Last mile delivery explained is the phrase most retail and e-commerce operators wish someone had pulled them aside about on day one. It is the slice of logistics that customers actually see, judge, and remember, and it is also the slice that quietly burns the biggest hole in your margin. If you have never thought hard about the last mile, this guide rebuilds the mental model from scratch.
In short
- The last mile is the final leg of a shipment, from a local hub or store to the customer’s door, locker, or curb.
- It typically accounts for 40 to 60 percent of total shipping cost despite being the shortest stretch.
- US retailers usually get it wrong by treating it as a freight problem instead of a customer experience problem.
- Speed matters less than predictability: a confirmed 3 day window beats a vague “1 to 5 day” promise.
- The single biggest lever for most brands is density, not technology: more orders per stop, per zip code, per route.
If you run a retail or e-commerce operation in the United States, you already pay for last-mile delivery whether you see the line item or not. This piece sits inside our modern retail logistics guide, which is the pillar where we map the full chain from warehouse to doorstep. Bookmark that one for the wider picture; come back here when you want the last-mile mechanics on their own.
What “last mile” actually means in retail logistics
The last mile is the final movement of goods from a local distribution point to the end customer. The local distribution point is rarely a literal mile away. It can be a city fulfillment center on the outskirts of Dallas, a retail store in Cleveland repurposed as a micro-hub, or a sortation depot a regional carrier shares with three other brands.
The phrase comes from telecommunications, where engineers used it to describe the expensive copper run from a neighborhood switch to a private home. Logistics borrowed it because the economics rhyme: most of the network is shared and cheap, but the final connection is dedicated to one address and stubbornly expensive.
Where the last mile begins and ends
Think of any parcel journey in three rough stages. First-mile pickup is the move from supplier or seller to a regional sorting point. Middle-mile linehaul is the long-haul truck or air movement between sort centers. Last-mile delivery is everything after the parcel leaves the final sort facility nearest the customer.
The exact handoff point varies. For UPS Ground, the last mile usually starts at a delivery center about 25 to 50 miles from the recipient. For an Amazon order coming out of a same-day fulfillment site, the last mile can begin five miles from the doorstep. For a furniture brand using a regional white-glove network, the last mile might start at a hub 80 miles out, with a two person crew handling installation at delivery.
Last mile vs. last touch
Some operators use “last mile” and “last touch” interchangeably. They are not the same. The last touch is the specific carrier or driver who hands the package to the customer. The last mile is the entire final operational leg, including the sort, the stop sequencing, the route, the failed attempt logic, and any return path. You can outsource the last touch and still control most of the last mile through how you design the upstream choices.
Why the last mile is the most expensive stretch of any shipment
Most operators assume the long-haul truck dominates cost because it travels the farthest. The math says the opposite. A semi pulling 50,000 pounds of mixed parcels 1,200 miles spreads its fuel, driver wage, and equipment cost across thousands of orders. A delivery van running 120 stops on a 60 mile loop spreads the driver’s eight hour wage across, at best, 120 packages, and that assumes every stop completes on the first attempt.
Industry analyses consistently put the last mile somewhere between 40 and 60 percent of total parcel shipping cost. The variance comes down to density, geography, and labor cost in the local market. A 1 pound parcel into dense Manhattan costs the carrier less per stop than the same parcel into rural North Dakota, even though the Manhattan parcel travels less than half the distance.
The four cost drivers you can actually move
Carriers and 3PLs will not hand you a transparent invoice that breaks down last-mile cost by driver. You can still influence the underlying drivers:
- Stop density. Orders per stop, stops per route, routes per neighborhood. Higher density crushes per-package cost.
- First-attempt success rate. Every redelivery roughly doubles last-mile cost for that parcel. Address quality, delivery windows, and ID-on-delivery rules drive this.
- Parcel profile. Volume, weight, dimensions, fragility, and special handling (signature, ID, age check) push you into more expensive service tiers.
- Service promise. Same-day, next-day, two-day, and standard pricing diverge sharply. Pushing customers from same-day to two-day on non-urgent SKUs is one of the fastest ways to cut last-mile spend.
Notice that “carrier choice” is not on this list. Switching from UPS to FedEx to USPS to a regional carrier shifts the price tag around but rarely changes the underlying physics. The four levers above do.
How a typical last-mile delivery works, step by step
Walk through a single parcel as if you were standing in the depot. Once you have followed one through the chain, the rest of the operation makes sense.
