Exploring opportunities and challenges for aerial drones at the interface of warehouse and last-mile delivery operations

The retail landscape is undergoing a significant transformation driven by technological advancements, which are reshaping fulfillment models and customer expectations (Ta et al., 2025). Demand for warehousing and last-mile delivery has grown substantially in recent years, driven by e-commerce, globalization and the increasing complexity of retail supply chain networks. With consumers expecting quick delivery, often within one or two hours (Snoeck et al., 2023; Zhang et al., 2019), warehouse managers need innovative technologies that can reduce excess inventory and ensure precision (Arkabaev et al., 2024). For retail warehousing, margins are low (Principal Logistics Technologies, 2015; Sides and Skelly, 2021); thus, maintaining a competitive edge requires minimizing operational costs and delivery time (Companik et al., 2018). As retail supply chains grow increasingly complex, third- and fourth-party logistics (3PL/4PL) providers have emerged as important enablers of retail fulfillment, managing warehousing and last-mile delivery. To ensure sustained competitive advantage, 3PL/4PL providers aim to remove bottlenecks from their supply chains. This puts operational efficiency at the core of their warehousing strategies, especially for providers seeking to eliminate inefficiencies and streamline internal logistics.

Aerial drones have great potential to increase operational efficiency and could fundamentally change warehouse operations and last-mile delivery (Filiopoulou et al., 2025). In practice, however, early efforts to use drones in retail logistics have been highly fragmented. This has resulted in an industry characterized by siloed trials rather than a cohesive network. Large firms such as IKEA have begun deploying autonomous drones for inventory counting in their warehouses (Maghazei et al., 2022). Likewise, DHL has pioneered drone deliveries with its Parcelcopter trials, collaborating with Microdrones to deliver pharmaceuticals in Germany (Yigit et al., 2024). While these examples demonstrate the ability of drones to enhance logistics performance on a larger scale, such implementations remain sparse. Moreover, initiatives such as Amazon's drone delivery arm Prime Air have faced significant challenges, including regulatory hurdles, technical limitations and high operational costs, leading to multiple site closures (Palmer, 2023). Therefore, a lack of clarity persists around the economic feasibility and real-world applicability of aerial drones across integrated supply chain processes (Merkert and Bushell, 2020).

Existing research on drones in logistics has tended to emphasize consumer perception (Merkert et al., 2022), technical performance and algorithmic optimization (e.g. Mourtzis et al., 2024), with comparatively limited attention to drones as viable business tools embedded in warehouse and delivery operations. As a result, scholarly understanding remains fragmented regarding how drones are perceived, evaluated and adopted by warehouse operators and drone service providers, particularly at the organizational interface between these actors. This represents a consequential gap, given the interdependence of warehousing and last-mile delivery services in shaping the operational and organizational feasibility of drone adoption across retail supply chains.

While empirical studies have examined drone applications in retail warehousing and last-mile delivery as largely discrete domains (Kembro and Norrman, 2022; Maghazei et al., 2022; Merkert et al., 2022; Perotti et al., 2022), integrative analyses spanning these contexts remain scarce. In particular, the B2B operational interface governing coordination between warehouse operators and drone service providers, encompassing both intra-warehouse activities and outbound last-mile delivery processes, has received limited scholarly attention. This omission is nontrivial, as successful drone integration requires coordinated decision-making across multiple organizational boundaries and operational domains.

To address this gap, our study examines aerial drone technology adoption and feasibility of collaborative processes at the B2B warehouse–delivery operations interface of the retail supply chain from a dual-stakeholder perspective, encompassing both warehouse operators and aerial drone-based last-mile delivery providers. Drawing on the technology–organization–environment (TOE) theoretical framework, we investigate how TOE factors shape functional adoption within warehouses, coordination across delivery operations and the organizational arrangements required for effective drone implementation at this interface, with implications for the performance of retail last-mile delivery systems.

Our research is grounded in the TOE framework, which provides a structured approach for examining technological innovation at the organizational level (Baker, 2011). While alternative theories, such as diffusion of innovation or the unified theory of acceptance and use of technology (UTAUT), primarily emphasize individual user adoption, the TOE framework focuses on organizational decision-making. This enables systematic identification and analysis of the technological, organizational and environmental factors that influence adoption outcomes at a broader, inter-organizational scale (Awa et al., 2017; Benchis et al., 2025; Tornatzky and Fleischer, 1990).

The TOE framework has been applied across various industries to explore technology adoption. Prior studies have employed TOE to examine warehouse automation (Ali et al., 2021), supply chain blockchain adoption (Benchis et al., 2025; Chittipaka et al., 2023), enterprise resource planning systems (Awa et al., 2017) and cross-country organizational studies (Malik et al., 2022). However, traditional applications of the TOE framework have predominantly adopted a firm-centric perspective, focusing on technological innovation and adoption within individual organizations (Baker, 2011). For instance, studies applying TOE to drone technologies and warehouse automation have primarily examined adoption decisions at the level of single firms (Ali et al., 2024; Maghazei et al., 2022). These studies emphasize how organizations assess technological readiness, organizational capacity and environmental pressures when deploying autonomous systems within their own operations. In contrast, this single-firm focus overlooks the critical B2B interactions between warehouse operators and drone service providers, as well as the collaborative arrangements through which technologies are implemented and operational integration is coordinated. The TOE framework is well suited to address this limitation, as it offers both theoretical grounding and empirical support across technology adoption contexts (Benchis et al., 2025; Tornatzky and Fleischer, 1990). Importantly, adopting a dual stakeholder perspective enables a structured evaluation of adoption factors across organizational boundaries, extending the application of TOE beyond individual firms to examine collaborative B2B adoption dynamics in the retail supply chain and their implications for last-mile delivery.

Specifically, we identify promising use cases by examining how warehouse operators and drone service providers perceive adoption decisions and collaborative arrangements at the technological, organizational and environmental dimensions. Our focus is on the B2B segment of retail supply chains, where decisions by 3PL/4PL providers, together with drone operators, determine whether drone logistics can transition from isolated pilot projects to integrated and scalable operations. These decisions have a direct impact on warehouse processes such as dispatch readiness, inventory management and handoff coordination, as well as on the operational feasibility of drone-enabled last-mile delivery.

Our empirical analysis draws on semi-structured interviews with senior managers from large-scale warehouse operations and drone service providers primarily engaged in last-mile delivery.

We contribute to the literature in several important ways. First, to our knowledge, our study is the first to provide firsthand managerial insights from both warehouse managers and drone service providers regarding demand for and feasibility of aerial drone use at the warehouse–last-mile delivery interface in retail supply chains. Second, drawing on the TOE framework, we analyze how organizational readiness, technological constraints and external regulatory conditions jointly influence drone adoption in this inter-organizational context. Third, we offer practical implications for managers and policymakers by identifying governance and coordination structures that either facilitate or impede drone adoption at the warehouse–last-mile delivery interface.

Our study focuses on drone operations management in retail warehouses and last-mile delivery, with emphasis on the operational and governance interface between operators and drone service providers, where drones support both intra-warehouse activities and downstream delivery operations.

As retail fulfillment and distribution warehouses grow in size and complexity, traditional manual methods are becoming increasingly labor-intensive and error-prone. Drones are increasingly viewed as part of broader smart warehouse solutions, particularly when combined with technologies such as radio frequency identification (RFID) and advanced sensor systems. Drones offer significant advantages in automating various functions. For instance, Afandi et al. (2024) demonstrated how using drones reduces the time required for inventory tracking from 15 to 2 min. Equipped with cameras and sensors, drones can achieve high detection accuracy. With collision avoidance systems and RFID technology, drones can safely navigate across aisles. Drones now also have the ability to use advanced positioning systems to safely navigate confined warehouse environments (Geetha et al., 2023).

