Configuring ethical food consumers: understanding the failures of digital food platforms

ABSTRACT

Digital food platforms are being launched, offering consumers convenient access to ‘ethical’ food. Ensuring the viability of these platforms hinges on attracting and retaining loyal customers. However, recent studies show that despite considerable efforts from market actors, stable platform-consumer relationships are rare in these digitalized food markets. The aim of this paper is to explore and explain why digital food platforms fail to produce stable ethical consumers. The paper draws on an ethnographically inspired study of the meal box market to explore the socio-material configuring of consumers and the resulting consumer arrangements. The analysis shows that while the market devices of meal box providers worked to produce loyal and ethical food consumers, the consumers in this study were not loyal to any one meal box provider. This instability, the paper argues, was the result of both a restless market and restless households. The multiple, often conflicting, and, in some cases, unspecific consumer configurations intensified the households’ dynamic tendencies, leading to changes in diets and sustainability focus, the breaking of routines, and the continuous chasing of offers. Because of this, the consumer arrangements formed tended to unravel over time, making the production of stable ethical food consumers difficult.

KEYWORDS:

• Ethical consumption
• digital platforms
• marketing
• food consumption
• consumers
• configuring

Introduction

Food markets are being reorganized by digitalization. Digital devices such as smartphones, tablets, and computers are increasingly becoming intertwined with everyday life, giving us access to websites, web shops, social media, QR codes, and other digital artefacts (Cochoy et al. 2017). Among other things, the digitalization of everyday life has meant that consumers are now reachable in multiple ways throughout the day. Not only company-consumer communication, but also consumer-to-consumer communication has been greatly facilitated, and as a result intensified. Food companies have now new means to reach, entice, and connect with consumers, often with little effort and at a low cost, which has given way to several new modes of food provisioning and food consumption. Meal box schemes, consumer-to-consumer food swapping networks, food sharing apps, and online food stores are all made possible by these new digital devices. Digitalization has set the scene for the contemporary reorganization of food markets.

These newly launched digital food platforms are often also marketed as promoters of sustainable and ethical food consumption: some examples include meal box schemes promoting vegetarian consumption, food swapping networks to reduce food waste, and online stores focused on the marketing of ecological food products. The assumption made by those who design and launch these new digital platforms is that they will be able to reconfigure food consumption, enabling various forms of ethical food consumption in the process.

Crucial to these efforts to digitally promote ethical food consumption is the issue of recruiting consumers and turning them into recurring customers. These new digital food platforms are designed to capture consumers by enticing them to break with their normal, often highly routinized food practices, and try a new mode of acquiring food (Samsioe and Fuentes 2021). Many of these digital food platforms offer subscription services and are designed to encourage and establish long term relationships with their customers; they seek to reinforce the platform-consumer relationship.

However, recent studies show that despite the new opportunities afforded by digitalization and considerable marketing efforts from market actors, stable platform-consumer relationships are rare in these new digitalized food markets (C. Fuentes, Cegrell, and Vesterinen 2021; C. Fuentes and Samsioe 2021). While consumers do buy food digitally and engage in ethical food consumption – consuming, e.g. more ecological food, shopping local food, or reducing their meat consumption – they neither stick to one digital food provisioning platform nor a single ethical diet. Instead, they switch between them, often changing both food ethics and preferences in the process, much to the dismay of food platform owners. Why is this so? Why are these market actors, and the various market devices they mobilize, incapable of turning consumers into stable ethical food customers? Why do these efforts to promote ethical food consumption fail?

In trying to answer these questions, this paper takes a somewhat different theoretical approach. Ethical consumption research (both the psychological and sociologically influenced research) tends to take for granted the existence of ethical consumers (C. Fuentes and Sörum 2019). Whether ethical consumers are seen as a social and cultural meaning maker or individual rational decision-maker, powerful actors capable of shaping markets or a powerless victim to marketing and market actors, these studies assume the existence of ready-made ethical consumers ‘out there’ with a set of built in capacities (Stigzelius 2017).

