal sequence. This involves complex orchestration of labor and machine capacity.
Over a decade ago, Amazon made a pivotal move by acquiring a robotics company called Kiva Systems. That acquisition, one former head of systems and products at Amazon Robotics told us, was based on the fact that “roughly 50% to 60% of the labor costs in a warehouse were due to the time spent walking to pick up inventory.” Automating this was the first step.
Those orange Kiva robots – now an integral part of “Amazon Robotics” – revolutionized the picking process. Instead of workers walking miles of aisles to find products, Kiva robots carry shelves, called pods, to the workers. Today, Amazon has deployed more than 750,000 robots across its operations network to help fulfill orders.
As soon as your vitamin order comes in, a robotic drive unit similar to the Kiva bot, nicknamed Hercules, zips off to fetch the pod containing that item. Hercules can lift and move entire shelves of products, bringing the right section of inventory to a stationary human picker. For larger or heavier items, a beefier robot called Titan performs a similar role, capable of lifting double the weight, so even bulk goods can be robotically delivered to workers.
By having robots do the legwork, Amazon dramatically reduces the time it takes to locate an item and eliminates the need for workers to roam the warehouse. This also helps Amazon manage demand in a tight labor market. “The cost of turnover within fulfillment centers is high, and companies face limits in meeting customer demand,” the former head of systems and products at Amazon Robotics, said in an InPractise interview. “Robotics is now more about enabling volume than just saving costs.”
Once the shelf arrives, a human worker, or in some cases, a robotic arm, picks the correct item for your order. Amazon’s system uses computer vision and scanners to verify the picker grabbed the right product and quantity. In some cutting-edge sites, robotic arms like Sparrow can identify and pick individual items from bins, handing them to humans for packing.
Similarly, Amazon’s new Sequoia robotic system combines AI, mobile robots, and automated storage units to optimize how inventory is organized and delivered to workers. It can store inventory up to 75% faster than previous methods which means products are shelved and ready for picking more quickly after they arrive at the warehouse. Sequoia also consolidates items and frees up space, ensuring popular products (like your vitamins) are positioned for quick retrieval.
When inventory is stowed or retrieved by robots, it’s often delivered to employees at ergonomically convenient stations, where they don’t have to climb ladders or bend down to bottom shelves. This reduces strain and speeds up work, saving precious time per order.
After someone picks your bottle of vitamins, it’s placed into a yellow tote and whisked via conveyor to the packing station. Here too, Amazon employs automation, with machines that can scan items and auto-pack orders into custom-sized packages. In many facilities, human packers still place items in boxes or padded mailers, but technology assists them. Computer screens may suggest the optimal box size and packing materials for each order, and in newer sites, automated packaging machines measure items and create a perfectly fitted box or sealed paper bag on the fly.
This not only saves time spent hunting for the right box, but it also improves sustainability by reducing excess packaging. This approach has helped Amazon avoid over 130 million plastic bags a year – a nice environmental win alongside faster output. “The goal has always been to minimize the amount of air being shipped,” said the former head of systems and products at Amazon Robotics in an InPractise interview.
Once packed and labeled, the package is ready for outbound sorting. Robotic arms come into play again: Amazon’s first robotic arm, Robin, was designed to grab packed parcels off the line and put them onto another conveyor or into sorting carts, speeding up what used to be a manual process of lifting boxes. Another robot arm called Cardinal uses computer vision to select one package out of a jumble, read its label, and insert it into the correct cart for loading. Cardinal can handle up to 50 lb boxes, sparing human workers from heavy lifting and potential injury.
If a machine learning system detects a backlog at one packing station, it can divert incoming work to another. It’s a dance of humans and robots, managed by AI in real-time. Notably, Amazon’s use of robotics isn’t about replacing humans so much as augmenting them to work faster and safer. A joint study with MIT found a majority of workers believe robotics and AI make their jobs better and more efficient, supporting the company’s messaging that automation and employees together can improve productivity.