Researchers have examined the adoption of drone technology in warehouses, but most findings are based on single case study observations (e.g. Maghazei et al., 2022). There is a need for a more comprehensive framework that can help decision-makers assess the practical implementation challenges and timing of this technology adoption. Although scholars have developed theoretical models and algorithms for drone operations (e.g. Moshref-Javadi et al., 2020; Mourtzis et al., 2024), there is limited research on the practical challenges of implementing these systems. Current studies focus on ideal conditions rather than dynamic and complex real-world warehouse environments (Stanko et al., 2022). Issues such as demand, operational requirements, operational models, payload capacity, battery life and integration with low-altitude airspace operations remain underexplored. Moreover, researchers have not fully addressed the practical integration of drones into daily operations or explored organizational, regulatory and economic constraints. Appendix A summarizes key studies on drone operations in warehousing, defined here as those that are highly cited.

Interest in drone applications has also increased in the last-mile delivery segment, largely driven by the expansion of e-commerce. Consumers are increasingly time-sensitive (Zhang et al., 2021), and warehouse managers need innovative technologies to reduce delivery times and ensure precision (Kembro and Norrman, 2022). Practical examples of companies such as Walmart, Zipline, DoorDash, Manna Drones and Wing have shown that drones are not only technically viable but are starting to shape last-mile delivery practices (Merkert et al., 2022).

Studies (e.g. Merkert and Bushell, 2020; Rejeb et al., 2023) have emphasized the operational advantages of drones in urban and rural contexts, including faster delivery times, reduced road congestion and lower emissions. Recent research has advanced the technical foundations of drone-enabled delivery, focusing on route optimization and infrastructure planning. For example, Nguyen and Dang (2024) proposed a system that manages individual drones and swarms using advanced air traffic management and flight control systems. Building on route optimization, Chour et al. (2023) addressed the multi-route weighted packaged delivery problem, and Murray and Chu (2015) developed a mathematical model for drone scheduling and the optimal distribution of parcels. Complementary studies have examined infrastructure considerations. For example, Aurambout et al. (2019) modeled the optimal location of drone beehives, and Yurek and Ozmutlu (2018) focused on minimizing the combined delivery completion time.

Although these contributions enhance the technical feasibility of drones in last-mile delivery, most remain focused on isolated operational aspects. Very few scholars (Maghazei et al., 2022; Purtell et al., 2025) have explored how organizational, infrastructural and ecosystem-related factors affect the deployability of aerial drones in real-world settings. Existing research instead prioritizes business-to-consumer (B2C) applications and operational optimization. Most of the research streams focus on algorithmic optimization, routing efficiency, vehicle–drone collaborative systems, consumer acceptance and environmental sustainability (Agatz et al., 2018; Boysen et al., 2018; Dorling et al., 2017). Consequently, the literature overlooks an important aspect: the organizational decision-making and inter-organizational coordination mechanisms that determine how such partnerships are formed and governed within the B2B segment of retail supply chains prior to last-mile distribution. This gap underscores the need for integrated frameworks that can account for multi-actor coordination, especially at warehouse–last-mile delivery interface.

The practical significance of this research gap is immediate. Drone adoption in the B2B segment of retail supply chains requires alignment between warehouse operators who manage inventory and order fulfillment operations and drone service providers who manage autonomous delivery operations (Merkert et al., 2021; Purtell et al., 2025). These are separate organizations with different operational needs, technical capabilities, resource limitations and business priorities (Ali et al., 2024; Maghazei et al., 2022). Within warehouses, initial implementation projects including stocktaking and order picking for retail/e-commerce items require close collaboration between warehouse operators and drone service providers (Maghazei et al., 2022; Perotti et al., 2022). The warehouse-to-delivery interface introduces additional complexity, such as determining whether the same drone fleets serve both warehouse delivery functions, as all require governance structures between warehouse operators and last-mile service providers (Hofmann et al., 2019; Kellermann et al., 2020). Without understanding these collaborative processes, scaling drone operations from fragmented pilots to viable commercial systems, a transition that depends on effective B2B collaboration between warehouse operators and drone service providers, remains underexplored.

Although drones have been studied in both warehouse and last-mile delivery contexts, the literature is largely technical or algorithmic in nature, with limited exploration of real-world business feasibility. Researchers rarely examine how operational decision-makers perceive drone adoption in terms of demand, risks and implementation strategies, especially at the critical warehouse–last-mile delivery interface. Most scholars treat drone adoption as a firm-level decision that is independent from the surrounding ecosystem and inter-firm adoption dynamics. In response to calls to explore practical and innovative solutions in logistics and supply chain management (Purtell et al., 2025; Rejeb et al., 2023; Ta et al., 2025), we evaluate the deployability of, and potential demand for, drone technology in (integrated) warehouse and last-mile delivery operations and identify potential use cases at that crucial interface. We adopt the TOE framework to examine organizational adoption of drone technology in the B2B segment of retail supply chains, focusing on the coordination interface between warehouse operators and drone operators. The TOE framework provides a structured approach to understand how technological characteristics, organizational capabilities of each party and environmental conditions interact to enable or constrain adoption and coordination. Appendix A presents an overview of key studies on drone applications in the last-mile delivery context.

The TOE framework (Tornatzky and Fleischer, 1990) offers a useful lens to examine factors influencing the adoption of technological innovations (Baker, 2011). It includes three dimensions that shape an organization's ability and willingness to implement new technologies: technology, referring to the characteristics of existing and new technologies; organization, capturing firm-level structural and managerial attributes; and environment, representing external regulatory, competitive and industry conditions (Baker, 2011). By integrating these three dimensions, the TOE framework enables researchers and practitioners to systematically assess technological opportunities, evaluate firms' internal readiness and understand external constraints. Oliveira and Martins (2011) reviewed firm-level technology adoption models and found the TOE framework most suitable for studying technology adoption, as it includes the environmental context.

From a theoretical perspective, we aim to build on the TOE framework to examine the interplay between technological, organizational and environmental factors that are important to the successful integration of drones in warehousing and last-mile delivery. Importantly, traditional applications of TOE have largely been firm-centric and focused on a single firm or organization (Baker, 2011). Despite literature on inter-firm innovation in the supply chain integration context (Wang and Zhang, 2024), evidence on the degree of coordination and collaboration at operational interfaces (e.g. warehouse management and last-mile delivery) is sparse. Moreover, the research designs of nearly half of the published studies (for an overview, see Appendix A) lack a theoretical foundation, thereby limiting the advancement of knowledge within the rapidly evolving field of drone-based systems in logistics (Garg et al., 2023).

Using the TOE framework as a sensitizing lens, we structured our interview guide across three broad dimensions. To address the technology dimension, we explored how actors evaluate characteristics of drone technology and its existing capabilities in warehouses and delivery contexts, including perceived benefits, technical constraints and integration challenges. For the organization dimension, we examined firms' internal capacity to adopt drone technology, including resource availability, governance structures and organizational readiness. To address the environment dimension, we investigated external factors that influence drone adoption, including regulatory frameworks, industry standards, infrastructure availability and ecosystem alignment.

Given the exploratory nature of drone adoption at the warehouse–last-mile delivery interface, we adopted a theory-informed inductive approach (Varpio et al., 2020), using the TOE framework to structure our research design and guide data collection and analysis, while allowing themes to emerge inductively from participant responses.

We collected data primarily via semi-structured expert interviews, using an interview guide based on the three domains of the TOE framework (Tornatzky and Fleischer, 1990). We interviewed 38 managers representing 12 logistics service providers, mainly 3PL and 4PL firms operating warehouses serving e-commerce and retail activities (see Appendix D for the interview guide) and 26 drone operators (see Appendix E for the interview guide). Participants were based across Australia, Brazil, various European and Asian countries and the USA. Our sampling strategy focused on countries and organizations leading in drone logistics innovation, targeting regions where commercial drone use is actively piloted or operational and where operators were willing to talk to us. This approach facilitated insights across diverse regulatory and operational contexts. The sample sizes of logistics service providers and drone operators differed due to purposive sampling and data saturation. Interviews with warehouse managers achieved thematic saturation early, as logistics service providers in large 3PL and 4PL environments consistently expressed similar views on the demand for drone-supported warehouse functions and last-mile integration. Consequently, additional warehouse interviews yielded limited new insights. In contrast, the drone ecosystem has significant functional and institutional diversity, including drone logistics providers, technology developers and airspace management organizations. Expanding the drone operator sample enabled us to capture varied operational roles, technological maturity and regulatory environments across different regions in the implementation phase of drones.

