Out-of-Home Delivery in Europe: How Lockers and PUDO Became Default and What AI Routing Now Has to Solve

A consumer in Stockholm orders a pair of running shoes online. The default checkout option isn’t home delivery, it’s the InPost locker around the corner. Two days later, the shoes don’t fit. The return process is equally locker-centric: print a label at home, drop the parcel in a locker, walk away. The retailer behind the transaction runs operations across six European markets and now sees over half its volume in some Nordic and Benelux markets flowing through out-of-home (OOH) channels rather than home delivery.

This is European e-commerce reality in 2026. Out-of-home delivery (OOH) in Europe is not a fallback channel or a sustainability feature — it is the default consumer expectation across a substantial share of the continent, particularly Nordic markets, Benelux, Poland, and parts of DACH. The infrastructure has evolved through five distinct waves over twenty-five years, and current operational challenges sit at a layer most retailers were not architected for: multi-network integration, region-specific density variation, and a returns flow that mirrors forward volume but moves backwards.

As of late 2025 and early 2026, 46% of regular online shoppers in Europe prefer Out-of-Home (OOH) delivery options, marking a significant 15-point increase since 2019. This equates to tens of millions of users, with 1 million new users switching from traditional home delivery to OOH solutions annually

This is a guide for European Heads of E-Commerce Operations, Heads of Logistics, and VP Supply Chain leaders running OOH-relevant operations — covering how OOH delivery evolved, current operational challenges, and how AI-powered routing addresses them, with a concrete use case in locker returns operations.

According to the PostNord E-commerce in Europe Report, consumer preference for OOH delivery varies materially across European markets — strong in Nordic and Benelux, growing in DACH and Poland, more nascent in Southern and Eastern Europe.

Part 1: How European OOH Delivery Evolved

European OOH delivery moved from postal initiative to mainstream default through five distinct waves.

Wave 1 — Postal pioneers (1990s–2003). National postal carriers built early pickup networks leveraging existing post office and retail partner footprints. Deutsche Post launched the first modern automated parcel locker — Packstation — in Germany in 2003. PostNL, La Poste, and Royal Mail built parallel collection-point networks. The dominant model was postal-carrier-operated, integrated with national post infrastructure.

Wave 2 — Carrier expansion (2000s–2010s). Private carriers built OOH as competitive differentiation. DHL expanded ParcelShop and Packstation networks across Germany and beyond. DPD built Pickup networks across multiple EU markets. GLS scaled ParcelShop networks across DACH and Benelux. Bpost added cyclamen lockers across Belgium. By the late 2010s, every major European parcel carrier operated a meaningful OOH network alongside home delivery.

Wave 3 — Independent operators emerge (2010s–early 2020s). Operator-agnostic locker networks broke the carrier-bundled model. InPost — founded in Poland — became the European OOH champion, scaling rapidly across CEE and Western Europe. Quadient (formerly Neopost) and SwipBox built networks targeting Nordic and Western European markets. The independent-operator model decoupled locker access from carrier choice and enabled retailers to integrate OOH without committing to a single carrier.

Wave 4 — Covid mainstreaming (2020–2023). Pandemic restrictions accelerated consumer adoption of OOH dramatically. Contactless preference, restricted movement, and overloaded home delivery networks shifted consumer behaviour from “OOH as alternative” to “OOH as preference” in many markets. Network expansion accelerated to meet demand, and retailers integrated OOH as a primary checkout option rather than a niche alternative.

Wave 5 — Returns and maturation (2023–present). OOH delivery has matured into a returns channel as well as a forward channel. Locker returns address one of the most expensive operational problems in e-commerce — the 8–10% of volume that flows backward — by giving consumers low-friction return drop-off and operators consolidated collection points. Multi-carrier locker integration is now standard. The current frontier is operational architecture supporting OOH at scale.

Part 2: Five Current European OOH Operational Challenges

European retail and logistics leaders running OOH operations at scale face five recurring operational challenges.

Network fragmentation. No single OOH network dominates Europe. InPost, Quadient, SwipBox, carrier-operated networks (DHL, DPD, GLS, La Poste, PostNord), and retailer-operated networks (Amazon Hub, Zalando lockers) compete for share. Operators wanting full European coverage typically integrate with multiple networks — each with its own APIs, label formats, capacity reporting protocols, and SLA tiers.

