What Actually Works for Sub-2-Hour Urban Delivery in US Markets

A Head of E-Commerce Operations at a US retailer reviews the latest proposal for adding sub-2-hour delivery to checkout. The pitch focuses on what is possible: micro-fulfilment, dense urban networks, and a convenience tier that can increase conversion. The customer expectation argument sounds reasonable. The competitor pressure feels real.

Then the operational question lands: what does sub-2-hour delivery actually mean in US markets — compared with what the marketing language implies — and which categories, geographies, and customer segments can support it profitably?

Sub-2-hour urban delivery means an order is accepted, picked, packed, routed, dispatched, and delivered within two hours or less inside a dense city or metro service area. In practice, the model depends on short delivery distances, accurate inventory, dispatch capacity, real-time route optimisation, curb access, and a delivery promise that reflects actual operational capacity.

That distinction matters. In 2026, sub-2-hour urban delivery in the US is real but specific. It is operationally feasible in selected categories, service areas, and customer segments. It is not a universal shift across all e-commerce. Treating it as universal creates delivery promises that fail at the level of routing capacity, pick-pack speed, driver availability, on-time delivery, and cost-to-serve. Treating it as a targeted capability can improve customer experience and conversion where the economics support the SLA.

For Heads of E-Commerce Operations, VPs of E-Commerce, and Directors of Online Operations, this is a 2026 operating framework for sub-2-hour urban delivery US markets: what it is, how it affects the customer experience, where it fits, which models work, and how to evaluate whether the promise belongs in checkout.

According to McKinsey & Company research on US e-commerce delivery economics and CSCMP State of Logistics Report research on the US last-mile operating environment, sub-2-hour delivery economics work in specific category and network configurations and fail in others. The customer experience value is highest when the promise is offered only where fulfilment, routing, and dispatch can consistently meet it.

What Sub-2-Hour Urban Delivery Actually Is in US Markets

The honest starting point: sub-2-hour delivery in US markets in 2026 is not “15-minute residential micro-fulfilment” at scale. That category has appeared more often in investor decks and marketing narratives than in durable US operating models. Multiple well-funded quick-commerce attempts — Getir exited the US in 2023, Buyk shut down in 2022, Jokr exited in 2022, and Gorillas merged into Getir before that exit — showed that sub-15-minute residential dark-store economics generally do not work under US market conditions.

What does work is more specific.

Commercial dark-store quick commerce

Commercial dark-store quick commerce operates from commercial real estate in residential-adjacent areas. Gopuff and regional players have used this model in selected metros, typically with 15–30 minute windows rather than sub-15-minute promises.

The model depends on proximity to demand, tight delivery radii, controlled assortment, fast pick-pack processes, and disciplined service-area boundaries. For operators planning this model, hyperlocal routing for urban same-day delivery becomes central because small geography changes can alter SLA feasibility and cost per drop.

In-store micro-fulfilment for grocery

In-store micro-fulfilment uses existing retail footprints and automation to reduce picking time and increase store throughput. Walmart, Kroger, and Albertsons have worked with automation providers such as Takeoff, Fabric, and AutoStore to support 1–2 hour fulfilment from stores.

The operating advantage is that stores already sit close to demand. The constraint is that stores were not originally designed as high-velocity fulfilment nodes. Inventory accuracy, picker productivity, staging discipline, substitutions, cold-chain handling, and driver dwell time determine whether the model can meet the SLA consistently.

Restaurant delivery batching

Restaurant delivery operates through marketplace and gig courier networks such as DoorDash, Uber Eats, and Grubhub. These networks rely on dispatch automation, courier availability, batching logic, and real-time reallocation to maintain sub-1-hour windows.

The complexity is not only the delivery route. Merchant preparation time, handoff delays, courier arrival timing, order batching, traffic, and customer availability all affect performance.

Same-day and 2-hour delivery from urban-adjacent fulfilment centres

Same-day and 2-hour delivery from urban-adjacent fulfilment centres uses nodes close enough to dense demand to support faster SLAs. Amazon Same-Day, Whole Foods, Target Shipt, and similar models depend on inventory placement, service-area logic, and routing capacity.

This model often sits between ultra-fast quick commerce and traditional parcel delivery. It is closer to same-day and slot-based delivery models than to sub-15-minute residential quick commerce.

Each model has different economics, operational risks, and customer expectations. For e-commerce leaders, the first decision is not “should we offer fast delivery?” It is “which fast-delivery model fits our category, network, and margin structure?”

Also Read: Beyond In-House Fleet: When Should Enterprise Shippers Move to Multi-Carrier Orchestration?