Our informants held strategic or operational roles with direct knowledge of drone technology, warehousing systems or last-mile logistics. Logistics participants were senior practitioners from 3PL and 4PL warehouses, with prior exposure to, or informed knowledge of, drone applications. Drone participants were managers involved in the application of drones within logistics operations, including professionals from drone start-ups, drone manufacturing firms and organizations developing commercial drone use cases. This dual-perspective enabled analysis of drone adoption across intra-warehouse operations and last-mile delivery interfaces. Through purposive sampling, we were able to recruit managers with a diverse range of perspectives. Informants were purposively selected and recruitment continued until thematic saturation was reached (Fusch and Ness, 2015). Table 1 provides details about the managers interviewed.

Following ethics approval, interviews were conducted between August 2024 and March 2025, either in person or via video conferencing, recorded with consent and transcribed verbatim. Secondary data, including industry reports and regulatory documents, were collected to support contextualization and triangulation.

We deployed a theory-informed inductive approach (Varpio et al., 2020), where the TOE framework served as a sensitizing lens for thematic analysis. The coding process commenced with TOE-aligned categories but allowed for additional codes to be created when participants' narratives extended beyond the existing framework. This approach ensured that factors not encompassed by the original TOE framework could be incorporated into the analysis.

The TOE framework provided a general structure without predetermining specific findings, enabling representatives of both warehousing companies and drone operators to share their experiences within these theoretical dimensions. The TOE framework was “let in” during interpretation to help organize the inductively derived themes around technological, organizational and environmental dimensions, while ensuring informants' voices and unexpected findings could emerge organically (Nguyen et al., 2022). Open-ended questions within the TOE structure allowed for unexpected insights to surface about technology adoption benefits, organizational challenges and environmental factors. This theory-informed inductive approach aligns with recent methodological guidance advocating for flexible application of theoretical frameworks in qualitative research (McGowan et al., 2020; Proudfoot, 2023), allowing theoretical insights to guide analysis while remaining open to emergent themes that may extend beyond existing frameworks (Sarraf-Yazdi et al., 2024).

To identify and categorize recurring themes and patterns, we analyzed the collected data using NVivo software and Python (Maher et al., 2018; Saldaña, 2021). Interview transcripts were anonymized and imported into NVivo. Python was used as a complementary tool to organize coded outputs, support comparison across categories and assist in grouping governance models based on economic, operational and infrastructure-related factors. Intercoder reliability was established through a collaborative coding process involving the researcher and two high-level supervisors. Emerging codes and thematic interpretations were regularly reviewed and discussed, with discrepancies addressed through iterative dialogue and refinement of the coding framework until consensus was achieved.

We thematically analyzed our data by following the coding process proposed by Braun and Clarke (2006). We first familiarized ourselves with the transcripts, generated initial codes and iteratively grouped these into higher-order themes aligned with the TOE dimensions. Themes were then reviewed and refined to ensure internal coherence and conceptual clarity. To improve transparency in how participant insights were organized and presented, the resulting coding hierarchy was structured in a manner consistent with the data organization logic of Gioia et al. (2013). This iterative process resulted in the thematic structure presented in Appendix B, mapping themes to TOE dimensions and emergent sub-themes. Figure 1 illustrates the coding progression using selected sample quotes, from initial codes to refined themes.

We also applied the following strategies to ensure rigor:

1. Triangulation: We collected data from multiple sources (logistics service providers and drone operators; primary and secondary data) to enhance the credibility and validity of our study (Meydan and Akkaş, 2024).

2. Saturation: We interviewed representatives of relevant firms in the industry, and our sample size was determined by conceptual saturation.

3. Audit trail: We maintained detailed memos in NVivo during coding and theme development.

4. Transparency: We documented our coding framework and interpretation steps.

5. Reflexivity: We minimized interviewer bias by employing bracketing and consistent protocols.

Table 2 presents a synthesis of the thematic findings derived from the 38 interviews, organized around the TOE framework. Each dimension captures the key themes, subthemes and illustrative quotes that emerged from the data, alongside summarized findings. The following sections discuss each identified theme in turn, beginning with demand assessment within warehouse environments.

Most informants confirmed the potential to adopt drones for warehouse functions, agreeing that drones have emerged as a transformative technology. By automating repetitive tasks and reducing human error, drones offer a cost-effective solution for 3PL and 4PL warehouses. Several companies already have advanced projects in development, with demonstrations taking place as early as four years ago (in 2025) (informant D-20-Q). Drone pilot programs for warehouse automation and systems integration were implemented as early as 2016 (W-8-H). Acknowledging potential applications, one CEO said their company had invested in drone testing with cameras and engineering support and was conducting proof-of-concept studies with 2–3 customers: “The concept was proven; it was more about how we make it compelling to actually roll this out” (W-5-E). Operators said drones were found to be very useful for verifying inventory (W-8-H). They were open to drone adoption for warehouse operations and cited labor shortages as a major challenge that drones could address (W-7-G).

Drone adoption is also influenced by the scale and complexity of warehouse operations. Larger facilities with higher throughput have greater incentives to integrate automation for cost savings: “It depends on the scale of your manufacturing. If the scale is large, then definitely you need to have innovative technologies. You need to keep your inventory in an order where drones can go and scan the inventory” (W-1-A). Moreover, most small- and medium-size enterprises are likely to become subcontractors of a 3PL provider. Companies like Google Wing, OK and Amazon are particularly well placed to be 4PL providers, as they have their own warehouses and fulfillment centers (D-7-G) where drones are most suited for stocktaking. Furthermore, regulatory changes in the USA and Australia in 2025 and 2026 are making indoor drone operations more feasible:

Discussions about the implementation of drones in warehouses were guided by the technological dimension of the TOE framework. Operators indicated that the primary functional applications for drones in warehouses are stocktaking, inventory cycle counting, item retrieval and shelf audits. These functions are labor-intensive and are important in inventory management. A logistics team lead explained: “In our operations, one of the biggest challenges is aligning physical inventory with system records, especially in our case like fast-moving environments. There have been instances where a SKU couldn't be found, despite being shown in the system” (W-12-L). Drones could also assist in small item order picking or transferring small parts between warehouse zones (Thomaidis and Zeimpekis, 2024). This would reduce reliance on manual labor for time-consuming tasks (W-11-K). Operators emphasized that in regions like Australia, where labor costs are high, automation can help reduce high operational costs for warehouses (W-2-B).

A supply chain manager described their experience integrating drones into warehouse operations systems. Warehouses at clients' sites require regular verification of stock accuracy in terms of total stock, pallet locations and SKU details. Tasks include weekly stocktakes (subsets of stock), full stocktakes (2x/year) and identifying obsolete stock and shrinkage. These activities are typically done by hand and are very labor-intensive. Now, autonomous drones are used to map pallet positions by scanning and taking images of each aisle. Pallet IDs are obtained from barcodes in these images and compared against warehouse management system (WMS) data. A drone can check 3,000–4,000 pallets per day and operate outside normal operating hours. This amounts to substantial cost savings, as the annual cost of a stock counting employee can be up to USD 80,000 (in any jurisdiction). Drones can also record images as proof of product condition, collect additional data on pallet or racking conditions, ensure additional surveillance to minimize product theft and scan inventory at high speed (W-8-H).

