Designing for High Responsiveness: When Omnichannel Networks Become a Demand Strategy

Consumers now expect their online orders to arrive fast, ideally the next day. Meeting that expectation is not just a logistics challenge. It requires retailers to rethink where they position their inventory, which facilities they open, and how they design their entire distribution network. The investment is significant. The question is whether it always pays off. In our research, we sought to establish under what conditions building a more responsive omnichannel network creates value, and to develop a framework that jointly optimizes delivery promises and the network structure.

Service offer shifts demand

Most supply chain models assume that demand exists and is known in advance. The network’s mission is to fulfill it efficiently. The service offer is a constraint to be met, not a decision.

Our research challenges this view. The service offer not only serves existing online demand. It influences whether customers choose to buy online in the first place. A shopper who would typically visit a store may shift to the online channel if next-day delivery is offered. Conversely, a long service offer quietly pushes customers back toward physical stores. This is the central idea our work is built around: demand is not a fixed input to the network design problem. It is a variable that the network itself helps determine.

We posed a different question: not where to position facilities given a predetermined delivery service offer, but which delivery service offer to make, zone by zone, to maximize profit, and what network structure that implies. Answering it requires accounting for a component that has received little attention: the role of brick-and-mortar stores in omnichannel fulfillment.

Stores as a strategic logistics asset

Physical stores were originally designed for shopping. Increasingly, they are becoming fulfillment platforms for online orders, what we call ‘omnistores’. This shift is not just operational. It is strategic.

Unlike dedicated urban fulfillment centers (UFCs), which are automated facilities requiring significant capital investment, omnistores leverage what already exists: proximity to customers, established inventory, staff, and supplier relationships. They serve both channels simultaneously; customers can still shop in person while the store’s backroom handles online order preparation and dispatch.

This dual role makes omnistores particularly attractive. Retailers do not need to choose between offline and online customers. The omnistore serves both. And because it is already embedded in the urban area, it reduces the distance between inventory and the customer’s door, directly supporting higher responsiveness. Our research shows that, alongside urban fulfillment centers, omnistores are fundamental to building a responsive and profitable omnichannel network.

Building the model

Our research was motivated by a collaboration with a major cosmetics retailer in France that operates from a single central distribution center and has limited responsiveness.

We built a model that jointly optimizes two decisions: the selection of a delivery service offer per demand zone, next-day, two-day, or longer, and the number and location of urban fulfillment centers (UFCs) and omnistores to deploy. These two decisions are deeply intertwined. The delivery service offer shapes how many customers choose to buy online, and that volume determines whether a given network configuration is profitable.

To capture this, we embedded a customer choice model that estimates how demand shifts between online and offline channels in response to different delivery service offerings. It is applied to 34799 zip code-level areas, aggregated into 94 demand zones across France. We then optimized the network under demand uncertainty, accounting for inventory, transportation, and facility setup costs.

The key insight is that customers do not respond uniformly to service offerings. In some regions, shoppers are highly sensitive to how quickly their orders arrive and will shift online if a more responsive service offer is proposed. In others, that sensitivity is lower, making accelerated delivery difficult to justify economically. The objective, therefore, is not to maximize responsiveness everywhere, but to provide the right responsiveness in the right place.

When does higher responsiveness pay?

The first result is that jointly deploying UFCs and omnistores, with a flexible, zone-by-zone service offer, can increase profits by up to 8% compared to relying solely on a central distribution center. That gain is driven by a channel shift: greater responsiveness draws customers away from physical stores and toward online purchasing. But this advantage holds only when the online channel is more profitable per unit than the offline channel. When it is not, the network naturally pulls back from next-day delivery coverage and prioritizes in-store sales.

The delivery landscape is redrawn just as sharply. In some configurations, up to 60% of demand zones shifted from three-day to next-day delivery once a nearby omnistore or urban fulfillment center was introduced. The network did not just become more responsive. It allocated responsiveness to where it could generate the most value.

Across all configurations, the model consistently preferred omnistores over dedicated UFCs as they serve both channels, provide broader geographic coverage, and cost less to deploy. UFCs become more attractive when omnistore capacity is insufficient to meet tight delivery windows.

The future of retail

The most counterintuitive finding: imposing next-day delivery universally across all regions reduced profit by more than 8% compared to a flexible, zone-adapted approach.

The cost of expanding the network to guarantee one-day delivery everywhere outweighs the revenue gains, particularly in zones where customers are less sensitive to delivery service offerings.

The right delivery service offering is not the most responsive. It is the one calibrated to where customers will respond, and where the economics support the investment. Responsiveness is a deliberate choice, made zone by zone and channel by channel.

For omnichannel retailers, the implication is clear: network design and service strategy cannot be decided separately. The delivery service offer, the customer, and the network infrastructure form a single system, and optimizing that system is where the competitive advantage lies. The most valuable capability for an omnichannel retailer is not to build the most responsive network, but to know where responsiveness creates value and where it does not.