Same-day vs. 2-hour delivery: the operational difference

Dimension	Same-day delivery	Sub-2-hour urban delivery
Customer promise	Delivered later the same day, often by slot	Delivered within two hours or less
Primary constraint	Cut-off time, linehaul timing, delivery capacity	Pick-pack speed, dispatch latency, routing feasibility, driver supply
Best-fit categories	Broader e-commerce, apparel, electronics, personal care, grocery	Grocery urgency, prepared food, convenience, pharmacy-adjacent, high-need replenishment
Fulfilment model	Store, fulfilment centre, 3PL, parcel carrier, local courier	Dark store, store-based fulfilment, micro-fulfilment, gig courier, owned or hybrid fleet
Routing challenge	Route consolidation and slot adherence	Real-time assignment, ETA accuracy, batching, exception recovery
Risk profile	Late delivery, missed cut-off, customer dissatisfaction	SLA breach, support spike, refund/appeasement cost, failed promise perception

How Sub-2-Hour Delivery Affects the E-Commerce Experience

Sub-2-hour delivery changes the full e-commerce experience. It is not only a delivery execution problem. It affects checkout, conversion, basket composition, customer support, post-purchase visibility, and returns.

Promise design

The customer-facing commitment sets the operating standard. A 2-hour promise at checkout becomes an SLA across picking, packing, dispatch, routing, driver assignment, handoff, and delivery confirmation. If serviceability logic is weak, customers see promises that the operation cannot fulfil.

The strongest operating models connect checkout promises to inventory position, store or node capacity, dispatch availability, and route feasibility before the promise is displayed. This is where delivery experience optimization in the last mile becomes more than a post-purchase capability. It starts before the customer places the order.

Conversion impact

Sub-2-hour availability can drive conversion in specific categories: grocery emergency replenishment, prepared food, convenience essentials, and some health and wellness use cases. In categories where urgency is weaker, the conversion benefit is less measurable and may not justify the higher delivery cost.

Market data also reinforces that fast delivery is concentrated in e-commerce and lighter shipments. Mordor Intelligence reported that e-commerce accounted for 53.05% of the United States same-day delivery market in 2025, while 75.88% of US same-day delivery shipments were lightweight parcels. That matters because sub-2-hour economics are easier to defend when order size, handling complexity, and delivery density align.

Basket size dynamics

Faster delivery often correlates with smaller baskets. That changes cost-to-serve. A low-value basket delivered inside two hours can quickly become unprofitable if pick cost, driver cost, failed delivery risk, and support cost are not controlled.

Retailers need to evaluate whether the delivery fee, basket margin, loyalty value, or retention impact covers the incremental cost. If the basket cannot support the SLA economically, the promise should be limited by SKU, zone, customer tier, or minimum order value.

Post-purchase experience

ETA accuracy matters more when the promise window is short. A 15-minute ETA drift on a standard delivery may be tolerable. On a 2-hour SLA, it can trigger customer contacts, refunds, appeasements, or churn. Real-time tracking, exception alerts, and proactive notifications are not optional.

The support burden can also compound quickly. When customers repeatedly ask “Where is my order?”, the hidden cost of WISMO in last-mile delivery becomes part of the unit economics, not a separate customer service issue.

Customer expectation spillover

Once customers see a sub-2-hour option, they compare other delivery tiers against it. That means fast delivery can raise expectations across the entire delivery portfolio, even where standard or next-day delivery remains the operationally appropriate choice.

The implication for e-commerce operations: do not launch sub-2-hour delivery as a broad brand promise unless the operation can explain why it appears in some contexts and not others.

Return economics

Sub-2-hour delivery creates different return patterns from standard delivery, particularly in grocery and convenience categories. The operational cost is not limited to the forward journey; reverse logistics, failed handoffs, substitutions, refunds, and support contacts must be included in the economic model.

From a Locus point of view, the core issue is promise-to-execution alignment. Retailers should not optimise routing in isolation after a promise has already been made. They need serviceability, inventory position, dispatch capacity, fleet availability, and route optimisation to inform the promise before it reaches the customer.

The Suitability Profile That Determines Fit

Sub-2-hour delivery is a capability for specific intersections of category, geography, and customer segment.

Category fit

Grocery, prepared food, convenience essentials, emergency replenishment, pharmacy-adjacent items, and some health and wellness categories typically fit the use case. The customer need is immediate, the basket is often mission-driven, and the convenience premium can be defensible.