Step 1: Arrival at the local hub
The middle-mile truck pulls in overnight, usually between 1 and 4 a.m., carrying parcels from a national sort center. Each parcel has a barcode that encodes the destination zip code and a routing key. The local hub scans the truck in, breaks down the load on conveyors, and feeds parcels into a sort engine.
Step 2: Sort to route
The sort engine reads the barcode and assigns the parcel to a specific delivery route, not just a zip code. A route is a planned sequence of stops a single driver can complete in one shift. Modern sort engines use historical density data, road network maps, and live driver availability to redraw routes daily.
Step 3: Loading the vehicle
Parcels for a single route are staged in a cage or roll cage, then loaded into the van or truck in reverse-stop order, so the first stop of the day is at the door of the cargo area. Loading quality matters more than people realize. A poorly loaded van adds 30 to 90 seconds per stop while the driver hunts for the right parcel; multiply by 120 stops and you have lost two hours.
Step 4: Route execution
The driver leaves the hub between 7 and 10 a.m. Route execution is a continuous trade off between time, fuel, and safety. Modern routing software updates the sequence in real time as stops succeed or fail, traffic shifts, or new orders are injected (in the case of same-day networks).
Step 5: Proof of delivery
At the doorstep, the driver scans the parcel and captures proof: a photo, a signature, a geotag, or an in-app confirmation. Proof of delivery is what closes the operational loop and triggers downstream events: payment release, fraud-check clearance, return window start, customer notification, NPS survey.
Step 6: Exceptions and the second attempt
Not every stop succeeds. The customer is out, the buzzer is broken, the package is too large for the parcel locker, the driver cannot find safe placement. Exceptions go into a queue and trigger a second attempt the next day, a hold for pickup, or a return to sender. Each exception is a hidden cost line; well-run operations measure first-attempt success rate at the route, driver, and zip code level.
The five biggest mistakes US retailers make on the last mile
Once you understand the mechanics, the recurring mistakes are easy to spot. We see the same patterns at brands doing 5,000 orders a month and at brands doing 5 million.
1. Treating the last mile as a freight problem
Most retailers manage shipping out of finance or operations, optimizing for cost per parcel. The last mile is a customer experience problem first and a freight problem second. The brand that ships in six days for $4.50 loses to the brand that ships in three days for $6.20, every time, on repeat purchase rates.
2. Promising speed they cannot consistently deliver
“Order by 2 p.m. for next-day delivery” sounds great until your cutoff fails on the Friday before a holiday weekend. A 95 percent on-time promise customers can trust is worth more than a 70 percent same-day promise that constantly slips. Build the SLA you can hit on a bad day, not on a perfect day.
3. Hiding shipping cost in product price
“Free shipping” hides the real economics from both the customer and your internal teams. The CFO sees a stable product margin; logistics sees a quietly rising shipping spend. Shipping cost should at minimum be modeled separately, even if the customer-facing price is “free.”
4. Ignoring address quality
A surprising share of failed deliveries trace back to bad address data: missing apartment numbers, misspelled streets, autocomplete picking the wrong county. Address validation at checkout, using a maintained dataset like the US Census geography program or a commercial provider, pays back within weeks for any operation past 10,000 orders a month.
5. Optimizing carriers instead of mix
The single biggest unlock for most brands is moving from “one carrier for everything” to a mix: regional carriers in regions where they are strong, a national for the tail, USPS for low-weight residential, and a same-day partner only where it actually moves conversion. Carrier mix is a quarterly review, not a one-time RFP.
Real examples from US retail and e-commerce
Abstractions are useful; concrete operators ground them. Here is how four common retail archetypes structure their last mile, with the trade offs that shape each choice.
| Retailer archetype | Typical order profile | Last-mile approach | Why it works for them |
|---|---|---|---|
| National marketplace (Amazon-style) | High volume, mixed sizes, urgent | Owned last mile in metros, DSP partners on the fringe, USPS for the tail | Density justifies dedicated routes; control over driver experience compounds |
| DTC apparel brand | Light parcels, 2 to 5 day expectation | Single national carrier with a USPS overflow lane | Simplicity beats optimization until volume hits 100k orders a month |
| Big-ticket furniture | Large, fragile, two-person handling | Regional white-glove network with appointment scheduling | First-attempt success and damage rates dwarf raw delivery cost |
| Grocery and convenience | Fresh, perishable, ultra-fast | Store-fulfilled with a gig courier platform on the last touch | Existing store footprint becomes the local hub at near-zero marginal cost |
The pattern across all four: the last-mile strategy follows the product and the customer expectation, not the other way around. None of these brands picked “fastest possible” as the design goal. Each picked the right speed at the right cost for what the customer would actually buy on repeat.
