Omnichannel retailing has become the new norm in the retail sector as companies reinvent their supply chains to offer customers multiple buying channels.
Although the omnichannel model is now relatively well established, a challenge that retailers continued to grapple with is how to reconfigure their last-mile supply networks (LMSNs) to achieve better alignment between delivery responsiveness, product variety, and convenience.
A study of LMSNs* using a multiple, in-depth case study approach to unpack different network configurations identified four types of LMSNs and helps the retail industry to understand how these different options perform. The research also proposes a six-step process for planning and developing LMSN models for omnichannel operations.
Quest for alignment
Over the last decade, pure-play e-retailers that offer consumers a wide assortment of products with low-priced or free direct postcode delivery have become a major force in retailing. Correspondingly, click-and-mortar retailers have sought to integrate their physical store and online fulfillment operations as a way to counter the threat of the pure-plays.
As traditional companies navigate this fast-evolving landscape, they must address the notion that conventional distribution center (DC)-to-store replenishment is incompatible with the online DC-to-consumer delivery approach. The former is based on full truckload delivery and bulk packaging to fixed store locations. This model enables retailers to capture economies of scale. The latter model involves high volumes of singles packing and unit delivery to end consumers, often with nonfixed locations. This option allows retailers to be more responsive and to customize to specific consumers’ requirements.
These conflicting realities make modern last-mile delivery more challenging. For example, retailers must accommodate small purchase quantities, erratic purchase frequencies, delivery window constraints, and the prospect of customers not being at home to receive orders. Many retailers develop omnichannel supply chains to meet these challenges. However, such solutions jeopardize operational efficiencies in areas such as order picking, and these issues have forced some retailers out of business.
Previous studies of these problems have focused on the role of technology in the changing retail business. While these studies provide valuable insights into how retailers can leverage technology and last-mile delivery innovations in an omnichannel world, they do not offer guidance on how to configure the critical last-mile delivery segment. Retailers need to establish effective LMSNs that align their marketing efforts with operational distribution activities to achieve success in the broader e-retailing environment.
The LMSN study’s* findings center on four network configurations encapsulated in a model called SHOP: Simple LMSN, Hyperlocal LMSN, One-Stop LMSN, and Protean LMSN. Here is a brief description of each one.
Slow delivery responsiveness and low product variety characterize Simple Networks. They aim to create cost efficiencies in order fulfillment (typically through the use of highly automated DCs and well-integrated processes) and emphasize convenience to consumers through the scheduling of well-suited home delivery windows. High levels of inventory aggregation and centralized logistics infrastructure are typical features of these networks. In the study, online grocer Ocado is attempting to provide this model.
Slow delivery responsiveness and high product variety are hallmarks of One-Stop LMSNs. Typically, the focus is on pooling inventory, sharing resources, and seeking tighter collaboration and involvement with partners to manage uncertainties and disruptions as well as to deliver multiple service options. Delivery flexibility and reliability are core logistics capabilities in this category, and the supporting networks are structured to handle moderate to high levels of inventory aggregation and centralized logistics infrastructure. One-Stop LMSN’s typically stock a core product range and source the remaining products from suppliers to extend their product ranges. In the study, UK supermarket Tesco Direct and Amazon illustrate this option.
Fast delivery responsiveness and low product variety distinguish these networks. A central goal is to offer responsive deliveries; Protean LMSNs can handle small and frequent orders via carefully curated product portfolios. Retailers in this category also tend to be highly responsive to customer demands by, for example, leveraging established physical stores to fulfill orders and provide pick-up/delivery services. Logistics capabilities are geared to providing high responsiveness in target markets, and stocks are typically stored and fulfilled in local inventory points resulting in highly decentralized network structures. UK high street retailer Argos is among the companies in the study that have embraced the Protean model.
Hyperlocal networks offer fast delivery responsiveness and high product variety. Logistically, they aim to provide fast delivery to consumers through networked ecosystems of actors connected via marketplaces or platforms. The retailers that choose this configuration leverage crowdsourcing to provide localized fulfillment of products stocked locally. They employ a highly decentralized network structure, and the short lead-times of deliveries require these retailers to rely on direct point-to-point distribution mechanisms. Two retailers in the study, Amazon, and Snapdeal, employ this configuration.
A six-step approach can be used to apply the SHOP framework, illustrated in the study by a hypothetical case that involves a retailer called Omnichannel Co. In the first step, the retailer constructs a product portfolio matrix, seeks to understand target consumer profiles, and reviews the business model. Subsequent steps involve evaluating the LMSN’s it has mapped, conducting a hotspot analysis, conducting an LMSN (re) configuration and options analysis, developing an action plan, and reviewing actions.
The LMSN SHOP framework created by the study provides retailers with a tool to help them succeed in an omnichannel economy. The tool can be used to map existing LMSNs and asses their performance and offers decision support in critical areas.
Moreover, the research underscores the notion that the best retailers will win in an omnichannel environment by evolving their LMSNs across the permeable boundaries of delivery responsiveness and product variety.
*The LMSN study was conducted by Stanley W. T. Lim, a doctoral candidate at the Institute for Manufacturing, University of Cambridge, UK, and Matthias Winkenbach, a research scientist at the MIT Center for Transportation & Logistics and director of the MIT Megacity Logistics Lab. For more information, contact Matthias Winkenbach at firstname.lastname@example.org. This piece is derived from an article written by the study authors and published in California Management Review 2019, Vol. 61(2) 132–154.