Most apparel, most electronics, most home goods, most big-and-bulky items, and much general merchandise typically do not fit. The urgency is lower, fulfilment complexity can be higher, and standard 1–2 day or same-day delivery is often more economically appropriate.

That does not mean these categories can never use fast delivery. It means operators should test the promise by SKU economics, delivery radius, inventory position, and customer willingness to pay rather than applying the same promise to the full catalogue.

Geographic fit

Dense urban areas with sufficient order density can support dark-store or in-store fulfilment economics. The route density exists to reduce miles per drop, improve batching, and keep driver utilisation within range.

Suburban zones with moderate density may support 2-hour or same-day delivery from urban-adjacent fulfilment centres, but they typically do not support sub-15-minute dark-store economics. Rural zones generally do not support sub-2-hour delivery on standard economics because travel time, stop density, and labour utilisation do not align.

The geography skew is visible in same-day delivery market data. Mordor Intelligence reported that three US megaregions — California metro clusters, the Northeast Corridor, and Texas urban hubs — captured 64.72% of same-day delivery market value in 2025. That concentration underscores why fast-delivery economics are not evenly distributed across the US.

Customer base fit

Customers paying for a convenience tier, premium loyalty members, and customers in high-urgency use cases can justify the higher operating cost. A universal sub-2-hour promise to the general customer base usually fails the unit economics test.

The operational test is simple: can the business meet the SLA repeatedly, at zone level, without hiding the cost through unsustainable subsidies? If not, the promise should not be exposed universally.

---

What Actually Works Operationally

The models that work in US sub-2-hour urban delivery share common operating characteristics: proximity to demand, controlled inventory, dispatch visibility, service-area discipline, and flexible workforce orchestration.

Commercial dark stores in residential-adjacent commercial real estate

Gopuff and regional players have operated this model in selected US metros. The model depends on demand density, tight assortment control, fast pick-pack processes, and dispatch rules that prevent delivery radii from expanding beyond the SLA.

For dense urban quick-commerce planning, operators often use tight catchment areas around high-demand nodes rather than broad metro coverage. The smaller the promise window, the less tolerance the network has for demand leakage, traffic delay, or driver repositioning time.

In-store micro-fulfilment for grocery

Walmart Spark, Kroger, and Albertsons use existing store footprints and automation to improve fulfilment speed. The critical operating questions are store inventory accuracy, picking productivity, staging discipline, driver dwell time, and route sequencing for fresh and time-sensitive goods.

Available industry estimates point to why micro-fulfilment is attractive for fast delivery. FleetRabbit reported that micro-fulfilment operations supporting same-day or sub-2-hour windows achieved 35–40% last-mile cost reduction versus traditional store-based fulfilment in 2025, and lifted picking efficiency from 15–25 orders per hour to 60–80 orders per hour for fast delivery missions.

These numbers should not be treated as universal benchmarks. They show the direction of the operational advantage when inventory, picking, staging, dispatch, and routing are tightly coordinated.

Restaurant delivery batching

DoorDash, Uber Eats, and Grubhub operate sub-1-hour windows through gig courier networks. Success depends on real-time driver supply, merchant preparation time, batching constraints, and the ability to reassign orders when ETAs slip.

The restaurant model shows both the power and limitation of crowdsourced delivery. It can scale capacity quickly, but it also depends on marketplace liquidity, courier incentives, merchant readiness, and customer location accuracy.

Same-day fulfilment from urban-adjacent centres

Amazon Same-Day, Whole Foods, Target Shipt, and Instacart Priority Delivery use fulfilment nodes close enough to demand to support faster windows. The model requires accurate service-area logic, capacity-aware routing, and delivery orchestration across owned, gig, or third-party capacity.

Also Read: Why Most Driver Retention Strategies Miss the Operational Layer – Locus

The common operational denominator is not “speed at any cost”. It is controlled execution:

• Commercial real estate positioned close to demand, rather than residential premises that create zoning and operating friction.
• AI route optimization that reflects delivery windows, traffic, capacity, driver skills, order priority, and batching constraints.
• Dispatch automation that assigns orders to the right fleet, courier, or partner based on SLA risk and cost-to-serve. This is where auto-dispatch logistics software becomes critical for protecting short delivery windows.
• Workforce fit across owned fleets, W-2 drivers, gig couriers, 3PLs, or hybrid models.
• Real-time exception management when fulfilment delays, courier shortages, address issues, or traffic disruption put the 2-hour SLA at risk.
• Integration with the e-commerce experience across checkout promises, post-purchase tracking, customer communication, returns, and support workflows.