Warehouse operators emphasized that the successful integration of drones depends on technological compatibility with existing warehouse management systems (D-6-F). A CEO and founder introduced the concept of a 4PL warehouse ecosystem, where aerial drones work in coordination with ground-based robots:

Some drone operators further highlighted the potential application of drones for order picking, although they acknowledged that this would require more complex integration into warehouse operations compared to inventory management (D-4-D). They also mentioned that drones could be used for small item picking or internal warehouse transfers (D-16-M).

Despite the potential of drones in 3PL/4PL warehouse operations, managers revealed several challenges. Technical limitations in barcode reading and quantity verification present challenges for accurate stock counting (W-4-D). Unless drones can achieve the required accuracy in quantity verification, warehouse operators are reluctant to invest in the technology (W-4-D). They explained that while drones can read front-facing barcodes, they might struggle to count items in multi-layer pallets (D-5-E). Warehouse operators also emphasized the importance of integrating drones with existing WMS to ensure the technology is effective (W-2-B). For instance, DHL closed its Parcelcopter division due to such technical constraints:

Informants also highlighted infrastructural challenges, including current warehouse layouts and storage methods. In conventional layouts, items are stacked in multiple layers on pallets, which are difficult to lift with drones (W-9-I). Facilities must be redesigned to maximize the benefits of drone technology. A partnerships manager also pointed out environmental issues in traditional warehouses:

Challenges of battery life and payload limitations have also stalled full commercial deployment: “We had a few challenges from a safety perspective. We also faced issues with battery life and payload. Flying a drone is not as simple as people might think” (W-5-E). Safety considerations and operator proximity within confined spaces were also mentioned as significant barriers:

Although six managers remained skeptical of drone deployment within warehouses, all managers acknowledged growing demand for last-mile delivery via drones, especially for e-commerce goods and food items. They expressed confidence that drone delivery systems are already being implemented and are expected to scale rapidly in the near future. Where drones are unconstrained by high-density urban environments and traffic congestion, they can provide rapid, on-demand delivery. This capability is becoming increasingly important since customers now prioritize speed and reliability for last-mile fulfillment. Although scholars have examined drone delivery from a customer perspective in detail, demand from the supply side remains underexplored (Maghazei et al., 2022). We address this gap in the next section, with findings from organizations that are actively contributing to the growth of the last-mile delivery sector.

A few of our informants represent organizations that are actively and publicly contributing to the growth of the last-mile delivery sector. Manna, Wing, Zipline, Matternet and Speedbird Aero have all created logistics concepts that structure the current model for successful drone delivery. In the B2C market, Manna has a defined delivery radius that contributes to profitability. As the delivery fee is fixed, the number of flights per hour is decisive. Wing flies drones faster and farther (8–15 km) than most competitors. Although their drones make fewer flights, the catchment area per hangar is larger. Both Manna and Wing place their hubs close to retail locations. Wing carries a sample stock and can pick up parcels from partners and deliver them. Matternet provides standardized parcel boxes for drones to pick up and an integrated workflow for medical deliveries. Speedbird Aero has collaborated with Skyports as its global drone delivery partner. During interviews, operators mentioned that B2B deliveries have been slow to scale, but they do expect deliveries between warehouses and consumption points to grow. Nevertheless, at this point:

Most operators reported growing demand for drone-enabled last-mile delivery, particularly for e-commerce goods, food items and other time-sensitive deliveries like medical supplies. This demand is primarily driven by faster speed, traffic avoidance and the ability to serve locations where traditional road-based delivery is less efficient. As one operator explained, moving parcels “from the warehouse to the end user” or “from hub to node” with a drone provides an operational advantage because drones “can move faster” and are “not subject to traffic” (D-1-A). This advantage is particularly significant in congested urban environments, where one manager noted that customers are willing to “pay a premium for a three-hour delivery over a three-day delivery” (D-12-J). Operators also identified specific use cases where drone delivery could provide greater value than conventional delivery modes. These include food delivery, urgent medical supplies and premium products (W-3-C). Demonstrating the capability for urgent deliveries could help prove the technology and business case for broader adoption:

In addition to direct consumer delivery, operators identified hub-based and corridor-based models as more practical options for implementation. One operator suggested that last-mile drone delivery could be effectively managed through “two or three key distribution points” in cities, especially as operators can now manage multiple drones with fewer regulatory constraints (D-3-C). Another operator described a fixed-route model, explaining that “we build bridges between points, so we have point A and point B, and there is an air corridor that connects these points”, while also reporting more than “30,000 successful flights” (D-24-U).

Operators also suggested that several concerns commonly associated with drone delivery, particularly noise and privacy, are becoming more manageable as the sector matures. Although customers, industry stakeholders and authorities often perceive noise and privacy as barriers, operators argued that these concerns can be reduced through appropriate measures:

Despite the perceived benefits, operators identified several barriers that may limit the near-term scalability of drone-enabled last-mile delivery. The most frequently mentioned barriers were related to technical readiness, regulatory approvals and supporting infrastructure. Technical concerns focused on aircraft conspicuity and detect-and-avoid capabilities, with one participant noting that drones will need to “share the airspace with many other vehicles that we don't know where they are” and that current systems are not yet “super reliable” (D-20-Q). Regulatory capacity and the processes for obtaining BVLOS approvals were also seen as significant constraints. Participants expressed concern that regulations are “2–3, maybe four steps behind”, as technology is evolving faster than the approval processes can keep up (D-22-S). Additionally, regulatory understaffing was noted as a major obstacle, with some approvals taking longer than nine months (D-15-L).

Operators also highlighted the need for appropriate ground and operational infrastructure. One operator argued that drone delivery depends on “safe landing zones or ditch zones or crash zones on route”, comparing these to rest areas for heavy trucks and emphasizing that the service “will rely on the provision of ground infrastructure” (D-7-G). Others pointed to geolocation infrastructure, anti-collision systems, battery charging, downtime and the need to monitor drones during operations (D-11-G; W-2-B). Operators also emphasized that the economics must match the use case. As one participant noted, high-value or critical deliveries may justify drone costs, but “most likely an e-commerce customer would not be willing to pay” very high delivery charges (D-13-K).

An important yet underexplored theme is the role of drone operators' expertise in advancing drone operations (Companik et al., 2018; Kumbhani and Kant, 2024; Rejeb et al., 2023). Drone operators expressed strong support for integrating drones into logistics networks. Interestingly, the chair of a drone company described an innovation that supports vertical integration: “We have a fully autonomous drone that integrates into the pick and fill. After locating the product, it picks the product and then autonomously departs towards the destination” (D-23-T). Discussions revealed three interrelated themes: technology readiness, regulatory alignment and infrastructure maturity. Technologically, drones are now capable of handling complex warehouse operations and last-mile deliveries. However, logistics firms often lack the technical frameworks to integrate them into existing supply chains.

Drone operators are emerging as pioneers in advancing the technological adoption of drones in logistics and supply chain processes. A CEO of a drone operating company indicated that their company is enabling adoption of long-range aerial data solutions at scale by being the first in Australia to manufacture all-carbon-composite airframes for unmanned aerial vehicles and the first to commercialize hybrid drones in the >7 kg category (D-5-E). A market development director indicated that progress in drone technology is no longer confined to controlled trials; drone delivery systems are now deployed in complex, real-world scenarios, such as medical supply delivery in remote areas and time-sensitive e-commerce fulfillment (D-11-G). It is noteworthy that many drone operators reported that their businesses are profitable, despite the common conception to the contrary. One operator disclosed: “Typical delivery in less than 3 min, and profitable—truly inspiring and at a level of professionalism that will help the industry move forward” (D-4-D).