In the present paper, I will turn the issue around and take the stance that all consumers, ethical or not, must be made. Starting from this vantage point, the success or failure of new digital food platforms is not a question of finding and targeting the ‘right’ group of consumers but rather of using these new digital platforms to configuring consumers (Woolgar 1998) to consume food ethically. Consequently, in this paper, understanding the success or failure of digital food platforms is tied up with the understanding the work and processes of consumer configuration.

Therefore, the aim of this paper is to advance our understanding of digitalized food consumption by empirically examining and conceptualizing the digital food platforms’ efforts and failure to produce stable ethical consumer arrangements.

Drawing on and combining Science and Technology Studies (STS) and market studies literature, I take an interest in the socio-material configuring of consumers performed by digital food platforms. More specifically, this paper sees the configuration of consumers (Woolgar 1998) as the work of market devices, which are used to both construct an identity and enable and delimit a set of actions for the consumer (for a similar argument, see C. Fuentes and Fuentes 2017).

The analysis of consumer configuring developed is based on an ethnographic study of 15 households that signed up to meal box schemes. The study combines ethnographic interviewing, on-site kitchen observations, informant digital walkthroughs, and close readings of digital devices to produce rich data regarding how different meal box providers work to configure consumers in efforts to produce specific customers of their services.

The analysis shows that while the market devices of Meal box providers are organized to produce stable ethical food consumers, these efforts fall short as the meal box consumers in this study switch both meal box provider and modes of food consumption. The paper argues that the instability of these consumer arrangements can be explained both by the dynamics of household food consumption and the marketing efforts of the Meal box companies which, in their efforts to attract meal box consumers, end up promoting and enabling switching consumers. Because of this, the consumer arrangements sought tend to unravel over time, making it impossible to produce stable ethical food consumers.

The work of marketing: capturing and configuring consumers

Marketing is often conceptualized as a set of ideas, techniques, and practices aimed at identifying and creating value for a segment of consumers or target market at a profit (Grönroos 2008; Kotler et al. 1999). Mainstream marketing sees marketing as a value creating activity, but often assumes that consumers or groups of consumers already exist and it is merely a matter of finding them, targeting them, uncovering their needs and wants, and then satisfying these in ways that creates value for consumers and the firms/organizations providing the service or product.

In contrast, critical marketing scholars, Consumer Culture Theory (CCT) scholars as well as scholars from sociology, anthropology, geography and the broader social sciences have shown that consumers do not exist ‘out there’ independent of marketing but are instead, at least partly, constructed by marketing practices, discourses, and devices (e.g. Caruana and Crane 2008; C. Fuentes and Sörum 2019; Giesler and Veresiu 2014; Stigzelius 2018). So, while the notion that consumers are made in and through marketing is not part of the mainstream marketing literature, it is far from novel.

One of the areas in which this notion has gained traction in recent years is via constructive market studies (CMS). This interdisciplinary body of research, focused on understanding markets and their practical and material organization, has taken an interest in how consumers, as market actors, are made and gain agency (Cochoy 2007, 2008; Harrison and Kjellberg 2016; Kjellberg 2008). This research is often preoccupied with the practices and processes behind the formation of consumers as market actors and how these actors in turn work to shape markets in various ways and through various means.

However, while consumers and their making as actors has started to receive some attention within market studies, this is still a relatively under-researched area (Hagberg 2016). Moreover, with a few exceptions (see, e.g. Sörum 2020), studies looking into the making of consumers have focused on more ‘successful’ consumer constructions, exploring, for example, how market devices work in order to enrol consumers in the consumption of vintage goods (Brembeck and Sörum 2017) or how food packaging configures consumers to promote the emergence of alternative food markets (C. Fuentes and Fuentes 2017).

Inspired both by STS (Latour 2005; Woolgar 1998) and market studies (Cochoy 2007; 2015; C. Fuentes and Fuentes 2017), I outline below a theoretical approached to discuss how marketing devices work to configure consumers in different ways and to varying degrees.