By the time your vitamins are in a box with a shipping label, the fulfillment center’s job is essentially done. In some of Amazon’s sub-two-hour facilities, this picking process takes between 15 to 30 minutes, before orders are packed and handed off to a driver.
The package is now prepared for the next leg of its journey: getting from the warehouse to your local area, a phase known as the middle mile.
Part 3: Navigating the “Middle-Mile”
If the fulfillment center is the kitchen where the meal is prepared, the middle mile is the food runner bringing it to the waiter. In Amazon’s case, this often involves sortation centers, trucks, and sometimes even airplanes working in concert to move packages rapidly across regions. For same-day delivery, the middle mile is extremely compressed. It might even be just a quick drive from a local mini-warehouse directly to your neighborhood. But to understand why that’s so novel, it’s worth exploring how Amazon’s broader middle-mile system works.
In many cases, after items are packed at a fulfillment center, they are shipped to a sortation center. Sortation centers are specialized facilities where packages from various fulfillment centers are sorted by destination region. Think of a sortation center as a postal distribution center. Its job is to group packages headed to the same area. For example, your vitamins from a fulfillment center in your metro area might be sent to a regional sortation center that handles all packages for your part of the city.
At the sortation center, conveyor systems and workers, often aided by scanning machines, sort parcels by ZIP code. Packages then are organized onto pallets or into bins destined for the specific delivery stations that serve those ZIP codes. This ensures that each last-mile delivery route is as efficient as possible, delivering only packages meant for the immediate area.
For same-day deliveries, Amazon relies on new mini-fulfillment centers, which combine the fulfillment and sortation functions under one roof. These are smaller warehouses embedded near city centers specifically for one-day and same-day orders. By building out dozens of these urban fulfillment hubs, Amazon reduces the distance to customers, which tackles both speed and cost. As of 2023, Amazon had 46 sub-same-day facilities in the U.S.
Each of these sites is designed to serve a tight radius, with up to 100,000 of the fastest-moving items stocked on-site. By collapsing the middle mile, Amazon cuts out a time-consuming handoff. An order can go from being picked to a delivery vehicle in a matter of minutes. “They are no longer going from fulfillment center to sortation center to delivery station,” an executive at a leading Amazon logistics consulting firm said in an InPractise interview. “This takes out headcount, cost, and transportation expense.”
When packages do need to travel from a fulfillment center to a sortation center or from sortation to a delivery station, Amazon utilizes its transportation network: a fleet of Amazon trailers, semi-trucks, and sometimes cargo planes. Amazon has developed a vast middle-mile capability, operating its own Amazon Freight trucks and partnering with external carriers to shuttle goods between facilities. On longer distances (say, overnight from one state to another), Amazon Air might be involved. Amazon Air is the company’s dedicated air cargo service, with a fleet of about 95 aircraft serving over 70 airports worldwide.
These planes allow Amazon to fly packages between regions extremely quickly. Amazon Air’s ability to reposition inventory overnight is a backbone of one-day delivery and makes the network resilient. Some Amazon Air hubs can handle 450,000+ packages on 12 or more flights daily.
Trucks, however, remain the primary middle-mile workhorses. Amazon’s trucking routes are optimized by algorithms to ensure minimal delay. They’ve even developed internal software for dynamic route scheduling and load planning. A system that manages trailer handoffs uses multiple drivers to keep long-haul trucks moving with minimal stops. One driver takes a load halfway, then another driver takes over. AI identifies the most efficient exchange points.
This kind of optimization helps Amazon move goods cross-country faster. On a regional scale, predictive analytics help plan truck departures from fulfillment centers to sortation centers. Amazon’s systems forecast how many packages for each area will be ready by a certain hour, so they can dispatch a truck at the right time. That ensures trucks aren’t leaving half-empty, but also aren’t waiting so long that packages miss their delivery windows.