A potential barrier voiced by drone operators was that logistics operators, often lacking technical knowledge, are concerned about the risks of integrating drones into their operations and the capabilities of a drone ecosystem. In response, drone operators are focusing on implementing practical solutions and ensuring their trainers are qualified and have backgrounds in the industry (D-3-C). All operators mentioned the importance of regulatory alignment and maintaining compliance with fast-changing airspace regulations. Regulatory frameworks have been lagging behind, as confirmed by a CEO in the USA who said their firm had the capability long ago to launch regular flights in a controlled airspace between cities up to 100 km, but the time-consuming and costly process of obtaining certifications ended their ambitions (D-15-L). To overcome such challenges, drone operators have been collaborating with regulatory authorities on the development of unmanned traffic management and wireless connectivity (D-4-D). A CEO explained: “We are registered with the Civil Aviation Safety Authority [CASA, Australia] and the Australian Skills Quality Authority [ASQA] training organization” (D-3-C). An operator was optimistic about recent developments:

Two interconnected themes emerged from the interviews concerning ownership and, more specifically, governance models for drone operations at the warehouse–last-mile delivery interface. First, uncertainty remains as to whether warehouses should develop in-house drone capabilities or rely on third-party service providers to manage drone logistics. Second, it is unclear whether the same drones should be used for both internal warehouse operations and last-mile delivery or whether specialized drones should be deployed for each task, requiring coordination between drones and/or human operators at the interface. Drawing on the organizational dimension of the TOE framework, we examine how successful drone implementation depends on governance structures and alignment with business objectives.

Warehouse operators indicated that economic considerations play a central role in decisions regarding whether to adopt in-house drone systems or rely on third-party service providers. In particular, return on investment is a key criterion in assessing the economic viability of automation initiatives. One logistics lead emphasized that partnering with a third-party provider offers the flexibility required to scale operations: “Investing in a drone fleet involves substantial upfront capital, alongside ongoing maintenance costs that can significantly affect budgeting. By contracting third-party providers, we can achieve greater cost control and flexibility to scale” (W-11-K). Similarly, a founder suggested that offering drone operations as a service may be advantageous, as specialized providers are better positioned to manage technical and regulatory complexities (D-6-F). In such a model, third-party providers would manage delivery operations in a manner comparable to traditional carrier services.

Operational considerations are also key in determining whether warehouses should maintain in-house drone fleets or outsource drone operations. In-house ownership requires internal capabilities in drone management, including ongoing maintenance, personnel training and compliance with aviation regulations. A logistics manager noted that many companies face operational challenges due to a lack of specialized expertise. This capability gap may be mitigated by engaging consultants or specialized firms capable of managing procurement and implementation processes. At the same time, one operator highlighted the potential for in-house management, citing Amazon as an example. However, large firms may encounter difficulties when organizational leadership is grounded in traditional logistics expertise but lacks specific operational capabilities in drone systems (D-25-V). Operators also emphasized that maintenance requirements are likely to remain a critical challenge.

Regarding the second theme, one approach discussed was an integrated drone system, in which the same drone performs both internal warehouse operations and last-mile deliveries. However, interviewees noted that this approach faces several challenges. Warehouse tasks typically involve short-range movements within confined spaces, such as between aisles, and require high levels of precision in navigation, with minimal payload requirements. In contrast, last-mile delivery tasks generally require drones with extended flight ranges and greater payload capacity (D-21-R). These differences suggest a potential trade-off between integration and specialization, with implications for governance design and operational coordination at the interface. Acknowledging the higher initial capital cost, a warehouse operator suggested using different drones:

Warehouses could also leverage a hybrid model by implementing a docking station handover system. Retrieved packages can be transported to designated docking stations at the warehouse exit via drones. At these docking stations, last-mile drones can pick up parcels and proceed with deliveries. Relationships between warehouse operators and drone delivery providers can be established based on service-level agreements or performance-based contracts that define operational expectations, insurance coverage and payment terms. Compliance with aviation safety standards, airspace permissions and privacy requirements is an important element in agreements. One informant whose company has trialed such a system explained: “Our contracts outline responsibilities [for] who maintains docking stations, what happens in case of drone collision or exact timeframes for drone handovers” (D-15-L).

Interestingly, discussions with the drone operators centered largely on regulatory constraints, particularly restrictions on flights beyond visual line of sight (BVLOS). Overcoming these barriers would require the introduction of a low-altitude airspace management (LAAM) system to govern commercial drone operations (Merkert et al., 2021). Such a system could enable safe and scalable operations in increasingly congested airspaces, for example, by coordinating drone traffic through designated air corridors. One operator noted that their firm is currently the only provider authorized to conduct BVLOS flights over urban areas in Switzerland:

The collected evidence suggests that drone operators are currently focused on establishing BVLOS capabilities, which consume substantial managerial attention and resources. While warehouse applications are recognized as promising, they are not an immediate priority.

Another major challenge highlighted by operators concerns collaboration with existing aviation stakeholders who are often reluctant to invest in drone-related infrastructure (D-20-Q). Understaffing in regulatory bodies was also identified as a major bottleneck for drone implementation. An operator explained: “There are too few assessors in CASA's [Civil Aviation Safety Authority, Australia] drone team, and it takes forever to get the approval. We have got to wait 9 months, and we'll pay $30,000 or $40,000 to get that” (D-18-O). These constraints point to the need for regulatory frameworks to evolve alongside technological capabilities in order to support safe and efficient operations. “There is a need for regulation on this. For example, who gets the priority when two drones want to cross each other? If you address the safety properly, there is a lot of scope in the market” (D-6-F).

Prior research suggests that successful technology adoption in complex retail logistics ecosystems often depends on collaborative governance arrangements, such as aligning commercial proximity between stakeholders to enable coordinated investment and operational integration (Bushell et al., 2025). Consistent with this, themes centered on inter-organizational and ecosystem coordination and collaboration emerged from our interviews, as managers frequently discussed the practical arrangements required to integrate drones into existing logistics systems.

Our interview findings indicate that drone integration is unlikely to progress through isolated organizational efforts. Instead, it requires coordinated ecosystem development across logistics operators, drone service providers, regulatory authorities and supporting industries. Multiple participants emphasized cost efficiency as a driver of collaborative arrangements over independent drone deployment. Recognizing the technical complexity of drone operations as a barrier to in-house deployment, one operator noted:

Moreover, the complexity of drone approval and certification processes further reinforces the need for collaborative approaches, with participants noting that “drone approvals are difficult to get and difficult to understand and therefore it would cost a lot to achieve that” (D-4-D). This dynamic creates a natural division of labor, with warehouse operators focusing on inventory management while relying on specialized drone operators for flight operations.

Our interview findings also highlight the intermediate role of consultants in bridging capability gaps between logistics providers and drone operators. As one operator explained:

This approach may also facilitate initial infrastructure setup, with ongoing support provided for maintenance and technological development. Such a phased model would involve warehouse operators initially partnering with drone service providers, before potentially transitioning to greater ownership as volumes and internal capabilities develop (W-2-B).

Drone operators also emphasized the mismatch between logistics expertise and aviation knowledge. One operator, for example, noted Amazon's challenges with “leaders that are in the logistics area, but they have limited knowledge about drones” (D-16-M). This capability gap encourages some drone operators to pursue vertical integration, developing “a vertically integrated service that starts with transportation capability and has software that plan, track and follow up shipments, taking care of the flight operations” (D-11-G). The interdependence between flight operations and service design was frequently emphasized, with one participant arguing that “you need to co-develop them” rather than attempting to separate these functions (W-6-F).

Our interview findings also indicate that competitive advantage in drone logistics depends on ecosystem development rather than sole ownership:

These inter-organizational dynamics suggest that drone logistics is likely to evolve through collaborative ecosystem development rather than isolated firm-level deployment. However, such ecosystem development does not follow a single configuration. Across our interviews, participants described three approaches to organizing drone integration at the warehouse–delivery interface, each reflecting a different distribution of capabilities and responsibilities between warehouse operators and drone service providers. The first configuration centers on a single actor maintaining full control over fleet operations. The second combines in-house operations with selective outsourcing of specific functions. The third involves collaborative partnerships, in which warehouse operators work with specialized drone providers to coordinate activities, particularly in last-mile delivery. Table 3 presents illustrative quotes for each configuration. We return to these configurations in Section 6.3, where we develop their implications at the organization–environment intersection.