Where same-day actually moves the needle
Same-day delivery is overused as a marketing flex and underused as a targeted tool. The categories where same-day reliably moves conversion and repeat purchase are narrow: groceries, pharmacy, pet food, baby supplies, and a few niche consumables. In most apparel, accessories, home decor, and electronics, customers consistently choose “free in 3 days” over “$8 same day” when given a clear choice. We covered the broader 2026 outlook for this in our 2026 last-mile delivery outlook for US retailers, which is the sibling piece to this one inside the same cluster.
Vendors and partners worth knowing in 2026
The last-mile vendor landscape in the US has shifted significantly since 2020. The pandemic-era boom in same-day startups thinned out, regional carriers consolidated, and the big three (UPS, FedEx, USPS) restructured their residential networks. Here is a snapshot of who actually matters today.
National networks
UPS Ground, FedEx Home Delivery, and USPS Ground Advantage remain the default backbones for most US retailers. UPS still leads on B2B and high-value parcels; FedEx Home Delivery competes hard on suburban density; USPS Ground Advantage replaced First Class Package and Parcel Select, and it is now the cheapest residential option for anything under three pounds.
Regional carriers
OnTrac (West and now expanding nationally), LSO (formerly Lone Star Overnight, strong in Texas and the Southeast), and GLS US (West Coast) are the names worth a serious look. They typically beat national rates by 15 to 30 percent in their core regions and offer better service in many zip codes. The trade off is integration cost: setting up label generation, tracking, and exception handling for a regional carrier takes engineering time most brands underestimate.
Same-day and gig platforms
Roadie (UPS-owned), DoorDash Drive, Uber Direct, and Shipt cover the same-day and on-demand layer. Pricing is volatile; integrate via a multi-carrier rate-shopping platform if you plan to use more than one. We dig into the full vendor toolbox in our tools and vendors for last-mile delivery in 2026, which is the practical companion to this primer.
Marketplace-native last mile
If you sell on a marketplace, last mile is often bundled. Amazon FBA, Walmart Fulfillment Services, and Temu’s fulfilled-by-marketplace programs take the last-mile decision off the seller’s plate, for better and for worse. Sellers used to pure-play DTC tend to underestimate how much operational complexity those programs hide; our piece on tools and vendors for Temu in 2026 walks through what a marketplace-native last mile looks like from the seller side.
Building a last-mile operation: a starter checklist
If you are inheriting a last-mile operation or building one from scratch, work this list in order. Each step is a prerequisite for the next.
- Pull 90 days of order data and map zip codes to parcel volume and weight.
- Identify your top 20 zip codes by volume and ask whether they justify a regional carrier or zone-skipping arrangement.
- Measure first-attempt success rate by carrier, zip code, and weekday. Anything below 92 percent has a fixable root cause.
- Validate addresses at checkout, before the order enters the warehouse.
- Publish a delivery promise customers can read in three seconds. “Arrives Wed, Mar 18” beats “Standard shipping, 3 to 7 days.”
- Instrument exceptions: every failed delivery, redelivery, damage, and return should land in a single dashboard.
- Negotiate carrier contracts annually with real volume data, not last year’s spreadsheet.
- Tie last-mile metrics to a customer NPS or post-delivery survey. Operational and emotional outcomes need to be on the same page.
None of this requires a heroic technology investment. The shipping module inside any modern OMS, a half-decent address validator, and a willingness to look at the numbers weekly will get a 5,000-orders-a-month operation to a solid baseline within a quarter. The broader logistics context, including warehouse design, middle mile, and returns, sits in our retail logistics pillar; this article is the deep dive on the slice closest to the customer.
Returns and the reverse last mile
Most last-mile conversations focus on outbound, but the reverse path is just as expensive and twice as likely to be ignored. Returns in US e-commerce now average somewhere between 15 and 25 percent depending on category, with apparel routinely above 30 percent. Every return is a last-mile event in reverse, with the added complication that the customer chooses when, where, and how to send it back.
