Collection and delivery points for e-commerce last-mile logistics: a study of consumer acceptance using a unified theoretical framework

Abstract

Last-mile delivery is the most cost-intensive and environmentally harmful segment of the e-commerce supply chain. It faces persistent challenges such as delivery failures, congestion, and high emissions, particularly in urban areas. In India, per-parcel emissions are significantly higher than global averages, and parcel volumes are expected to grow sharply in the coming year reaching over 40 billion annually by 2030. The 2.6 billion parcels delivered in 2022, produced around 500,000 tonnes of CO₂ emissions in India. Globally, the carbon footprint of the last-mile delivery sector is equivalent to emissions produced by one million petrol-powered vehicles as per 2021 estimations. Financially too, the last-mile segment is equally critical, accounting for an estimated 40 to 50 percent of overall delivery costs. To address these concerns, innovative delivery models such as Collection and Delivery Points (CDPs) have emerged. CDPs offer operational benefits by reducing failed deliveries and enabling parcel consolidation. They also provide secure, flexible pickup options, thus improving both efficiency for logistics players and convenience for consumers. Despite their promise, CDPs remain underexplored in India. Most existing research is concentrated in high-income countries, with limited understanding of how consumers in emerging markets perceive such alternatives. The literature also highlights the lack of theoretically grounded research in the context of last-mile delivery innovations. Moreover, the interaction between individual-level attributes and broader contextual factors appears to have received limited attention in existing research. Thereby, the following research question was formulated: How do behavioural factors, individual attributes, and broader contextual factors shape consumer’s intention to use Collection and Delivery Points (CDPs) as an alternative to home delivery and return of e-commerce parcels within the Indian context? In order to systematically address the research question, three specific objectives were formulated. The first objective is to review the existing literature on consumer acceptance of Collection and Delivery Points (CDPs), with a focus on identifying relevant constructs and empirical gaps. The second objective is to examine how behavioural factors, drawn from established theoretical frameworks, influence consumers’ intention to use CDPs in a context where such services are not yet operational. The third objective is to explore the role of demographic characteristics, online shopping behaviour, and built environment attributes in shaping consumer intentions, thereby incorporating both individual-level and contextual dimensions into the analysis. This study focuses on the Indian city of Guwahati to analyse consumer intentions toward CDPs, using a mixed-methods design. The thesis positions itself at the intersection of two conceptual paradigms: logistics innovation and value co-creation. First, it recognizes Collection and Delivery Points (CDPs) as a logistics innovation that reconfigures traditional last-mile delivery models by shifting from direct-to-door distribution to decentralised service point fulfilment. This transition holds potential for enhancing operational efficiency and reducing delivery-related externalities. Second, the study draws on the value co-creation paradigm, viewing CDPs as part of a broader category of self-service technologies that involve active consumer participation. In this model, users are not merely recipients of services but co-producers who contribute to service delivery by undertaking tasks such as parcel collection and returns. To address Objective 2, which focuses on behavioural factors, the study employed Partial Least Squares Structural Equation Modelling (PLS-SEM) to examine how selected latent constructs influence consumers’ intention to use Collection and Delivery Points (CDPs). Five antecedent behavioural constructs were hypothesised to influence behavioural intention: Habit, , and Social Influence were derived from Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) and Perceived Relative Advantage from Diffusion of Innovation Theory; Perceived Risks was based on Risk Theory; and Perceived Neighbourhood Environment was drawn from built environment and neighbourhood perception literature. Each construct was operationalised using multiple observed indicators. Objective 3 was addressed using binary logistic regression to assess the influence of demographic variables, online shopping variables, and built environment attributes on consumer intention. Separate models were estimated for service points and parcel lockers. Significant latent variables from the PLS-SEM analysis were included as composite scores. To enrich the analysis and mitigate the limitations of purely quantitative modelling, a thematic analysis of open-ended interviews was conducted. Using a hybrid deductive - inductive coding approach, the