Our interviews with warehouse and drone operators revealed several technological, organizational and environmental factors that shape the adoption of aerial drone operations at the warehouse–last-mile delivery interface in retail supply chains. It is evident that it is technologically possible to adopt and deploy aerial drones not just in last-mile delivery services, but also in warehouse operations, and more importantly, in a system that integrates both logistics components. Deploying such integrated systems at scale requires addressing technological, organizational and environmental concerns.

Our findings indicate that drones are viewed as a transformative technology with the potential to automate labor-intensive functions in a retail warehouse, including stocktaking, cycle counting, item retrieval and shelf audits. Operators have reported significant cost savings, especially in larger, high-throughput facilities. This aligns with evidence showing that automation technologies can deliver efficiency gains in retail logistics environments where scale facilitates growth and maturity (Boysen et al., 2018; Fragapane et al., 2021). The readiness for adoption was found to be contingent on the technological compatibility of drones with existing WMS. This underscores the idea that successful integration with current systems, rather than just the technology's capabilities, determines the outcomes of adoption (Oliveira and Martins, 2011; Tornatzky and Fleischer, 1990). However, operators also identified some barriers to full deployment in retail warehouses. A primary concern is technical limitations regarding barcode reading accuracy and quantity verification. Operators expressed hesitation to invest until these accuracy issues are resolved. Additionally, infrastructure challenges have been identified as factors that slow down the commercial rollout. These challenges include the need to redesign warehouse layouts, battery life constraints, payload limitations and safety concerns in confined spaces. These findings are consistent with the broader literature on technology adoption, which shows that technical immaturity and integration complexity can hinder adoption, even when perceived benefits are recognized (Rogers, 2003; Rejeb et al., 2023).

In the context of last-mile delivery within retail supply chains, informants expressed strong confidence in the growth of drone delivery systems, especially for e-commerce and food retail distribution. Unlike traditional road-based delivery, drones can operate without the hindrances of traffic congestion and geographic limitations, allowing for faster order fulfillment. Most of the operators we interviewed represented organizations that are actively participating in the last-mile delivery sector. This reflects a growing body of evidence indicating that last-mile drone delivery is transitioning from pilot programs to operational deployment (Aurambout et al., 2019; Goodchild and Toy, 2018; Stolaroff et al., 2018). Interestingly, some managers who were skeptical about the use of drones in warehouses recognized the increasing demand and momentum behind last-mile delivery. This distinction highlights a critical technological interface in the retail supply chain: the handoff points between warehouse operators and drone delivery operators. At this B2B interface, the technical requirements extend beyond the capabilities of individual systems. They also encompass the need for interoperability between WMS and drone fleet management systems, which would involve real-time coordination of dispatch and retrieval operations. Operators noted that this integration would enhance the overall value that drones can provide across the supply chain. This represents a key finding of our study: the adoption of drones in logistics cannot be evaluated in isolation, whether from the warehouse or the delivery perspective. It must also consider the technical infrastructure that connects the two.

From an organizational perspective, drone operators have emerged as key drivers of adoption in retail supply chain operations. Their roles have evolved beyond early trials, with operators confirming that drone systems are now being deployed in complex, real-world scenarios for retail fulfillment. Operators have also indicated that drone delivery businesses are already profitable, contrary to the common perception that commercial viability has yet to be achieved (Kumbhani and Kant, 2024; Rejeb et al., 2023). However, a significant organizational challenge identified was the knowledge gap in logistics. Warehouse operators, often lacking technical expertise, expressed concern about the risks of integrating drones into their operations. In response, drone operators have focused on building qualified training teams with industry backgrounds to bridge this gap. This aligns with research demonstrating that inter-organizational knowledge transfer plays an important role in adopting emerging technologies (Autry et al., 2010; Oliveira and Martins, 2011). Regulatory alignment has also emerged as a significant concern. Operators noted that regulatory frameworks have not kept pace with technological advancements and certification processes are often described as time-consuming and costly. Even though many drone operators have the technical capability to conduct flights up to 100 km between cities, they are unable to progress due to regulatory bottlenecks. As a result, firms are actively collaborating with regulatory authorities on unmanned traffic management, with 2025 identified as a transitional year for BVLOS and cross-border flights.

At the organizational level, the B2B interface between warehouse operators and drone delivery providers raised two important governance questions. First, whether warehouses should develop in-house drone capabilities or rely on third-party service providers. Second, whether the same drones should serve both internal warehouse operations and last-mile delivery or whether specialized drones should handle each separately, requiring structured coordination at the handoff point. Informants indicated that both economic factors, particularly return on investment and operational factors, including maintenance requirements, personnel training and regulatory compliance, shape these decisions. To manage these, warehouse and drone delivery operators are structuring their relationships through service-level agreements and performance-based contracts (Purtell et al., 2025). These findings highlight that successful drone adoption at the warehouse and last-mile delivery interface depends on how organizations structure their roles, responsibilities and relationships.

From an environmental perspective, regulatory barriers have emerged as the most significant challenge affecting drone adoption, particularly regarding BVLOS operations. To enable BVLOS at scale, an LAAM system would need to be introduced to safely coordinate commercial drone operations in increasingly congested airspaces, as proposed by Merkert et al. (2021). Additional environmental pressures include understaffing at regulatory bodies and the reluctance of current aviation stakeholders to allocate funds toward drone infrastructure. These factors have been identified as major bottlenecks that slow down certification and approval processes. Our findings support broader calls in the literature for regulatory frameworks to evolve alongside technological developments to unlock the potential benefits of drones in retail logistics (Clothier et al., 2015; Rejeb et al., 2023).

Moreover, our results highlight the necessity for ecosystem-level coordination across warehousing and delivery. Operators indicated that drone integration within retail logistics cannot progress through isolated organizational efforts alone. Instead, it requires collaborative development among logistics operators, drone service providers, regulatory authorities and supporting industries. Our findings also reveal a knowledge gap between logistics expertise and aviation knowledge, with operators noting the challenges faced by organizations whose leaders understand logistics but have limited understanding of drone technology. The interdependence between flight operations and service design was also emphasized in our discussions. Operators argued that these functions must be co-developed rather than treated separately. Consistent with evidence that successful technology adoption in complex logistics ecosystems relies on collaborative governance arrangements (Bushell et al., 2025), our findings suggest that the advancement of drone logistics will be most effective through shared ecosystem development rather than siloed efforts.

A comparison of warehouse manager and drone operator perspectives across the major themes identified through our interview program is provided in Appendix C.

Our findings also suggest that large-scale adoption depends on how warehouses and service providers organize themselves within the broader logistics ecosystem. This raises an important question of whether warehouses and service providers will continue to operate in fragmented and siloed configurations (Maghazei et al., 2022; Purtell et al., 2025). During our interviews, key stakeholders indicated that standalone experimental implementations of drones for warehouse activities and last-mile delivery services are largely seen as underperforming due to poor ecosystem alignment. These findings highlight the importance of examining interactions between TOE dimensions when studying inter-organizational systems and collaborative adoption of emerging technologies like drones. Governance decisions cannot be understood through isolated analysis of organizational or environmental factors alone; rather, they emerge from the dynamic interplay between organizational capabilities (O) and environmental ecosystem (Ali et al., 2021) requirements (E). This underscores the value of TOE as an integrative framework that captures multidimensional adoption dynamics.

As indicated in Section 5.2.2, a central question emerging from the interview findings and worth discussing in further detail is the choice of governance model for drone adoption: whether vertical integration represents a viable and strategic option for logistics firms or whether collaborative models offer a more suitable pathway. Our analysis identifies three distinct configurations: vertical integration, inter-organizational collaboration and partial deployment with intra-organizational collaboration (see Table 3).

This governance choice reflects a key interaction between the organizational (O) and environmental (E) dimensions of the TOE framework. While the decision is made at the organizational level, it is influenced by environmental factors, including ecosystem maturity, partner availability, regulatory frameworks and infrastructure dependencies. As such, it is inherently inter-organizational, highlighting the need to examine drone adoption beyond individual firm boundaries.