SmartRoutes reported that last-mile delivery costs represented more than 50% of total shipping expenses for many logistics operations by 2025. That cost structure makes precise routing, batching, dwell-time control, and failed-delivery reduction essential for any urban 2-hour SLA.

Apartment package infrastructure such as Luxer One, Amazon Hub, and Package Concierge operates as receiving infrastructure. It improves the last-metre handoff, but it does not solve fulfilment, routing, or dispatch capacity for sub-2-hour delivery.

---

The Urban Constraints That Can Break a 2-Hour SLA

Sub-2-hour delivery is shaped by city-level constraints that rarely appear in checkout messaging.

Curb access

Delivery speed depends on the ability to stop, unload, complete handoff, and leave without excessive dwell time. In dense areas, curb competition from passenger vehicles, rideshare, buses, bikes, and other delivery fleets can create hidden delay.

Loading zones

Dedicated loading zones can improve delivery reliability when they are available, enforced, and located near actual demand. Without them, drivers lose time circling, double-parking, or walking longer distances from informal stops.

Off-peak delivery

Off-peak delivery can reduce congestion and improve route reliability for some categories. It works better for business deliveries, replenishment, and unattended handoffs than for customer-dependent residential deliveries that require presence or age verification.

Smaller vehicles and cargo bikes

Smaller electric vehicles, cargo bikes, and consolidated pickup models can help in dense urban zones where congestion and parking constraints reduce van productivity. The Urban Freight Lab identified smaller electric delivery vehicles and consolidated pickup models as key levers for US cities and carriers seeking zero-emission urban delivery while maintaining service performance.

---

The E-Commerce Operations Evaluation Framework

For Heads of E-Commerce Operations evaluating sub-2-hour delivery as a customer-facing capability in 2026, seven dimensions matter more than broad technology claims.

1. Category suitability assessment

Which categories in the assortment have urgency, margin, fulfilment speed, and substitution logic that support sub-2-hour delivery?

Operators should segment SKUs by urgency, gross margin, pick complexity, temperature requirement, return risk, and substitution tolerance. A product may be attractive for fast delivery from a customer standpoint but unattractive operationally if it requires complex handling or has low contribution margin.

2. Geographic fit assessment

Which markets have the demand density, fulfilment nodes, driver supply, and service-area shape required to hit the SLA without excessive cost per drop?

The answer should be evaluated at zone level. A citywide launch can hide the fact that only specific ZIP codes, store catchments, or neighbourhood clusters can support consistent 2-hour delivery.

3. Customer base fit assessment

Which segments value the speed enough to justify the operational cost — premium loyalty customers, convenience-tier customers, emergency replenishment buyers, or high-frequency urban households?

Fast delivery should not be evaluated only as a logistics cost. It should also be evaluated against retention, share of wallet, subscription value, and strategic account value where applicable.

4. Promise design decisions

What should be shown at checkout by SKU, inventory position, location, customer tier, cut-off time, and available capacity? Promise design should be dynamic, not a blanket banner.

This is where capacity planning for omnichannel retailers becomes essential. If capacity is not reflected in the promise, the operation can create demand that the fulfilment and delivery network cannot absorb.

5. Operational partnership decisions

Which model fits the operation: in-house fleet, 3PL, gig courier, marketplace partner, or hybrid? The right model depends on service level, labour availability, brand control, and cost-to-serve.

Owned fleets provide more control. Gig and crowdsourced networks provide flexibility. 3PLs can add capacity and coverage. Hybrid models are often strongest when they allocate demand by SLA risk, cost, geography, and customer value.

FleetRabbit reported that 40% of urban last-mile deliveries in 2025 were handled via crowdsourced networks such as Amazon Flex, DoorDash Drive, and Uber Direct. For e-commerce leaders, the takeaway is not that crowdsourcing is universally better. It is that flexible capacity has become part of the operating toolkit for dense urban delivery.

6. Integration with the broader e-commerce experience

How will the offer affect conversion analytics, basket mix, post-purchase communication, substitutions, returns, refunds, and customer service load?

Fast delivery is not successful if logistics hits the SLA but customer support costs spike, substitution dissatisfaction rises, or customers receive inaccurate availability promises at checkout.

7. Unit economics evaluation

Economics must be tested by category, geography, customer segment, and delivery zone — not averaged across the network. A metro-level average can hide unprofitable zones, poor batching performance, high failed-delivery rates, and SLA leakage.

Per Pitney Bowes global parcel research with North America coverage, operators that succeed with sub-2-hour delivery treat it as a targeted capability rather than a universal feature.