Inside sortation centers, Amazon employs both people and machines to rapidly organize packages. Workers stack packages onto pallets for specific delivery stations, and those pallets get loaded onto 18-wheelers or box trucks for the short trip to the station. Amazon has been experimenting with automating parts of this process. In some centers, robotic arms and systems (nicknamed Pegasus) scan and divert packages, reducing human error.
Amazon also developed patent-pending technology for “dynamic sortation zones.” Traditionally, a sortation center is divided into fixed zones, where there is one zone per group of ZIP codes. But this can be inflexible if one zone suddenly has a surge of packages. Amazon’s technology uses AI to adjust sortation zones in real time based on incoming package volume and forecasts.
Once sorted, the packages are loaded for the final journey. By now, perhaps 90 minutes after you ordered, your vitamins are on a pallet with other local orders bound for a delivery station in your area. In many same-day cases, the delivery station might be the same building as the fulfillment center. Either way, we’re now at the last stage: getting your order to your doorstep.
Part 4: Getting to your door
The “last mile” is often the most visible part of the journey – the part where a driver pulls up to your house and drops off the package. It’s also one of the most complex and costly segments in the logistics chain. A decade ago, Amazon relied heavily on carriers like UPS, USPS, and FedEx to deliver packages from its warehouses to customers. But as Prime’s promise shifted from two-day to one-day and same-day, Amazon realized it needed more control over the last mile to meet tight deadlines.
Amazon still uses UPS, USPS, and regional carriers for some deliveries – particularly in rural areas or for certain types of shipments. But those third-party carriers are now a smaller slice of the pie. UPS’s revenue from Amazon parcels, for one, has dropped to the lowest level since 2019. Amazon’s strategy has been to bring delivery in-house to ensure speed and reliability.
Today, Amazon’s in-house logistics team delivers the majority of Amazon packages in the U.S. In 2022, Amazon delivered about 5.2 billion packages in the U.S., more than FedEx, and by 2023 it had leapfrogged UPS as well. It’s an astonishing transformation from customer to competitor. In addition to its sophisticated tech stack, Amazon now runs the largest parcel delivery operation in the country.
Amazon’s in-house delivery encompasses several programs. One is the Delivery Service Partner (DSP) program, where Amazon contracts with small independent businesses to run fleets of Amazon-branded vans. There are over 3,000 such DSP companies worldwide, employing more than 275,000 drivers. These drivers wear Amazon uniforms, drive Amazon blue vans, and exclusively handle Amazon packages, but they technically work for the local DSP business owner. DSP drivers deliver over 10 million packages every day for Amazon ranging from routine Prime deliveries to ultra-fast orders like your same-day vitamins. Amazon’s software, often running on a device the drivers call a “rabbit,” sets the order of deliveries, and the driver follows the device’s instructions, which use GPS and algorithms to direct drivers stop-by-stop.
This system factors in promised delivery times, distance, and other priorities. For same-day deliveries, the process can involve multiple waves. Amazon might send out a first wave of vans in late morning for noon deliveries, then a second wave in mid-afternoon for evening deliveries. Many Amazon delivery stations have dispatches throughout the day to accommodate same-day orders that keep coming in. Because same-day delivery can be ordered as late as 10pm, Amazon needs a flexible system where drivers are doing continuous loops or new drivers pick up shifts later in the day.
One challenge of last-mile delivery is unpredictable obstacles. Traffic accidents, bad weather, address issues, or a customer not being available for a signature can gum up the works. But Amazon’s routing system is built to be dynamic. If one driver is running behind, another nearby driver may be rerouted to take an extra package. The dispatchers at delivery stations monitor the fleet’s progress in real time on a screen, aided by algorithms that flag delays.
Amazon also introduced a feature for drivers called “Delivery Zero” that can reorder stops by, say, skipping to another stop if there’s a holdup at the current one. Amazon equips its delivery vans with telematics and cameras, not only for safety but also to gather data that can improve routes over time. Some vans even have an AI system called Vision-Assisted Package Retrieval (VAPR). This augmented reality system helps drivers find the right package in their cluttered van faster by projecting indicators on the packages. VAPR can cut about 30 seconds off each delivery stop. That might not sound like much, but multiply by 100 stops and you have 50 minutes saved.