The following three subsections discuss how governance models may drive the integration of drones into the broader logistics chain. We define each governance model and discuss its application at the warehouse–last-mile delivery interface, highlighting illustrative examples as well as associated benefits and risks, thereby positioning governance as a boundary-spanning mechanism linking organizational decision-making with environmental constraints.

The vertical integration governance model aligns with a central multi-use platform where a single actor maintains full control over multiple stages of its supply chain (Beal-Partyka and Laureano-Paiva, 2024). In drone operations, a vertically integrated company would operate its own drone fleet and related infrastructure as part of its logistics operations. These operations include inventory tasks, route optimization, last-mile delivery and customer experience.

Vertical integration requires seamless integration of drones, minimizing interoperability issues and reliance on technology from external providers. For example, a warehouse could have drone launch pads on the roof or vertiports where orders are dispatched for last-mile delivery. Amazon is a good example of a vertically integrated firm with significant capital resources and advanced technological capabilities. From the service provider side, Wing, Zipline and Speedbird Aero have end-to-end control over their flight operations.

The vertical integration model with in-house drone operation reflects a firm's strategic commitment to automation, which in our context would require significant restructuring. Companies such as Amazon may prioritize confidentiality and competitive positioning favoring internal control over collaboration with external providers. Workflows would change, and firms would have to introduce new protocols. Employees would need to be trained to ensure they can handle drone operations smoothly, as any disruptive technological change creates significant uncertainty and risk.

Vertical integrations would present a higher regulatory burden for warehouses, as aviation certifications are required, which our findings suggest is no easy feat. BVLOS operations are critical for last-mile delivery; however, they are currently only permitted in low-risk areas, limiting the scalability of these operations (Purtell et al., 2025). In constrained urban airspaces or sensitive biopharma delivery corridors, predictability, regulatory compliance and rapid decision-making processes are important. The vertical integration model provides the clearest line of accountability in this regard, as drones equipped with AI enable real-time data flows that remain within a unified feedback architecture, reducing latency in optimization. Vertical integration may be advantageous at this stage, as the ecosystem may not yet be sufficiently mature to support multiple specialized operators across both drone operations and regulatory compliance. Aviation authorities also tend to prefer engaging with a single responsible entity, valuing consistency in procedures and risk assessment.

Drone operators should move beyond functioning solely as transport providers and expand their control across multiple stages, including in-house drone manufacturing, route planning software development, fulfillment center operations, vertiport management, selected warehouse functions and engagement with last-mile delivery customers. Such integration may facilitate scalability in delivery operations while enabling closer alignment between drone capabilities and logistics requirements. However, this model requires substantial resources and is therefore more feasible for larger firms with advanced capabilities and access to regulatory environments that support BVLOS operations. High capital intensity, combined with limited in-house technical expertise, presents a significant barrier for smaller logistics operators (Maghazei et al., 2022).

Firms that implement the partial adoption model use drones to enhance operational efficiency within their own logistics chains that include inventory management, cycle counting, order picking and last-mile delivery. Unlike vertical integration, the focus is on using drones for specific tasks rather than automating all operational functions and drone operations do not cross organizational boundaries. For example, a warehouse might deploy drones for stocktaking, but not for last-mile deliveries. Walmart has used drones to automate internal inventory management, GEODIS initially collaborated with DELTA DRONE prior to bringing operations fully in-house, and DHL has tested and deployed drones for inventory counting globally (Abrams, 2016; GEODIS, 2018; Yigit et al., 2024).

For last-mile delivery, the idea behind the partial adoption model is a network where multiple delivery mechanisms work together in a mesh. Rather than drones handling the entire process from end-to-end, mesh delivery involves smaller interconnected hops. For example, the Amazon Prime Air model involves human drivers delivering parcels to drone dispatch hubs, which tend to be located close to customers, and the last mile of delivery is provided by drones (Tarasov, 2023). In such systems, human–drone transitions must be tightly coordinated to ensure smooth and reliable operations. The partial adoption model may be a pragmatic approach that better captures how the industry will leverage existing last-mile delivery networks. It also offers a solution for deliveries in dense areas with many apartment and office buildings as opposed to suburban areas.

However, the partial adoption of drone deliveries does require a cultural shift where staff are trained on drone operations. Rather than replacing human labor, drones are intended to augment or complement existing logistics staff. In-house operations would be controlled primarily by managers within the firm, but flights beyond the premises would require staff to obtain certifications such as BVLOS approvals or remote pilot licenses. In practice, the partial adoption model is primarily driven by economic optimization.

Partial adoption is likely to be more suitable where full integration is constrained by regulatory uncertainty, infrastructure limitations or investment risk. It allows organizations to gradually build drone capabilities within selected warehouse functions or delivery segments, while maintaining operational flexibility and adapting to evolving regulatory and ecosystem conditions.

The inter-organizational collaboration model involves independent organizations establishing partnerships and pooling capabilities to implement specialized drone operations. For example, a warehouse could outsource last-mile delivery to operators offering drone-delivery-as-a-service (DDaaS) or outsource other core functions, such as inventory management or order picking. Given the novelty of the technology, warehouses often lack the organizational structure required for integrating drones (Maghazei et al., 2022), making third-party collaboration a practical path forward (Purtell et al., 2025). Inter-organizational models would become more prominent with increasing specialization, consolidation and the formation of accepted standards between stakeholders.

Although this model is intellectually rich, it is most prone to interface entropy. From a technological perspective, the major challenge to overcome is the drone-to-drone or drone-to-infrastructure handoff, which becomes a site of friction unless both actors have harmonized protocols, shared telemetry standards and real-time negotiation capabilities. Early-stage trials of this interoperability have been successfully conducted in Shenzhen (China) and Rotterdam (The Netherlands) in the context of coordinating ports' inbound drone freight with private last-mile drone operators. Organizationally, this model relies on flexible contracting arrangements, particularly for standardized or low-margin delivery segments. Large firms may outsource last-mile delivery where goods are of low value and customer information is not commercially sensitive, reducing the need for exclusive arrangements between warehouse operators and drone service providers. This points to a market-based coordination model, in which drone operators function as specialist carriers rather than being fully embedded within warehouse operations.

Such an arrangement may also reduce the need for immediate warehouse transformation, as dedicated sorting centers could enable drone carriers to dispatch goods more efficiently without requiring costly modifications to existing facilities. However, this model remains exposed to inter-organizational risks. Misaligned incentives, delayed information sharing, fragmented liability and differences in organizational culture can weaken resilience and reduce operational efficiency across the warehouse–drone interface. These risks are less pronounced in fully integrated models, where firms develop tightly coupled systems without relying on standardized application programming interfaces (APIs).

Inter-organizational models may offer efficiency gains for warehouse operators by optimizing delivery networks while introducing competition for last-mile delivery. For example, Amazon's road delivery operations rely on a network of delivery service partners (DSPs). An important design consideration concerns whether the interface or handover point between the two parties is fixed or mobile. For example, UPS has experimented with “motherships” concepts, using hybrid-electric delivery trucks equipped with drone launch and retrieval systems.

Inter-organizational collaboration is particularly pertinent where ecosystem coordination and specialized capabilities are required across multiple actors. It enables logistics providers and drone operators to leverage complementary expertise while reducing investment burden and technological risk. Collaborative arrangements also support shared infrastructure development and help organizations to adapt to evolving operational and airspace management requirements.

Our paper shows that successful adoption will require warehouses and drone operators to transform work practices across technological, organizational, environmental and inter-organizational–ecosystem domains. While barriers such as infrastructural constraints and regulatory coordination exist, both warehouse managers and drone operators see potential to integrate drone operations across the warehouse–last-mile delivery interface.