qualitative analysis provides deeper contextual insights and captures spatial variations in consumer perspectives across selected localities. The geographical scope is bounded by two levels of analysis: (1) the entire metropolitan region and (2) selected localities within it. The quantitative analysis presented is conducted at the scale of the metropolitan region, while the qualitative approach focuses on selected nine localities within the region. In both units, the unit of observation is the individual. The sample size for the quantitative analysis was 166, while for the qualitative analysis was 34. The hypothesis testing in PLS-SEM revealed that Perceived Neighbourhood Environment had the strongest influence on intention to use CDPs, followed by Relative Advantage, Habit, and Social Influence. Perceived Risk had a negative as hypothesised but statistically insignificant effect. The model demonstrated strong explanatory power, and bootstrapped confidence intervals affirmed the model’s robustness and predictive accuracy. The logistic regression analysis revealed distinct predictors for service point and parcel locker adoption. For service points, adoption intention was higher among individuals with lower educational attainment, greater satisfaction with delivery experiences, loyalty programme membership, and frequent high-value purchases. Built environment factors such as living in government quarters or independent houses were negatively associated with adoption. Behavioural constructs, particularly Habit and Relative Advantage, significantly increased the likelihood of service point usage. In contrast, parcel locker adoption was lower among graduates and postgraduates, and among urban residents or those living near high-traffic streets. Social Influence emerged as a significant positive predictor, while Perceived Neighbourhood Environment showed a marginal positive effect. This underscores the need for differentiated planning and communication strategies tailored to each CDP format. The qualitative analysis indicated that dissatisfaction with current home delivery, particularly failed deliveries and dependency on human presence, drives interest in Collection and Delivery Points (CDPs), especially in areas such as AEC Campus and Ambari. Perceived benefits including convenience, reliability, and reduced theft risk emerged as key enablers, with service points generally viewed more favourably than parcel lockers. However, satisfaction with existing services and concerns related to safety and privacy presented notable barriers, particularly in Bamunimaidam and Ambari. Some of these concerns, such as unfamiliarity with the CDP concept and privacy issues, appear to be addressable through targeted awareness and thoughtful design. Participants also expressed preferences based on product type and timing, highlighting the importance of flexible operational features. Location suitability was a consistently important theme, with preferences centring on proximity, walkability, and integration with daily routines. Spatial variation in responses indicated differing levels of awareness and readiness across localities, underscoring the need for place-based implementation strategies. Overall, adoption decisions were shaped by the interplay of behavioural perceptions, service experiences, and daily commuting patterns. The study make several theoretical contributions by adapting adoption models like Diffusion of Innovation and UTAUT2 to a developing country context. It introduces Perceived Neighbourhood Environment (PNE) as a key factor, showing that spatial context shapes behavioural intentions. The findings highlight the continued relevance of habit and social influence, while challenging the assumption that perceived risk always hinders adoption. By comparing service points and parcel lockers, the study shows service format influences adoption drivers. The integrated framework offers a context-sensitive, consumer-centric understanding of Collection and Delivery Points acceptance in India. This study offers practical guidance for scaling CDPs in Indian cities. Marketing should target active online shoppers with tailored messaging for different socio-economic groups. CDPs must be placed in safe, accessible, and familiar locations to align with users’ daily routines. Services should support local preferences like Cash on Delivery and provide reliable, user-friendly experiences to encourage habit formation. Incentives, trials, and education can increase adoption, while integrating CDPs with transit hubs supports convenience and sustainability. Policymakers should embed CDPs in urban planning and logistics policies, backed by multi-stakeholder collaboration for effective deployment. The study concludes by outlining the limitations and future research directions. Overall, the study advances theoretical understanding and offers practical pathways for the deployment of Collection and Delivery Points in India and similar contexts. Keywords: Collection and Delivery Points, E-Commerce, Urban Logistics, Last-Mile Delivery, Consumer Behavior