Economic profitability of last-mile food delivery services: Lessons from Barcelona

https://doi.org/10.1016/j.rtbm.2021.100659Get rights and content

Abstract

During the last years, the number of digital platforms offering allegedly environmental sustainable last-mile logistics services has been increasing fast all over the world. Their size and geographical spread are growing leaps and bounds. Some studies suggest that they are still operating at losses and relying on venture capital to carry on growing. In this paper we employ real-life data gathered from the largest food delivery platforms (Just Eat, Glovo, and Deliveroo) operating in the city of Barcelona (Spain) to analyse the profitability of these business models. We develop a Monte Carlo simulation model with several scenarios to estimate how many orders are needed to reach economic profitability. Using this simulation model, a second model based on multiple linear regression is built to investigate the relationship between ‘the minimum number of orders required to reach profitability’ and several independent variables, such as the share of the total purchase order or the delivery time-distance. The potential use of this tool for managers is discussed, and several lines of future research on the economical profitability of food delivery operations are highlighted.

Introduction

In a matter of years, many traditional businesses have been affected by an unprecedented transformation of economic activity worldwide. The combination of business innovation, strategy, and new technologies has spurred the emergence of new business models, which in turn has contributed to greater overall performance of many companies (Bouwman, Nikou, Molina-Castillo, & de Reuver, 2018). While the magnitude of such correlation requires more research, it is clear that many economic sectors have opted to digitise and outsource different parts of their value chain. These processes allow companies to focus all their productive capacity on their core activities, ensuring the maintenance of high levels of productivity.

In the catering industry, this trend has manifested itself in a striking way. The incorporation of ICT into restaurants has altered the customer-restaurant relationship, especially with customers who prefer to order remotely. In 2019, the online food delivery sector surpassed, for the first time, the barrier of 100 trillion US dollars worldwide, maintaining a sharp upward trend. China alone accounts for approximately 37% of the total, followed by the US with just over 20%, and Europe with 15%. The old continent shows the highest growth rates, however, standing at 9.5% per year, with countries such as France and Spain growing at a rate of more than 10%.1 A major transformation of economic activity is taking place at a frantic pace. More and more customers are adapting their habits, compensating their lack of time to cook with online purchases of meals. Restaurant owners are aware of such behavioural change and are trying to adapt to contemporary society trends. The problem is that their business models must be reassessed. In general terms, there are two main models of physical home delivery services on the market. Restaurants must choose between internalising in-house the home delivery service or outsourcing it to companies whose expertise lies in last-mile logistic services, usually through allegedly environmentally friendly modes of transport.2

The number of food delivery digital platforms has been rapidly increasing during the last few years all over the world. In the case of food delivery services, digital platforms operate as marketplaces that allow consumers to buy meals from local restaurants. These platforms base their business model on keeping a share of the total purchase. In turn, they offer restaurants the possibility of outsourcing the distribution of meals to the customer. The service works as follows: once a purchase order is received, the platform sends a message to riders that meet specific criteria, such as geographical proximity and a favourable status on the platform. The first to reply is allocated the service. The meal is then picked up from the restaurant and there is a time limit to deliver the meal to the customer (usually one-hour). Of course, the sooner the delivery is completed, the sooner the rider is available again for accepting new orders. While riders are paid per delivery, platforms charge the restaurant a share of the meal order. They also charge the customer for the delivery. Although these platforms are constantly increasing their size, number of users, and geographical spread, many doubts exist about their economic and financial sustainability. Some experts suggest they are still operating at a loss, relying on venture capital and complementary business initiatives such as ghost (or dark) kitchens or dark stores to continue to grow (Li, Mirosa, & Bremer, 2020)..3 We develop a simulation model with several scenarios to study how their potential benefits evolve as the number of orders increases. Our objective is to determine the minimum volume of orders required to reach profitability, so we can assert the financial sustainability of these business models. Our baseline model provides an estimate of the minimum number of purchases needed to ensure economic profitability, as well as the threshold above which these companies can continue to scale up and spread to other markets. Knowing the changing nature of these companies, our research questions investigate (1) the impact of using different income settings, (2) the impact of hiring riders as employees instead of freelancers, and (3) the robustness of our model to the choice of specific parameters. We also perform a sensitivity analysis, and a multiple linear regression model was built to investigate the relationship between the minimum number of orders required to reach profitability and the independent variables used in our simulation model. The regression model proved highly sensitive to changes in specific parameters such as the restaurant fee, as well as other aspects such as the total customer revenue or the legal status of the riders. This regression model can provide useful advice to policy-makers: it can be used, for example, to study the potential effect of new policies on the economic profitability of these companies when one or more variables are altered. To the best of our knowledge, this is the first work to develop interconnected simulation and regression models to study the profitability of food delivery companies and provide a tool for related policy-making.

The paper is organised as follows. In Section 2, we present the existing literature on the topic and specify some geographical context. Section 3 establishes the definition of the problem. While Section 4 describes the real-life data gathered, Section 5 presents the computational approach we followed to address the problem. Section 6 then exposes the results we obtained and discusses them according to current understanding of this phenomenon. Finally, Section 7 summarises the main findings and points to a number of open research lines.

Section snippets

Research context and related work

In this section, the context of our study is described in further detail, and a short review of related academic work is provided.