By building this private parcel network, Amazon has created a strategic asset. It can ramp up capacity during peak demand, it can fine-tune delivery practices, and it gains a treasure trove of data to improve its operations.
In addition to DSPs, Amazon uses Amazon Flex, a gig-economy platform where independent drivers use their personal vehicles to deliver packages. Much like Uber or DoorDash, Flex drivers use a smartphone app to manage their work. They arrive at an Amazon delivery station, often a small warehouse in a city. The app will assign them, say, 20-40 packages that are all destined for addresses in a certain area, and it will generate an optimized route for them.
Flex routes for same-day deliveries might be as tight as one neighborhood or a few zip codes. The route optimization behind this is powered by Amazon’s routing algorithms, which consider real-time factors like traffic and distance. Amazon’s systems use dynamic route planning to map out the most efficient path for drivers, aiming to avoid traffic jams and minimize drive time. If there’s heavy congestion on one side of town, the algorithm might sequence stops in a slightly different order or send drivers alternate ways. The driver’s app provides turn-by-turn directions and can update automatically if needed.
Prime Now and same-day deliveries rely heavily on Flex, especially for small, urgent orders that might not require a van. It also gives Amazon access to a flexible workforce that can scale up during peak times like Prime Day or holidays without permanently increasing staff. Amazon Flex is also heavily used in the evening wave, with gig drivers picking up slack for the last rush of the day.
The business impact of all of this is huge: Amazon turned what was traditionally a cost center — shipping — into a competitive advantage and even a potential revenue source. Amazon has even started delivering packages for other companies in some areas, effectively becoming a carrier itself.
For customers, the last-mile network means faster, more reliable delivery. The sheer density of Amazon’s deliveries in metro areas allows efficiency that smaller e-commerce players can’t match.
At last, the driver arrives at your doorstep for the final handoff of your vitamins. He takes a photo of the package on your doormat, which pings your phone. To you, it all feels simple. But that feeling is the result of an enormous amount of coordination and innovation that Amazon has invested in for decades. This last-mile mastery is one of the key reasons Amazon can confidently offer same-day delivery as a regular service, not just a rare luxury.
Part 5: Challenges and Innovations
Speed can be costly. Despite Amazon’s efforts to group orders, a driver might do fewer deliveries in a two-hour cycle than they would in a full-day route, and dispatching trucks half-empty just to rush packages erodes margin. When you demand such a quick turnaround, you lose some economies of scale.
Labor costs and intensity are another issue. Same-day promises put pressure on workers at every stage. Warehouse pickers have to hit higher throughput. Drivers have tighter deadlines. Keeping efficiency high without burning out employees is a delicate issue.
Another challenge: inventory complexity. Offering same-day delivery on millions of items means you must store those items close to customers. Sometimes, Amazon’s best estimate based on maximal data makes the wrong call, leading to overstocked items that don’t sell, or understocked items that sell out and break delivery promises. It’s a highly dynamic problem and part of the reason why same-day delivery is currently focused on metropolitan areas; expanding it to suburban or rural areas would require even more local nodes or mini-warehouses within a short radius of customers. But if Amazon scales same-day from, say, 120 metro areas to 300, those predictive capabilities will be further tested.
But Amazon isn’t standing still; it continuously introduces innovations to make same-day delivery faster, cheaper, and more sustainable. One major innovation has been the development of micro-fulfillment centers. At this point, a leading Amazon logistics consultant said in an InPractise interview, these facilities are the “exception rather than the rule.” But the company plans to build around 150 more in the coming years.