Our study makes several theoretical contributions to the literature on technology adoption in retail logistics. Following Makadok et al. (2018), we frame these contributions around the phenomenon examined, the level of analysis adopted, the constructs identified, the mechanisms explaining adoption and the boundary conditions under which drone-enabled logistics becomes more plausible and feasible.

First, our study conceptualizes drone adoption at the warehouse–last-mile delivery interface as an independent logistics phenomenon and an ecosystem-level coordination problem. Prior studies on technology adoption in logistics have largely focused on adoption within single firms or by individual users. While useful for firm-level technologies, these approaches do not fully capture drone adoption, as the latter cannot be explained without considering the wider network of actors involved (Maghazei et al., 2022; Purtell et al., 2025). Our findings show that drone logistics are inherently interdependent: adoption at the warehouse–last-mile interface depends not only on a focal firm's technological readiness but also on the alignment of supply chain partners, regulatory authorities and technology providers.

The second theoretical implication is that our study positions drone-enabled logistics at the inter-organizational level of analysis, where logistics service providers, drone service providers and regulatory actors must coordinate across operational boundaries. Applying the TOE framework to the warehouse–last-mile delivery interface reveals how organizational dependencies in drone adoption are shaped by ecosystem dynamics that extend beyond individual firm boundaries. Within the environmental dimension, drone adoption at the warehouse–last-mile delivery interface is shaped not only by exogenous factors (e.g. regulations, market trends) but also by endogenous variables such as infrastructural enablers and active collaboration between organizations. This highlights important interactions between the organizational and environmental dimensions of the TOE framework when innovations require coordination across multiple actors in the logistics ecosystem. This observation resonates with recent extensions of network theory (Bushell et al., 2025), which emphasizes the importance of stakeholder interdependencies, and with calls for further development of TOE theory to incorporate inter-organizational collaboration (Baker, 2011). Our results indicate that active stakeholder engagement with policymakers, supply chain partners and airspace regulators can lead to more realistic policies and pricing structures at the warehouse–last-mile delivery interface and the interconnected logistics ecosystem more generally.

The third implication is the identification of key constructs that shape drone adoption in logistics. Contributing to the literature on advancing TOE theory (Adade and De Vries, 2024; Ali et al., 2021; Wang and Zhang, 2024), our findings emphasize the significance of inter-organizational and ecosystem coordination in the context of logistics innovations. In particular, the findings identify stakeholder alignment, trust, collaboration, infrastructural readiness, regulatory engagement and governance model suitability as important constructs shaping drone adoption at the warehouse–last-mile delivery interface. Deploying TOE theory beyond static firm boundaries, our analysis demonstrates how technological, organizational and environmental dimensions interact in contexts where multiple actors must collaborate to adopt innovations. These findings capture the inter-organizational dependencies involved in adopting innovations in logistics, addressing a notable gap in the literature (Asare et al., 2016). By focusing on ecosystem coordination, our study moves beyond a firm-centric view of adoption and demonstrates how collective alignment, trust and coordination among multiple actors can influence the success or failure of drone adoption in retail logistics.

A further theoretical implication of our study is the identification of mechanisms and the boundary conditions that interact to shape the viability of drone adoption at the warehouse–last-mile delivery interface. Our findings suggest that the success of drone integration in logistics is not solely a function of one firm's technology, organizational capacity or external environment alone, but a collaborative outcome across the logistics chain. These mechanisms are reflected in three governance models for drone logistics: vertical integration, inter-organizational collaboration and partial adoption with intra-organizational collaboration. Examining these governance models through the interaction between the organizational and environmental dimensions of the TOE framework enables scholars to assess misalignment risks, identify where to focus primary coordination efforts and provide managers with guidance on selecting governance models based on adoption maturity and regulatory context. We have shown that the success of drone logistics relies not only on technological readiness or regulatory progress, but also on how firms organize themselves and collaborate across operational boundaries. Using real-world cases and drawing on our informants' experiences, we have considered the logistics chain as a whole rather than isolated segments, enabling us to explore the adoption and integration of drones in end-to-end logistics.

Our study also contributes to the retail and technology adoption literature by examining drone-enabled logistics as a B2B coordination problem at the warehouse–last-mile delivery interface. In the retail literature, prior work has emphasized the growing complexity of fulfillment models, increasing customer expectation and the role of technology adoption in improving warehousing and last-mile delivery performance. Our findings extend this discussion by showing that drone adoption in retail logistics depends not only on technological capability but also on effective collaboration among stakeholders. Our findings move beyond isolated warehouse use cases and technical demonstrations to examine how decision-makers assess the practical demand, risks, timing and implementation conditions for drone adoption at this crucial interface. In doing so, our study connects drone applications in warehouse functions with drone-enabled last-mile delivery, positioning the warehouse–last-mile delivery interface as a critical point of adoption, coordination and governance in retail logistics.

Our findings provide practical insights and actionable guidance for managers who are considering implementing drone technology at the warehouse–last-mile delivery interface. Building on Guan et al. (2023), we have shown that successful adoption to function effectively across stakeholders requires both formal management mechanisms (contracts, regulatory compliance, licensing and low-altitude airspace management integration) and informal management mechanisms (trust, shared risk management and cooperative planning). We show that drone adoption in retail logistics is constrained not by a lack of technical feasibility, but by the absence of shared operational standards and effective coordination among relevant stakeholders. For drone operations to create real value, drone company managers should aim to integrate their firms more deeply into logistics operations rather than functioning merely as transport providers. Warehouse operations managers should explore drone technology regarding current workflows, not just as a future concept and should avoid adopting drones solely for technological novelty. It is evident that drone operations cannot exist in a silo. Rather than solution spotting, there is a need to translate the great potential of drones into practical applications (Maghazei et al., 2022). We have further identified that collaboration between operators, such as by sharing real-time information about when a package is ready for collection, is important.

Regulatory constraints continue to be major barriers to scalability in drone delivery operations. Operators are subject to restrictions when flying in low-altitude airspaces. Collaborating with regulatory authorities and co-developing practical solutions would accelerate the adoption of drone delivery systems. As such, operators should not focus solely on the compliance aspect of regulation, but embrace solutions such as a low-altitude airspace management system to practically manage congestion in a network of their own aerial drones and perhaps more importantly, competitive and potentially non-cooperative, hostile drone operations.

In discussing governance models for the warehouse–last-mile delivery interface, regardless of the model pursued, an important consideration is how to fit drones into the broader logistics chain. Our findings suggest that the choice of governance model should align with firm size, adoption maturity and regulatory environment. Echoing one of our informants, we conclude that due to the complexities and the importance of local context, no single model will dominate globally. What we anticipate, and what aligns with the “ontological plurality” observed in other domains, is that deployment models will vary depending on legal maturity and cultural infrastructure (Pfaff et al., 2023; Tong et al., 2025). Aerial drone operations are not just a technology deployment problem; they involve the complex orchestration of trust, territory and transaction. Models that account for that multiplicity, especially those that allow flexible switching between modes based on context (congestion, load type, urgency, etc.), will likely emerge as more resilient. While vertical integration may be attractive to some large firms, many smaller operators and new entrants, and indeed, many large firms, may rely on successful intra- or inter-organizational collaboration to make drone operations work at the warehouse–last-mile delivery interface.

While our research has identified both enablers and barriers to drone adoption and integration at the warehouse–last-mile delivery interface, technology is advancing rapidly, and perceptions are changing. Our findings represent a snapshot in time. In addition, while our sample is global, some markets and use cases have not been covered. The cost of drone adoption and integration at the warehouse–last-mile delivery interface should be evaluated further for commercial viability. From a theoretical perspective, future research could further develop the application of the TOE framework, particularly from a longitudinal perspective. Investigating how organizational and environmental conditions evolve, especially in relation to governance models and inter-organizational coordination, represents a promising research avenue in the context of retail logistics and supply chains.

We thank the two anonymous reviewers and the Editors for their constructive comments and patience throughout the review process. We are also grateful to iMOVE Australia and Thales Australia for their financial support.

The supplementary material for this article can be found online.