Profit function of food delivery digital platforms

One of the main goals of this study is to estimate the profitability threshold at which food delivery platforms become cost-effective. The profit function of such firms is given as follows:Pvxf=IvxfExf,where P represents the profit of the digital platform firm, I represents its total income (estimated as the sum of revenue from restaurants and customers), E stands for total expenses (estimated as the delivery cost plus the fixed operating cost of the company), v = (v1, v2, …, vn) is a vector

Descriptive analysis of real-life data

This section provides a descriptive analysis of real-life data from takeaway food delivery platforms in the city of Barcelona (Spain). The analysis was performed with the programming language Python (3.7.2) and the following libraries: Pandas (0.23.4), Matplotlib (2.2.3), and Seaborn (0.9.0). The dataset was obtained from three of the main platforms operating in Barcelona: Glovo, Deliveroo, and Just Eat. To ensure a spatially homogeneous distribution of data, we selected 10 different locations

Computational approach: A Monte Carlo simulation model

In order to study the economic profitability of the food delivery firms, we developed a Monte Carlo simulation model (Rubinstein & Kroese, 2016). Simulation is a powerful and flexible methodology frequently employed in fields as diverse as transportation and logistics (Gruler, Panadero, de Armas, Moreno, & Juan, 2020) and reliability engineering (Faulin, Juan, Serrat, & Bargueño, 2008). By using simulation, we can consider the effect of different random variables (e.g., order quantities and

Results and discussion

This section introduces the results obtained for our baseline model and the different scenarios considered. These results are then discussed in relation to the existing literature and professional knowledge of the field.

Conclusions

This paper analyses the economic profitability of food-related last-mile delivery firms. To do so, we rely on real-life data scraped from companies operating in the city of Barcelona, specifically Glovo, Deliveroo, and Just Eat. Upon these data, we built a Monte Carlo simulation model using Excel/VBA, in which different scenarios were considered to assess the minimum number of orders required to overcome platform's operational costs. We define economic profitability as the difference between

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We would like to thank our research assistant, Josep Reixach, for his technical support with this article. This work has been partially supported by the Spanish Ministry of Science, Innovation, and Universities (PID2019-111100RB-C21/AEI/10.13039/501100011033).

References (66)

  • T. Letnik et al.

    Review of policies and measures for sustainable and energy efficient urban transport

    Energy

    (2018)
  • C.J. Martin

    The sharing economy: A pathway to sustainability or a nightmarish form of neoliberal capitalism?

    Ecological Economics

    (2016)
  • L.G. Marujo et al.

    Assessing the sustainability of mobile depots: The case of urban freight distribution in Rio de Janeiro

    Transportation Research Part D: Transport and Environment

    (2018)
  • J.B. Oliveira et al.

    Perspectives and relationships in supply chain simulation: A systematic literature review

    Simulation Modelling Practice and Theory

    (2016)
  • B.S. Onggo et al.

    Agri-food supply chains with stochastic demands: A multi-period inventory routing problem with perishable products

    Simulation Modelling Practice and Theory

    (2019)
  • J.P. Rodrigue

    The distribution network of amazon and the footprint of freight digitalization

    Journal of Transport Geography

    (2020)
  • G. Schliwa et al.

    Sustainable city logistics—Making cargo cycles viable for urban freight transport

    Research in Transportation Business & Management

    (2015)
  • J. Visser et al.

    Home delivery and the impacts on urban freight transport: A review

    Procedia-Social and Behavioral Sciences

    (2014)
  • L.v. Alberti-Alhtaybat et al.

    A knowledge management and sharing business model for dealing with disruption: The case of Aramex

    Journal of Business Research

    (2019)
  • Y. Benkler

    Sharing nicely: On shareable goods and the emergence of sharing as a modality of economic production

    Yale LJ

    (2004)
  • R. Bostman et al.

    What’s mine is yours. How collaborative consumption is changing the way we live

    (2011)
  • H. Bouwman et al.

    The impact of digitalization on business models

    Digital Policy, Regulation and Governance

    (2018)
  • D.J. Bowersox et al.

    Simulation in logistics: A review of present practice and a

    Journal of Business Logistics

    (1989)
  • C. Cant

    Riding for Deliveroo: Resistance in the new economy

    (2019)
  • V. Carbone et al.

    The rise of crowd logistics: A new way to co-create logistics value

    Journal of Business Logistics

    (2017)
  • C. Chen et al.

    Using the crowd of taxis to last mile delivery in e-commerce: A methodological research

  • P. Chen et al.

    Crowdsourced delivery for last-mile distribution: An agent-based modelling and simulation approach

  • V. De Stefano et al.

    European legal framework for’digital labour platforms’

    (2018)
  • J. Drahokoupil et al.

    Work in the platform economy: Beyond lower transaction costs

    Intereconomics

    (2017)
  • R. Ducruet

    Parcel deliveries and urban logistics: Changes and challenges in the courier express and parcel sector in europe: The french case

    Research in Transportation Business & Management

    (2014)
  • A. Estrada-Moreno et al.

    A biased-randomized algorithm for redistribution of perishable food inventories in supermarket chains

    International Transactions in Operational Research

    (2019)
  • European Commission

    The Challenges of Impact Assessment

    (2010)
  • S. Gonzalez-Martin et al.

    A simheuristic algorithm for solving the arc routing problem with stochastic demands

    Journal of Simulation

    (2018)
  • Cited by (0)

    View full text