Another frontier is drone delivery. Amazon has been touting its Prime Air drones for years, and while delays and regulatory hurdles slowed things, they are finally operational in test markets. In late 2024, Amazon launched its most advanced delivery drone, the MK30, and began drone deliveries in two U.S. locations: a part of metro Phoenix, AZ and College Station, TX. These drones can carry packages under 5 pounds and deliver them in 60 minutes or less, dropping them at designated locations at customer homes. The MK30 has double the range of earlier models and is quieter, addressing some past challenges. While drone delivery is not yet widespread, it represents a potential leap in same-day service, especially for small, lightweight items (like medicines, batteries, or that vitamin bottle) that someone might need urgently. However, drones come with their own challenges: they need clear airspace and suitable drop zones, and carry limited payloads. In all likelihood, they’ll augment, not replace, traditional delivery.
In addition to drones, Amazon has experimented with autonomous delivery robots (“Scout” was a sidewalk delivery robot tested in some neighborhoods) and is investing in self-driving vehicle technology that could someday power its delivery fleet. Imagine a world where an Amazon van drives itself through the night to a neighborhood, and maybe a drone or robot completes the last few feet to the door. It’s futuristic, but Amazon is exploring all angles.
Amazon also introduced technologies like the Packaging Decision Engine (PDE) to auto-choose optimal packaging, saving time and materials. A big challenge with rapid delivery is the environmental impact. Faster delivery often means more single-item shipments, less batching of orders, more vehicles on the road, and potentially more packaging waste. This can increase carbon emissions per package delivered. But Amazon, mindful of criticism, has paired its logistics expansion with sustainability initiatives, including embracing more recyclable materials. The shift to paper padded mailers and custom-sized packaging helps eliminate plastic and void filler.
But the crown jewel with regard to Amazon’s sustainability initiatives is the rollout of electric delivery vehicles (EDVs). Amazon partnered with EV maker Rivian to design custom electric vans, and they’ve committed to putting 100,000 electric vans on the road by 2030. As of late 2024, over 20,000 of these electric vans were already deployed across U.S. cities. These vans produce zero tailpipe emissions, meaning same-day deliveries in those routes no longer contribute to local air pollution or carbon emissions. It’s a major investment — those vans aren’t cheap — but it’s key to Amazon’s pledge to achieve net-zero carbon emissions by 2040. In the future, the company hints that its routing technology might prioritize not just speed but also carbon efficiency.
Despite these efforts, it’s an ongoing challenge to make same-day delivery truly sustainable. It’s essentially a trade-off of convenience and efficiency. The industry might move toward models like consolidated same-day deliveries, where customers choose to get all of their Amazon orders for the day in one drop. Amazon hasn’t done that yet for same-day, but it has an “Amazon Day” option for Prime where customers can choose to get all their orders on one day of the week.
Amazon’schallenges aren’t just related to tech and cost. There’s also a community impact. Neighborhoods see more Amazon vans cruising around than ever. That can create issues with congestion or noise. Amazon might face local restrictions if same-day traffic is deemed a nuisance.
Keeping the workforce motivated and safe is another challenge. Warehouse jobs under time pressure can be prone to injury, and delivery drivers under the gun might take risks on the road. Amazon has introduced programs and technology, like wearables that promote safe lifting and driver monitoring systems that give feedback on driving behavior to address these concerns. Ultimately, a challenge is ensuring that the pace of same-day delivery doesn’t come at the cost of worker well-being. This is an area where public scrutiny is high, and Amazon will need to continue innovating not just in robotics and AI, but in how it manages and supports its people.
Part 6: Business Impact and Competitive Edge
Amazon’s same-day delivery program isn’t just a logistics experiment – it’s a strategic business move. Amazon has conditioned shoppers to expect quick, reliable delivery, and same-day delivery added immense value to the Prime subscription. Now that shoppers can get many items the same day, they have even less reason to shop elsewhere. This drives customer loyalty and stickiness. In the company’s Q4 2022 earnings call, CEO Andy Jassy highlighted that Amazon’s push to speed up delivery has helped drive customer demand because it boosts usage.
