Temu’s last mile logistics challenges in the United States

Over the years, Amazon has built a large and sophisticated fulfilment network in the United States – a deep moat deterring potential competitors. 

As Temu is increasingly challenging the dominance of Amazon, how does the newcomer overcome the logistics challenges? What exact challenges does Temu face? 

A recent article by LatePost gives some vivid pictures. Below are translations of some excerpts from the article: 

The last mile cost challenge

Amazon and Temu are using the same factories to produce goods, the same sellers to sell them, and the same logistics to ship them to the United States. The most costly step in this process is also the same: the last mile. While everything else can be offshored or consolidated to reduce costs, the last mile is a real issue that must be handled locally.

This step is the most expensive part of the entire logistics process. According to a report by Capgemini, last-mile delivery accounts for 53% of the total transportation cost and 41% of the overall supply chain cost. A seller illustrated this with an example from his own experience: “One time, a courier drove to my house to deliver a single package and then left, taking an hour in total. With a $25 hourly wage, the cost of that single delivery was just that high.” While this situation is somewhat extreme, it highlights one thing: no matter how much costs are reduced at the production end, the savings could easily be wiped out by last-mile logistics.

Though most cases aren’t as extreme, the general cost per small package in the US is around US$3. Prices vary among different suppliers in the US. One supplier in the Midwest calculated that they could achieve a cost of $3.9 per package, while a West Coast supplier estimated they could lower it to $2.5 per package if the volume was high enough. Amazon’s costs are not transparent, but they charge customers $2.99 per shipment if they don’t meet the minimum purchase amount.

According to analysis from YipitData, the average order value on Temu in the US ranges between $30 and $40. This means that up to 10% of the cost could be consumed by last-mile delivery. Temu considers this too expensive and believes a reasonable cost should be one dollar, with an additional dollar subsidy in the early stages, totaling two dollars.

Most local service providers find this challenging, but some bold players emerge under Temu’s order volume. A last-mile contractor founded by Chinese entrepreneurs named UniUni came up with an idea: half of the last-mile cost comes from labour, so they decided not to work with full-time couriers. Their solution was to partner with part-time drivers, who would deliver packages while driving passengers. Initially, UniUni didn’t deal directly with the drivers but used an app called Beans Route to assign orders to them.

However, this didn’t work out. Drivers couldn’t deliver packages while transporting passengers, only managing to deliver 7-8 packages per hour. One driver showed a friend his route for a day, which involved delivering 59 packages. Beans Route planned a route for him, but following this route made it impossible to balance passenger and package deliveries. So, UniUni had to come up with another solution.

Ultimately, Temu had to gradually raise the price to $2.5 per package. For a time, Temu worked with US warehouses, almost entirely handing over this task of delivery to the US Postal Service. The larger the volume, the cheaper it became with the US Postal Service. But this approach couldn’t beat Amazon’s prices. A warehouse owner calculated that even if Temu’s cross-border costs are low, the overall cost, including the last mile, is about the same as Amazon’s.

In e-commerce, the pursuit of low costs remains centred on the last mile, which stands firm as the most significant cost factor. Temu cannot turn every aspect of this process into something it can fully control, as it does with its Chinese domestic suppliers. Anyone who wants to play this game better must face this challenge. 

Labour efficiency that can’t improve indefinitely

Logistics costs have long been a challenge in the United States. In the mid-19th century, it took twice as long to travel from New York to San Francisco as it did from Hong Kong to San Francisco. Nearly two hundred years have passed, and while the specific issues have changed, logistics as a major challenge continues to trouble American businesses.

The current specific challenge is the last mile. Whether it’s Temu, Amazon, Shein, or independent sellers, all US ecommerce players must face this cost behemoth that remains stubbornly high. Perhaps the entire logistics cost in the U.S. can be simplistically boiled down to one issue: how to reduce labour costs in the last mile.

Labour costs involve wages, time, and benefits. Cutting wages and benefits is difficult, so the only plausible solution is to improve logistics efficiency, delivering more packages in the same amount of time.

Taking Amazon, which handles the last mile most effectively, as an example: a courier typically covers 5-6 ZIP code areas. A ZIP code might include a few blocks or, in larger cases, several towns combined. Before starting deliveries each day, the courier sees their stops for the day—this means how many destinations the algorithm expects them to stop at.

If the day’s task involves 150 stops, it’s considered a relatively easy day. But in most cases, a courier’s daily workload is between 150-200 stops. However, this doesn’t represent the actual number of deliveries, as the system combines nearby locations into one stop. The actual number of delivery points could be over 200 or even 300.

These stops are separated by real physical distances. They might involve a 40-minute drive, congested streets, or closed reception rooms. A courier responsible for delivering to large houses in wealthy neighbourhoods may face driveways up to a mile long. Most driveways don’t allow other vehicles in, so the courier may need to be skilled at running to deliver the packages.

It’s not just the distance; there are a host of other issues that one wouldn’t understand unless experienced firsthand. Almost all couriers dislike night time deliveries because it’s hard to see the addresses on mailboxes or houses, especially in areas with standalone houses. They also have to worry about unrestrained dogs suddenly appearing—FedEx can refuse delivery in such cases, but Amazon couriers cannot. Temu is growing fastest among high-income groups, so Temu’s last-mile deliveries face the same issues as Amazon.

The challenge of labour costs lies in delivery efficiency, but efficiency can’t be improved indefinitely. Ultimately, the root of the challenge is American urban planning. Last century’s rationalist planners believed cities should be simplified, with reduced density, intersected by highways, and residents should move to the suburbs. As a result, America’s middle and upper classes are spread out around cities.

These suburban residential areas, along with the vast rural areas that people might drive past on highways but never stop to see, make up America’s last mile. Temu cannot solve this issue by pressuring sellers like Pinduoduo does domestically in China because the last mile is not a commercial problem but a social issue. 

However, just as Pinduoduo exerts control over suppliers in China, Temu is using a familiar approach overseas: bringing more players into the game and driving prices down. Now, besides UniUni, other last-mile carriers like SpeedX and RoRo have joined the competition. According to insiders, Temu currently pays $1.8 per package to carriers primarily run by Chinese-Americans, $1.6 to those mostly run by Mexican-Americans, and $2 to others.

One way to cut prices is to assign a single ZIP code to one driver, making it easier for them to learn the route and increase the number of deliveries per hour. Another method involves using relatively cheap labour, including many undocumented immigrants who can work with government-issued work permits while awaiting legal status. Some drivers say they can earn around $20-$30 per hour.

But this method only works in densely populated cities and doesn’t cover the vast, sparsely populated suburbs and rural areas of the United States.