Two concepts are of particular importance here: that of a market device and that of configuration. Now a commonly used concept in CMS, a market device is defined as ‘material and discursive assemblage’ involved in the making or arranging of markets (Muniesa, Millo, and Callon 2007, 2). Market devices can be artefacts, such as display arrangements, smartphones, or marketplaces, but also concepts and other less tangible entities (McFall 2015). Of particular importance here is how market devices interact with and shape consumers. While consumption is an underdeveloped topic in CMS, there is a growing body of work that explores the ‘devising of consumption’ (e.g. C. Fuentes and Samsioe 2021). This research has shown us that market devices can play a key role in the capture and shaping of consumers. For example, Cochoy (2008) argues and shows how mundane market devices such as shopping carts shape consumers calculative capacities but also how packaging can work to ‘capture’ consumers (Cochoy 2004). In a similar vein, Fuentes and Sörum (2019), analyse the role of smartphone apps in the shaping, or agencing, of ethical consumers. Focusing on the electric market, Grandclément and Nadaï (2018), show how the electric metre, a mundane market device, is inscribed with consumer configurations and then come to, partly, shape electric consumption. Market devices thus, in their work of arranging markets, also shape consumption, materially and semiotically (McFall 2015). The present paper follows this line of reasoning as it explores meal box schemes as market devices designed to configure consumers in different ways.

The second concept, configuration, is an often used but seldom explicitly defined concept. Originating in the STS literature, it has been used across the social sciences to discuss how products and material artefacts are inscribed with ideas of users by designers, marketers, and other actors and how these inscriptions in turn shape users and their range of possible actions. Examples include studies of how users¹ are configured in the development of electronic communication networks (Oudshoorn, Rommes, and Stienstra 2004), of how virtual worlds configure the child player (Grimes 2015), how the design of shavers configure users’ gender (Ooust 2003), and how the user-citizens of digital cities are configured (Lieshout 2001). It has also been used more broadly to, for example, discuss how medical research configure users of functional foods (Weiner 2010) or how teachers are configured as data test users (Ratner, Andersen, and Madsen 2019). Within the CMS literatures the concept has instead mainly been used to discuss the configuring of market actors (e.g. Andersson, Aspenberg, and Kjellberg 2008; Harrison and Kjellberg 2016). Here it has been closely linked to the ambition to understand how markets are made and, in line with this, how market actors come to be configured. In line with this, the present paper focuses on the (failed) configuration of meal box user as a specific type of consumer.

What is then meant by configuration? Drawing on an ethnographic study of computers, Woolgar explains that ‘configuring includes defining the identity of putative users, and setting constrains upon their likely future actions’ (Woolgar 1998, 59). To configure a user is to set parameters for their (future) actions; to make some actions more likely than others through material-semiotic means.

The concept of configuring clarifies that it is not only the identity of the user which is defined but that material – or design – elements define and delimit the range of possible user actions (Woolgar 1998). Configuring users is therefore both about defining who they are and what their possible actions should be.

In configuring the user, the architects of DNS, its hardware engineers, product engineers, project managers, salespersons, technical support, purchasing, finance and control, legal personnel and the rest are both contributing to a definition of the reader of their text and establishing parameters for readers’ actions. (Woolgar 1998, 69)

Configure means to arrange something into a particular form, and to configure users is, in a sense, to arrange users, materially and semiotically, in a particular way.

The scholars that have followed Woolgar and put the concept to use have also criticized and expanded on it, arguing, and showing that the configuration of users can involve multiple practices and actors and that configuration can take the form of text, images, and material design. Talking or writing about consumers within a company is one way of configuring consumers; these accounts are attempts to define the identity and actions of consumers (Oudshoorn et al. 2004). In the same way, writing instructions on a webpage dictates the proper mode of ordering something. These instructions both presuppose and prescribe who the user is and how they should act. User configuration can be based on consumer research (focus groups, for example, or big data), service interactions with consumers (support or retail practices for example), or ideas reproduced within the organization regarding who the company’s consumers are (Oudshoorn et al. 2004). An app or website, for example, configures users both regarding what they can do – possible actions on the site – and how they should understand/frame their actions – the message that the actions are accompanied by (C. Fuentes and Sörum 2019).