Also Read: How Routing Decisions Shape Dark Store Network Economics for North American Retailers

KPI Checklist for Sub-2-Hour Urban Delivery

Operators should measure sub-2-hour delivery with metrics that expose both speed and economic quality.

KPI	What it shows	Why it matters
On-time delivery rate	Share of orders delivered within the promised 2-hour window	Core SLA health metric
Cost per delivery	Total delivery cost per completed order	Shows whether speed is economically sustainable
Cost per successful first attempt	Cost adjusted for failed delivery and reattempts	Prevents undercounting failure cost
ETA accuracy	Difference between projected and actual arrival time	Critical for customer trust and support reduction
Pick-pack cycle time	Time from order release to ready-for-dispatch	Reveals fulfilment bottlenecks
Dispatch latency	Time from ready-for-dispatch to driver assignment	Exposes courier capacity and allocation issues
Driver dwell time	Time spent waiting at store, pickup, curb, or customer handoff	Major hidden constraint in dense cities
Stops per route or batch	Delivery density per courier run	Indicates batching quality and route productivity
Failed delivery rate	Orders not completed on first attempt	Directly affects cost, customer experience, and SLA performance
Support contact rate	Customer contacts per order	Captures WISMO, ETA anxiety, refunds, and exception load

Benefits of Sub-2-Hour Urban Delivery When the Fit Is Right

Sub-2-hour delivery can create real value, but only when the operating model matches the market.

Higher conversion in urgent categories

When customers need groceries, prepared food, medicine-adjacent products, essentials, or time-sensitive replenishment, speed can remove purchase friction. The benefit is strongest when the customer’s intent is immediate and the product is already positioned close to demand.

Stronger loyalty and convenience differentiation

Fast delivery can support premium loyalty tiers and convenience subscriptions. It gives high-frequency customers a reason to consolidate spend with one retailer or marketplace, especially when reliability is consistent.

Better use of urban inventory

Store-based fulfilment, micro-fulfilment, and dark-store models can turn urban inventory into a faster demand-response asset. This is valuable when inventory accuracy is high and store operations can absorb fulfilment work without hurting in-store customers.

More precise service-level segmentation

Sub-2-hour delivery does not need to be universal to be strategic. Retailers can offer it by SKU, customer tier, market, radius, time of day, capacity, or order value. That segmentation protects economics while still giving customers speed where it matters.

Improved operational learning

Fast-delivery operations expose weaknesses quickly: poor inventory accuracy, slow picking, weak dispatch logic, bad address data, driver dwell time, service-area overreach, and ETA inaccuracy. Retailers that measure these constraints can improve broader last-mile performance beyond the sub-2-hour tier.

---

Why Choose Locus for Sub-2-Hour Urban Delivery Execution

Sub-2-hour urban delivery is not solved by one capability. It requires connected decision-making across serviceability, routing, dispatch, tracking, exception management, and customer communication.

Locus helps logistics and e-commerce teams align promises with execution through:

• SLA-aware route optimisation that accounts for delivery windows, traffic, capacity, priority, and batching constraints.
• Automated dispatch and allocation across owned fleets, third-party carriers, gig capacity, and hybrid models.
• Dynamic serviceability logic to help teams expose fast delivery only where the operation can support it.
• Real-time visibility and exception control so teams can identify risk before a 2-hour promise fails.
• Customer communication workflows that reduce uncertainty and improve ETA transparency.
• Zone-level performance visibility so leaders can evaluate cost, SLA adherence, and operational leakage by geography.

For US e-commerce operators, the objective is not to promise speed everywhere. It is to make speed operationally credible where customers value it and where economics can support it.

Conclusion: Sub-2-Hour Delivery Is a Targeted Capability, Not a Universal Promise

Sub-2-hour urban delivery in US markets is real, operationally feasible, and increasingly expected in specific contexts. It is not universal, and it is not a revolution reshaping all e-commerce. Treating it as a universal promise creates cost and service failures. Treating it as a targeted capability for specific category, geography, and customer-segment intersections can produce customer experience and conversion benefits that justify the investment.

The strongest models combine micro-fulfilment, disciplined inventory placement, dynamic routing, flexible fleet capacity, and real-time exception management. Urban constraints such as curb access, loading zones, traffic volatility, and driver dwell time can determine whether the SLA is feasible.

The strategic question for US Heads of E-Commerce Operations in 2026 is: given that sub-2-hour delivery economics work in some operational and category configurations and fail in others, are we evaluating the decision based on honest category, geography, customer segment, SLA, and cost-to-serve fit — or are we accepting universal sub-2-hour positioning that will not survive contact with real operating economics?