Geographic density variation. OOH network density varies dramatically across European markets. Nordic countries, Benelux, and Poland have high density that supports OOH-as-default consumer expectations. DACH varies by region. Southern Europe — Italy, Spain, Greece — has lower density and more variable consumer preference. Rural areas across all markets are sparse. The implication is that OOH-first strategies don’t translate uniformly across European operations; what works in Stockholm fails in rural Sicily.

Returns-via-locker as a new operational pattern. Return volumes flowing through lockers can match a meaningful share of forward volume. The operational pattern is genuinely new for most retailers: consumers drop returns into lockers, lockers fill, collection drivers sweep them, returns flow back through reverse logistics to the retailer. Most operators are architected for forward delivery primarily, with returns handled as an exception flow rather than a peer to forward fulfillment.

Capacity management at peak. Locker capacity is physically fixed in the short term — you can’t add slots during Black Friday. Q4 and regional promotional periods regularly overflow capacity. Without real-time locker availability data flowing into routing decisions, retailers continue assigning parcels to lockers that can’t actually receive them. The result is downstream delivery failure: the carrier arrives, the locker is full, the parcel returns to depot, the consumer experience deteriorates.

Integration complexity. Multi-network integration multiplies API integration burden. Each network has different authentication mechanisms, different label requirements, different capacity reporting cadences, different returns flow protocols. Without an integration normalisation layer, operations teams manage each network individually, ongoing maintenance scales linearly with network count, and exception handling becomes operationally painful.

Part 3: How AI-Powered Routing and Dispatching Helps

Five capabilities at the routing and dispatch layer address the five operational challenges directly.

Multi-network orchestration. A routing engine that integrates with multiple OOH networks simultaneously — InPost, carrier networks, retailer networks — and selects per shipment which network and which specific locker or PUDO point fits the consumer location, network coverage, capacity availability, and cost. The orchestration layer handles network fragmentation as a routing decision rather than as an operations-team manual workflow.

Region-aware OOH-or-home decision logic. Routing engines that recognise OOH density variation across markets enable retailers to run different defaults by region — OOH-primary in Stockholm, home-delivery-primary in rural Sicily — without operational complexity at the team level. The same retailer can serve Nordic, Benelux, DACH, and Southern European markets with consistent operations and region-appropriate consumer experience.

First-mile pickup optimisation for returns. Routing engines that handle return collection sweeps as a first-class operational problem rather than as an afterthought. Collection frequency per locker tuned to capacity, volume patterns, and cost-to-serve. Multi-locker route planning for collection drivers solving the traveling-salesman problem across metropolitan locker networks.

Real-time capacity-aware routing. Locker capacity data flowing into routing decisions in real time. The system refuses to assign parcels to lockers that can’t receive them, diverting automatically to alternate lockers, alternate networks, or home delivery. This prevents the downstream delivery failure pattern that erodes consumer experience during peak periods.

API normalisation layer. A single integration interface abstracting away per-network differences. Operations teams work with unified data formats, capacity views, and exception flows. Changes in individual network APIs are absorbed at the integration layer rather than rippling through operational systems. Multi-network growth scales without proportional integration maintenance.

Part 4: Use Case — Locker Returns Sweep Optimisation

Consider a European retailer with 500 daily forward shipments across a metro area like Amsterdam or Warsaw, with returns running at category baseline through 150 to 200 locker locations across the metro.

Without intelligent routing: the retailer runs fixed daily collection sweeps. Some lockers fill between sweeps and reject incoming returns; other lockers are collected when nearly empty, wasting driver time. Collection routes are static, mixing high-volume and low-volume locations inefficiently. The operations team has limited visibility into collection cost per return parcel and limited ability to tune the operation.

With AI-powered routing: real-time locker capacity data informs collection priority — drivers visit nearly-full lockers first. Collection routes optimise for cost-to-serve per parcel, recalculated against current capacity state. Sweep frequency per locker tunes dynamically based on observed volume patterns rather than fixed schedules. Multi-locker routing solves the traveling-salesman problem across the metro, producing materially shorter collection routes for the same parcel volume. Operations teams see collection cost per return parcel as an operational metric they can manage rather than an aggregate they can only observe.

Routing platforms like Locus that handle multi-stop optimisation, real-time data integration, and exception management as core capabilities provide the operational substrate for this kind of OOH returns architecture.