The cheapest reverse last mile is the one that never moves a parcel. Refund-and-keep policies, where low-value or hard-to-resell items are refunded without requiring a physical return, save more money than any routing optimization. Above that threshold, drop-off networks at retailers like CVS, Walgreens, Staples, and Kohl’s are cheaper than scheduled home pickups, and they shift the labor cost to the customer who is happy to swap a stamp for a five minute errand.
The least visible cost is restocking. A parcel that comes back, gets inspected, repackaged, and put back into inventory has consumed labor that does not show up in the carrier invoice. Brands that track fully-loaded return cost rather than just the return shipping label tend to find their effective return rate hits margin harder than they assumed.
Why returns shape the outbound design
Designing the outbound last mile without thinking about returns is a common trap. If you commit to ultra-fast delivery on apparel, you also commit to a high return rate as customers buy multiple sizes and send back what does not fit. If your packaging is single-use and not designed to be reused for returns, you push customers toward inconvenience and resentment. The outbound and reverse last mile should be designed together, not by two different teams in two different quarters.
How carriers actually price the last mile
Last-mile pricing looks opaque because it is. National carrier rate sheets combine base rates, dimensional weight surcharges, zone-based fees, residential surcharges, fuel surcharges, peak season surcharges, and a long tail of accessorials. Two retailers shipping identical parcels through the same carrier can pay rates that differ by 40 percent depending on negotiated discounts, volume commitments, and account history.
The actual unit of pricing inside the carrier is something like cost-to-serve per stop. The carrier estimates how much it costs them to put a parcel of a given profile at an address in a given zone, then layers on margin and the discount you negotiated. When you negotiate, you are not really negotiating a discount; you are negotiating which parts of their cost-to-serve they absorb versus pass through. That framing leads to better contracts than the usual “we want a bigger percentage off the rate card.”
The metrics that actually matter
Most last-mile dashboards drown in vanity numbers. The ones that move the business are surprisingly few.
- On-time delivery rate against your published promise, not against the carrier’s promise.
- First-attempt success rate by carrier and zip code cluster.
- Cost per delivered parcel, fully loaded with packaging, label generation, and exception cost.
- Exception rate per 1,000 parcels, with root-cause tagging (address, access, refused, damaged).
- Customer-reported delivery satisfaction from a one-question post-delivery survey.
If you can hold five clean numbers on one screen and review them weekly, you are already ahead of most operators. The fancy dashboards come later, and only if the simple ones leave specific questions unanswered.
FAQ
What does “last mile delivery” actually include?
It covers the entire final operational leg from the local sort hub to the customer’s door, locker, or curb. That includes the sort, route building, vehicle loading, driver execution, proof of delivery, and any exception handling. It does not include the long-haul movement from a national sort center to the local hub.
Why is the last mile so much more expensive than the rest of shipping?
Long-haul movements spread fuel, equipment, and labor cost across thousands of parcels. The last mile dedicates a driver and a vehicle to one route, so cost per parcel is much higher. Industry data consistently puts the last mile at 40 to 60 percent of total parcel shipping cost.
Should a small DTC brand build its own last-mile operation?
Almost never. Below roughly 100,000 orders a month in a metro, density does not justify owned routes. The right move is to use national and regional carriers and put the engineering effort into address quality, promise accuracy, and exception handling instead.
Is same-day delivery worth the cost?
Only in narrow categories: groceries, pharmacy, baby supplies, pet food, and a handful of consumables where urgency drives the purchase. For apparel, home, and most electronics, a clear three-day promise outperforms an unreliable same-day option on repeat purchase rates.
What is the single highest-leverage change I can make this quarter?
Add address validation at checkout and instrument first-attempt success rate by carrier and zip code. Together they typically lift on-time delivery by 2 to 5 percentage points and cut exception cost meaningfully, with no carrier renegotiation required.
How do regional carriers compare to UPS and FedEx?
In their core regions, regional carriers often beat national rates by 15 to 30 percent and match or exceed service quality. The trade off is integration cost: tracking, label generation, and exception flows take engineering time. They are worth the effort once you have meaningful volume concentrated in their footprint.
How should I think about last-mile sustainability?
Consolidation and density do most of the environmental work that route-level electrification gets credit for. Combining orders into fewer stops, raising first-attempt success, and shrinking packaging cut emissions faster than swapping the vehicle fleet, and they cut cost at the same time.
Where can I learn the rest of the retail logistics picture?
Last mile is one slice of the chain. Start with our modern retail logistics guide for the pillar view, then come back to this article and the sibling pieces in the last-mile cluster for the deep dives.