By making online shopping nearly instantaneous, Amazon also captures sales that might otherwise go to a local store. This convenience can lead customers to spend more over time. According to Amazon, in 2024 it delivered at record speeds and delivered over 9 billion items on the same day or next day globally. Prime members worldwide, meanwhile, saved nearly $95 billion in shipping fees in 2024 by getting fast, free delivery. This kind of value keeps Prime membership renewal rates high, which in turn means steady subscription revenue for Amazon and more captive shoppers for its ecosystem.
Same-day capability also opens up new categories of purchases. Think of a $5 item. You normally wouldn’t buy it if shipping took two days or cost extra, but if you can have it in hours for free, you will. This drives incremental sales. It also makes Amazon an attractive platform for sellers. That can bring more merchants to Amazon, expanding product selection, which in turn attracts more customers.
Amazon’s success has already put pressure on traditional retailers to step up their fulfillment game. Walmart, for one, has leaned heavily into leveraging its biggest asset: 4,700 stores across the U.S. Walmart began offering same-day delivery on groceries and other items through services like Walmart Grocery and later Walmart+, its subscription program. As of early 2023, more than 3,900 Walmart stores offered same-day delivery, covering a huge portion of the population.
Walmart uses those stores as mini-fulfillment centers. In a way, Walmart has a denser network of “fulfillment nodes” than Amazon; a Walmart store is within 10 miles of 90% of Americans. For local one-hour type deliveries, that’s a big advantage. Walmart taps gig workers via its Spark Driver platform, which is analogous to Amazon Flex, to pick orders from shelves and deliver them to customers. This approach has led Walmart’s store-fulfilled delivery sales to nearly triple in two years, hitting over $1 billion a month.
Target took a slightly different approach: it acquired a same-day delivery startup, Shipt, and integrated it. It also uses stores as hubs, but has built sortation centers in key cities to aggregate orders picked from local Target stores. That way, contract drivers can deliver them efficiently. They even announced a $100 million investment into expanding next-day delivery capabilities via these sortation centers.
Not every retailer can afford dozens of warehouses or a fleet of vans. That’s why we’re seeing a proliferation of companies willing to handle that “last mile as a service,” including Instacart, DoorDash, or Uber, for example. But Amazon’s integrated model is still arguably more efficient at scale because it cuts out middleman fees and is optimized for its own operations.
Beyond the big names, the market saw a surge of hyperlocal delivery startups during the pandemic, including GoPuff and Gorilla, which promise 15 to 30-minute delivery of convenience items. Many of those were focused on snacks and essentials, and some have struggled or consolidated due to thin margins. Amazon had its own service in that vein. Prime Now, which launched in 2014, originally promised 1 to 2 hour delivery on a limited set of items, often from tiny urban warehouses or partner stores. But Amazon has since folded Prime Now into the main app, effectively making those ultrafast services part of the normal Amazon experience.
From a customer perspective, all of this competition means faster delivery is becoming a standard expectation. Amazon setting this bar means everyone from pharmacies to apparel retailers are exploring how to get products to consumers on the same day. It might not always be free, but it could become a norm for service.
Still, Amazon has the edge thanks to the integration of its systems. It owns the marketplace, the warehouses, the sortation, and much of the delivery network. This end-to-end control allows it to innovate quickly by, say, implementing a new AI routing system across the whole network, or launching a new robotics process in warehouses. Traditional retailers often have to patch together store networks, third-party delivery, and legacy systems, which can be clunky. Amazon’s also operating at a scale few can match, with billions of items delivered on the same or next day. That means more data to feed AI and more density to leverage. Amazon’s edge is cultural too: Same-day delivery is expensive, but Amazon bet that delighting the customer yields long-term loyalty and wallet share.
But it’s not an insurmountable lead. If a competitor finds a cheaper or smarter way to do hyper-fast delivery, customers would happily benefit. Amazon thus continues to invest heavily in research and development for logistics.