OOH delivery in Europe has moved from feature to default in many markets. The infrastructure is mature, consumer expectations are set, and the operational reality is that retailers running OOH operations need architectural capabilities most legacy systems were not built for. The strategic question is not whether OOH matters — that decision was made by consumers — but whether the operational architecture supporting OOH delivery and returns is purpose-built for the network fragmentation, density variation, returns volume, and capacity dynamics European markets actually produce.

To learn more, visit locus.sh

FAQs

What is out-of-home (OOH) delivery in European e-commerce?
Out-of-home (OOH) delivery refers to delivery channels that route parcels to consumer-accessible pickup locations rather than to home addresses. The two primary categories are automated parcel lockers (InPost, DHL Packstation, Quadient, SwipBox, retailer-operated networks like Amazon Hub) and PUDO points (pick-up drop-off points operated through retail partners — convenience stores, post offices, kiosks). In European markets — particularly Nordic countries, Benelux, Poland, and parts of DACH — OOH delivery is the default consumer preference rather than a fallback channel, with consumers actively choosing OOH at checkout when offered.

How did European OOH delivery evolve to become a default consumer channel? European OOH delivery evolved through five distinct waves. Postal pioneers built early networks in the 1990s, with Deutsche Post launching the modern automated parcel locker in 2003. Carrier expansion through the 2000s and 2010s saw private carriers (DHL, DPD, GLS) build competitive networks. Independent operators emerged in the 2010s, with InPost becoming a European champion scaling across CEE and Western Europe alongside Quadient and SwipBox. Covid mainstreaming during 2020-2023 accelerated consumer adoption from alternative to preference. Current maturation focuses on returns-via-locker, multi-network integration, and operational architecture for OOH at scale.

What are the main operational challenges in European OOH delivery?
European operators running OOH operations face five recurring challenges. Network fragmentation across multiple non-dominant operators (InPost, Quadient, SwipBox, carrier networks, retailer networks) requiring multi-network integration for coverage. Geographic density variation between high-density Nordic, Benelux, and Polish markets and lower-density Southern European and rural markets, making OOH-first strategies non-uniform. Returns-via-locker as a new operational pattern most retailers are not architected for, with locker collection sweeps requiring distinct routing from forward delivery. Capacity management at peak periods when lockers physically overflow. And integration complexity from per-network API, label, capacity, and returns differences that scale linearly without normalisation layers.

How does AI-powered routing help with European OOH delivery operations?
AI-powered routing and dispatching addresses European OOH operational challenges through five capabilities. Multi-network orchestration that integrates with multiple OOH networks and selects per shipment which network and locker fits each consumer based on coverage, capacity, and cost. Region-aware OOH-or-home decision logic that recognises density variation and runs region-specific defaults. First-mile pickup optimisation for returns, handling collection sweep frequency, multi-locker route planning, and cost-to-serve optimisation. Real-time capacity-aware routing that refuses assignments to full lockers and diverts automatically. And API normalisation that abstracts away per-network differences so operations teams work with unified data and exception flows.

What is locker returns sweep optimisation?
Locker returns sweep optimisation is the operational practice of dynamically scheduling and routing collection drivers to gather returned parcels from networks of OOH lockers based on real-time capacity, volume patterns, and cost-to-serve. Without intelligent routing, retailers run fixed collection schedules that produce both overfull lockers (rejecting incoming returns) and underfull lockers (wasted collection cost). With AI-powered routing, collection frequency per locker tunes dynamically to actual fill patterns, multi-locker routes optimise the traveling-salesman problem across metro networks, and operations teams gain visibility into collection cost per return parcel. The outcome is materially lower cost per return parcel handled, fewer rejected drops, and better consumer experience on the return side.

Why does OOH delivery density vary across European markets?
OOH delivery density varies across European markets due to a combination of consumer behaviour, infrastructure investment, urban density, and historical postal network footprint. Nordic markets (Sweden, Denmark, Norway, Finland) have high density driven by digitally mature consumers and heavy infrastructure investment from PostNord and independent operators. Benelux markets benefit from high urban density and strong PostNL, Bpost, and InPost networks. Poland has unusually high density driven by InPost’s home-market scale. DACH markets have variable density with stronger coverage in Germany than Austria or Switzerland. Southern Europe (Italy, Spain, Greece) and many rural areas have lower density, with consumer preference still tilted toward home delivery in many cases. European OOH operations need to recognise this variation rather than assume uniform consumer behaviour across the continent.