Why Last-Mile Delivery Now Costs More Than Half of Shipping
Between 2018 and 2023, the last mile went from 41% of total shipping cost to 53%. In five years, the final few kilometres of a delivery became the majority of what it costs to ship anything.
Last-mile delivery is the stage where goods move from a local distribution hub to the customer’s door. It is a small fraction of the distance a product travels and the largest share of what it costs to ship. Since 2025 it has also been a stage that some cities are closing to conventional vehicles outright. Dutch municipalities began designating inner-city zones that polluting vans and trucks may not enter, with rollout planned across 28 cities and Schiphol Airport by 2030.
Who does last-mile delivery matter to?
The last mile is unusual in how many different decisions it sits underneath. Six groups are acting on it right now, and each is asking a different question of the same market.
| Who | The decision in front of them |
|---|---|
| Logistics operators, carriers and 3PLs | Margin is leaking in the most expensive stage of their chain. Which interventions pay back on an existing fleet, and which require replacing it? |
| Retailers and e-commerce teams | How to keep a fast, cheap delivery promise as the cost of fulfilling it rises, and how to retain access to city centres that are closing to conventional vans. |
| Corporate innovation and sustainability teams | Which suppliers, partners and technologies to scout, pilot and scale, and how the last mile shows up in Scope 3 reporting. |
| Investors, corporate venture arms and growth funds | Where capital is concentrating, whether that matches where technologies are commercially ready, and which categories are gated by regulation rather than engineering. |
| Cities and policymakers | The last mile is roughly 13% of city emissions and a growing share of congestion. Which interventions actually shift behaviour, whether access zones, microhubs or cargo bike rules? |
| Advisors and consultancies | Building credible vendor long-lists and short-lists for clients, and keeping a defensible market view current as the technology set moves. |
Those questions have one thing in common: none of them can be answered by knowing the size of the market. They all require knowing which specific technologies work, under which conditions, and who is building them. That is what the second article in this series takes apart, mode by mode.
Why is last-mile delivery the most expensive stage of the supply chain?
Long-distance freight is efficient because it consolidates. Large volumes move along planned routes between fixed points, and cost per unit falls as the load grows. The last mile inverts every one of those advantages.
Instead of one vehicle carrying a full load between two points, the last mile requires many vehicles making frequent short trips to dispersed addresses, usually in congested streets. Each trip involves repeated stops, low vehicle utilisation, tight delivery windows and a meaningful chance that nobody is home. All of it adds fuel, labour hours, congestion and emissions at once. That is why the last mile is both the least efficient and the most expensive part of the chain, and why operators already working on thin margins find it so hard to fix.
The pressure compounds from the demand side. Consumers expect delivery to be fast, flexible, reliable and free, and those four are difficult to satisfy together. Same-day and next-day service require more local inventory, more vehicles, more labour and tighter route planning, while free shipping limits how much of that additional cost can be passed on.
How much of a city’s emissions come from last-mile delivery?
The same inefficiencies that make the last mile expensive make it emissions-intensive. On the World Economic Forum’s 2024 assessment of urban logistics, last-mile delivery accounts for approximately 54% of the transport sector’s emissions and around 13% of overall city emissions. The same report puts last-mile costs at 53% of total shipping cost in 2023, against 41% in 2018.
Both shares are set to grow. Without effective intervention, the number of delivery vehicles in cities is projected to rise by 61% by 2030 and delivery-related carbon emissions by around 60% globally. City-level variation is wide, from roughly +45% in Strasbourg to +84% in Bengaluru. The direction is the same everywhere.
One consequence is worth stating plainly, because it is unusual. In the last mile, the commercial case and the environmental case point the same way. Reducing unnecessary kilometres, raising vehicle utilisation and cutting failed deliveries lowers cost and emissions through the same intervention. Operators are not being asked to trade margin for carbon.
What regulations already affect last-mile delivery?
Policy has moved from encouragement to enforcement in some jurisdictions and not others, and the distinction matters for planning. Some measures set direction. Others set deadlines.
