What does an optimal e-commerce network look like? Is it the network with the fewest facilities, the fastest delivery promise, or the lowest transportation cost? In practice, the answer is rarely that simple.

Customers often ask us to optimize a specific part of their network rather than redesign it entirely. In one recent engagement, a large e-commerce shipper wanted to identify the optimal carrier mix while keeping its fulfillment footprint and service commitments unchanged. The model showed meaningful transportation savings by reallocating package volume among carriers. However, the lowest-cost solution on paper was not necessarily the best business decision. To determine the right path forward, we evaluated how those changes would affect contract commitments, incentive structures, penalty provisions, facility operations, and technology requirements.

That experience illustrates a broader reality: network design is not about optimizing one variable. Facility location, inventory positioning, carrier strategy, customer expectations, contracts, and operational execution are all connected. Improving one area often creates trade-offs elsewhere. The most effective networks are deliberately designed around customer demand, operational feasibility, and long-term business objectives.

What Does "Optimal" Really Mean?

An optimal e-commerce network is not defined by a single metric. It balances cost efficiency, service performance, customer experience, and resilience. A decentralized network may reduce transit times and improve delivery speed, but it can also increase inventory carrying costs and operational complexity. A centralized network may be simpler and less expensive to operate, but it may struggle to support aggressive delivery expectations.

The right answer depends on the shipper’s product mix, customer base, growth plans, and service promise. Optimal design comes down to alignment with business strategy.

Core Components of Network Design

Facility strategy forms the backbone of the network. Facility location decisions are among the most impactful choices a shipper can make, and among the most difficult to reverse.

Organizations must determine how many fulfillment centers to operate, where to position them, and whether alternative strategies can achieve similar benefits without adding new buildings.

As delivery expectations shift toward next-day and two-day service, many shippers are looking beyond traditional fulfillment center expansion. Zone-skipping, direct injection into carrier facilities near major customer clusters, and ship-from-store or ship-from-branch programs can reduce parcel zones and improve delivery speed while leveraging existing infrastructure. The optimal footprint often blends national coverage, targeted regional capacity, and strategic use of existing assets.

Inventory positioning is equally important. Centralized inventory can reduce working capital requirements and simplify replenishment, but it may limit speed. Distributed inventory can improve service levels, but it increases safety stock requirements and the risk of imbalance. For example, an industrial distributor may carry hundreds of thousands of SKUs, from frequently ordered fasteners and safety supplies to specialized replacement parts ordered only a few times per year.

Stocking every item in every facility would create excessive inventory cost and obsolescence risk. Instead, the distributor may position high-demand products across regional facilities while centralizing slower-moving items. This tactic improves service where demand is strongest without duplicating the entire assortment.

Transportation and carrier strategy connect the network. Parcel costs are heavily influenced by zone, weight, dimensional pricing, residential surcharges, and delivery area surcharges.

Reducing average zone can lower cost per package, but it may require additional facilities, new carrier relationships, or different induction strategies. Many large shippers are also moving from single-carrier strategies toward diversified models that match carriers to the shipments they handle best by geography, weight, service level, or cost profile.

Key Design Drivers

Strong network design begins with demand. Understanding where orders originate, and how those patterns change over time, directly affects how quickly and cost-effectively products can be delivered. Geographic concentration may justify placing inventory closer to major customer populations, while seasonal peaks or regional growth can create capacity constraints in some markets and excess capacity in others. Aligning capacity with demand helps control transportation costs, maintain service levels, and avoid unnecessary facility or labor investments.

Service requirements are another critical input. Customer expectations continue to rise, with next-day and two-day delivery becoming common in many markets. The 2026 GMT Benchmark Report found that 74% of online shoppers expect delivery within two days and 92% consider delivery windows when making purchasing decisions. Different channels, including direct-to-consumer, marketplace, and wholesale, may also require different service levels. These expectations influence where inventory should be positioned, how fulfillment capacity should be deployed, and which carriers should be used.

Cost and operational considerations must be evaluated together. Transportation savings can be meaningful, but they should not be considered in isolation. Labor costs, facility lease rates, utilities, taxes, inventory carrying costs, implementation timelines, carrier capacity, and facility readiness all affect the value of a proposed network.

A model may recommend a theoretically optimal location or carrier strategy. However, if labor is scarce, real estate is unavailable, or the carrier lacks pickup coverage in key markets, the recommendation may not be practical. The best design is not necessarily the lowest-cost design; it is the one that delivers the desired service levels and can be executed sustainably.

Technology and Execution

Technology has expanded what shippers can model and execute. Optimization tools can evaluate scenarios involving facility locations, inventory placement, carrier allocation, and service commitments. Integrated Order Management Systems (OMS), Warehouse Management Systems (WMS), Transportation Management Systems (TMS), and Parcel Shipping Systems (PSS) help execute those decisions at the shipment level.

For example, a retailer with inventory in both distribution centers and store locations can evaluate inventory availability, transportation cost, carrier transit time, labor capacity, and the delivery promise before selecting the best fulfillment point. One order may ship from a store to improve speed, while another may ship from a distribution center because store inventory or labor is constrained.

This execution layer is critical because large, complex shippers often struggle to convert a theoretically optimal design into a practical operating model. Common pitfalls include over-prioritizing cost, designing only for current conditions, ignoring demand variability, making decisions in silos, and underestimating implementation complexity.

A Framework for Effective Network Optimization

Successful optimization initiatives typically follow a structured but iterative process. First, define objectives across cost, service, customer experience, and long-term business goals. Next, gather and validate data on order patterns, shipment characteristics, inventory flows, constraints, and cost drivers. Then, model scenarios that include facility locations, inventory strategies, carrier allocation, and service commitments.

While the process itself is straightforward, the quality of the analysis ultimately depends on the quality of the data supporting it. Before evaluating trade-offs, organizations must have confidence in the data underlying the analysis. While technology can accelerate analysis and enable more sophisticated modeling, it cannot overcome poor data quality. Inaccurate data or assumptions can produce recommendations that appear optimal in a model but fail in practice.

From there, assess trade-offs across transportation cost, inventory requirements, contractual obligations, service performance, and operational complexity. Build an implementation plan that prioritizes achievable opportunities while preserving flexibility. Finally, track performance through KPIs such as cost, transit time, carrier performance, service levels, and network utilization. Continuous measurement helps validate results, identify emerging issues, and refine the network as business conditions evolve.

Ultimately, the most effective networks are built on data, not assumptions.

Sam Sealey is the Sr. Manager of Analytics at Green Mountain Technology (GMT). In this role, Sam partners with clients and internal Green Mountain teams to drive excellence in analytics by focusing on strategic decision-making, fostering innovation, and optimizing operational processes. He has over 20 years of parcel experience working in various engineering, analytical, and management roles at UPS, Johnson & Johnson, and GMT.

Sam holds a Master of Science Degree in Industrial and Systems Engineering from Auburn University, a Bachelor of Science degree in General Science from Morehouse College, and a Bachelor of Science degree in Electrical Engineering from the Georgia Institute of Technology.

Request a copy of the 2026 GMT Benchmark Report at greenmt.com/resource/2026-green-mountain-benchmark-report/ and learn how shippers are adapting to higher customer expectations, changing carrier dynamics, and increasing operational complexity.

This article originally appeared in the July/August, 2026 issue of PARCEL.