The future trends we might foresee include: wider use of automation where fully robot-run warehouses can operate 24/7 at high speed; a broader footprint of local hubs in new formats, like converting malls or abandoned retail stores into fulfillment centers; and anticipatory shipping, where Amazon predicts your order and ships it before you buy it, a concept Amazon patented.
Another trend is the integration of services. Amazon could one day integrate same-day delivery with its other offerings like Amazon Fresh grocery delivery. If you order bananas and a Kindle for same-day delivery, could one driver bring both? Possibly, as Amazon merges its networks.
What is clear is that logistics is now a competitive battleground and differentiator. It’s not just about what you sell or the price, but how and when you get it to the customer.
Part 7: Conclusion
The tale of clicking “Buy” and receiving your order two hours later encapsulates a groundbreaking shift in retail and supply chain management. We’ve seen that Amazon’s same-day delivery is not magic at all, but rather the result of meticulous planning, heavy automation, and iterative innovation across the entire logistics chain.
Even before an order is placed, Amazon’s AI-driven systems forecast demand and allocate inventory so that products are in the right place at the right time. In fulfillment centers, armies of more than 750,000 robots and human workers collaborate to pick and pack orders at blazing speeds. Packages are rapidly sorted and routed through a network of sortation centers and hubs that have been optimized with data analytics and even experimental AI. Finally, Amazon’s vast last-mile network of drivers, vans, and sophisticated routing software delivers the package to the customer, often beating the clock with minutes to spare.
The implications for business are profound. Amazon has demonstrated that controlling the supply chain end-to-end and investing in speed can yield higher customer satisfaction and loyalty, which in turn drives sales growth. Same-day delivery has become a competitive benchmark; many consumers now simply expect it as an option. This has pressured competitors to adopt new strategies and technologies, thereby transforming the logistics sector globally.
Supply chain professionals in all industries are watching Amazon’s playbook. Retailers are re-thinking store networks and considering hybrid models. And logistics providers are adopting AI for route planning and warehouse management, inspired by Amazon’s efficiencies.
Looking ahead, Amazon’s logistics machine will likely continue to evolve. We can anticipate more automation and the expansion of drone deliveries beyond pilot neighborhoods, which could further cut delivery times for small items to mere minutes. Micro-fulfillment centers could proliferate, potentially bringing same-day service to smaller cities and suburbs as Amazon fine-tunes the model of stocking popular items hyper-locally. The environmental impact will also take center stage. As Amazon scales fast delivery, it also aspires to do so sustainably, which means we’ll see a greener fleet and more innovative packaging and routing to reduce waste and emissions.
“Fast and free shipping” is no longer a perk; it’s a baseline expectation in many markets. In a way, Amazon’s same-day delivery is pioneering a new logistics paradigm, one that is customer-centric, technology-heavy, and continually improving.
The next time you receive a package just hours after ordering, you’ll know it arrived on the wings of algorithms and the wheels of countless innovations – a testament to how far supply chain technology has come, and an indicator that the future will only get faster.
Disclosures
The opinions expressed herein are those of Nightview Capital and are subject to change without notice. The opinions referenced are as of the date of publication, may be modified due to changes in the market or economic conditions, and may not necessarily come to pass. Forward-looking statements cannot be guaranteed.
This is not a recommendation to buy, sell, or hold any particular security. There is no assurance that any securities discussed herein will remain in an account’s portfolio at the time you receive this report or that securities sold have not been repurchased. It should not be assumed that any of the securities transactions, holdings or sectors discussed were or will be profitable, or that the investment recommendations or decisions Nightview Capital makes in the future will be profitable or equal the performance of the securities discussed herein.
Nightview Capital reserves the right to modify its current investment strategies and techniques based on changing market dynamics or client needs. Recommendations made in the last 12 months are available upon request.
This article contains links to 3rd party websites and is used for informational purposes only. This does not constitute as an endorsement of any kind. While Nightview uses sources it considers to be reliable, no guarantee is made regarding the accuracy of information or data provided by third-party sources.