| Jurisdiction | Measure | In force | Why it matters for the last mile |
|---|---|---|---|
| Netherlands | Zero-emission zones for city logistics | From 2025, rollout to 2030 | Municipalities may designate inner-city areas that polluting vans and trucks cannot enter. Planned for 28 cities and Schiphol Airport. The most direct constraint on conventional urban delivery anywhere. |
| European Union | New EU Urban Mobility Framework | Adopted 2021 | Recognises zero-emission urban logistics as a priority, pushing cities to plan for sustainable freight, logistics space and multimodal terminals. Directional rather than binding. |
| European Union | U-space Regulation | Applicable from 2023 | Creates the airspace-management framework for complex drone operations, which the EU links to future cargo and delivery services. An enabling regulation, not a delivery law. |
| United States | NYC DOT e-cargo bike rules | 2024 | Authorises pedal-assist e-cargo bikes for commercial delivery on city streets, with safety standards, explicitly to reduce reliance on large delivery trucks. |
| United States | NYC Microhubs pilot (Local Law 166) | Mandated 2021, pilot rules 2025 | Creates dedicated zones where trucks transfer goods to smaller modes such as e-cargo bikes, handcarts and small electric vans for the final leg. |
| United States | Personal delivery device laws | 2017 onward | Virginia, Florida, Texas and Washington created legal pathways for autonomous sidewalk robots on sidewalks and crosswalks, under speed, identification, monitoring, braking and insurance rules. |
| Japan | Level 4 drone flight rules | 5 December 2022 | Permits drone flight over inhabited areas, which is the operating condition most urban and suburban delivery models require. MLIT identifies package delivery to urban, mountain and island areas as a target use case. |
For anyone planning fleet investment, the Dutch zero-emission zones are the consequential row. They convert an environmental preference into an access requirement with a date attached.
How big is the last-mile delivery market?
| Market | 2025 | Forecast | CAGR |
|---|---|---|---|
| Last-mile delivery transportation | $186B | $487B by 2035 | 10.12% |
| Digital last-mile delivery platforms | $9.64B | $32.67B by 2034 | 14.55% |
The gap between those two growth rates is the more useful number. The physical delivery market is large and growing steadily. The digital layer that plans, dispatches and tracks those deliveries is far smaller but compounding faster, because it can be sold into fleets that already exist without requiring an operator to replace a single vehicle. That asymmetry is the through-line of our technology-by-technology comparison.
Where is innovation in last-mile delivery happening?
The market divides into two pathways. Last-mile delivery vehicles change what makes the delivery, replacing or complementing diesel vans with electric vans, cargo bikes and autonomous formats. Last-mile delivery operations change how deliveries are planned and managed, using software layered onto fleets that already exist. The companies building in each are mapped in our funding and ecosystem landscape.
- Electric light commercial vehicles (eLCVs)
- Cargo bikes
- E-cargo bikes
- Autonomous on-road vehicles
- Autonomous delivery robots
- Drones
- Delivery management platforms
- Route optimisation
- Real-time delivery tracking
- Fleet management
Alongside these, out-of-home delivery models such as parcel lockers and collection points are changing the customer side of the last mile. They reduce failed home deliveries and let operators drop multiple parcels at a shared point. Their environmental benefit is conditional rather than automatic: largest when customers collect on foot, by bike or as part of a trip they were making anyway, and it can disappear when collection requires a separate car journey.
What is slowing adoption?
Most of these solutions are already deployed somewhere. Scaling them across a full logistics network is the harder problem, and five barriers recur.
- High upfront cost. Replacing conventional fleets or redesigning delivery networks requires significant capital, in the part of the chain that is already most expensive and where margins are thinnest.
- Infrastructure readiness. Electric fleets need reliable charging access, cargo bikes need safe routes and secure storage, parcel lockers need suitable locations. Willingness to invest does not create these conditions.
- Customer expectations. Fast, flexible, reliable and cheap are difficult to deliver at once, and free shipping limits cost pass-through.
- Operational disruption. New solutions mean changed routes, relocated depots, retrained staff, new software integrations and mixed vehicle types. That is short-term complexity before efficiency gains land.
- Regulatory uncertainty. Drones, sidewalk robots and even e-cargo bikes need clear local rules. Where regulation is unclear or slow, operators hesitate to invest in technology whose legal operating conditions they cannot predict.
These barriers are not evenly distributed, and that is the most useful thing about them. Capital intensity and infrastructure dependence apply overwhelmingly to the vehicle pathway. The operations pathway is software layered onto an existing fleet, so it faces materially lower versions of both. That is why the commercially ready technologies are almost all software.
Which last-mile technologies are ready now, and where is the money going?
Net Zero Insights assesses ten last-mile technology categories for commercial maturity and tracks the companies building in each. Setting those two views side by side produces a result worth sitting with.
Seven of the ten categories are at full commercial readiness. Every one of those seven is either software or a bicycle. The three that are not ready are on-road autonomous vehicles, delivery robots and drones. They are also the three that attract the largest funding rounds and the most coverage.
Anyone using round size as a signal of where the last mile is being solved today is therefore reading the market backwards. The technologies producing measurable cost reductions right now are not the ones raising the biggest rounds. We take that pattern apart in the ecosystem and funding landscape.
Frequently asked questions
The full Last-mile Delivery Market Snapshot sets out the techno-economic benchmark for all ten categories, the maturity assessment behind it, and the legislative picture across the EU, Netherlands, US and Japan.
The 883 companies, their funding rounds and their investors are on the Net0 Platform. Book a demo →
The post Why Last-Mile Delivery Now Costs More Than Half of Shipping appeared first on Net Zero Insights.