While many studies of consumer configuration have tended to focus on the imagined/presupposed users/consumers enacted and materialized through design, marketing and market research practices (e.g. Ooust 2003; Schneider and Woolgar 2012), scholars have also discussed and investigated how and to what extent users are configured (Mackay et al. 2000). That is, studies have at times investigated both the encoding and decoding of configurations of users, and explored how and to what extent their identities and possible actions were shaped by the imaginary users built into material devices, systems, instructions, and other artefacts. These studies have sought to understand not only efforts to arrange a specific type of consumers but also the resulting arrangements.

Those that have explored how users are shaped by the user configurations thought up, described, and materialized by designers, marketers, and other actors, have noted a number of issues that are of interest for the analysis that follows.

To begin with, configurations can be of varying degrees of specificity. In some cases, you don’t want the consumer configuration to be too specific as to exclude potential users. This was, for example, the case with the development of a computer described by Woolgar (1998). The company behind the new computer wanted it to be used by various consumers and therefore the consumer configuration could not be so specific as to exclude an important category of users. However, in other cases, companies developing a product, or a service go to great lengths to adapt their offerings to specific types of consumers. In these cases, the idea is to make a consumer configuration as specific as possible, creating a close connection between product and user and even ‘personalize’ the service or product.

Configurations can also be of varying degrees of stability (C. Fuentes 2019; Ooust 2003). During the development of a product or service, the user configuration is typically fluid. Exactly what type of consumer is expected to use the service or product is still under negotiation (Woolgar 1998). During these times the consumer configuration is more malleable; the type of user built into the offering is still subject to changes. Similarly, configurations can be more or less stable in the user phase. Certain consumer arrangements stabilize over time, becoming difficult to change, while others remain unstable, continuously, and sometimes rapidly changing.

Configurations of users can also be in conflict. The configuring of users can be a complex process shaped by multiple actors, such as designers, marketers, policy makers, and users themselves. These various actors often draw on and work to reproduce different and, at times, conflicting user representations (C. Fuentes and Fuentes 2017; Oudshoorn et al. 2004).

Finally, while producers, designers, and marketers and the devices they develop can configure consumers, consumers or users are also capable of configuring devices, marketers, designers, and producers (e.g. C. Fuentes 2019). As previous research suggests, consumers can play a key role in the configuring of market actors and the enactment of markets (Harrison and Kjellberg 2016). The relationship between those who configure and those who are configured is therefore not always clear cut and configuration can go both ways (C. Fuentes and Samsioe 2021; Mackay et al. 2000).

Taking this into account, it becomes clear that configurations of the user made in design or marketing do not necessarily work to shape consumers’ actions and identities fully or even partly. It is, as others have pointed out when extending Woolgar’s notion user configuration, completely possible, and some would argue to be expected, that configurations of users made in design and marketing are challenged, adapted, or simply ignored by users (see, e.g. Mackay et al. 2000). Or phrased differently, efforts to arrange consumers in specific ways do not necessarily result in the envisioned arrangements.

Studying meal box schemes: an ethnographic approach

This analysis draws on an ethnographically inspired study of meal box schemes in Sweden. The study combines digital observations of the digital platforms providing the meal boxes, ethnographic interview with 15 households, and kitchen observations and digital walkthroughs conducted during the interviews.

In this study, we were interested in if and how meal box schemes, as a digitalized mode of food provisioning, became part of and shaped household practices. Results from this study have also been published in C. Fuentes and Samsioe (2021). We wanted to understand not only the use of meal box schemes but also, more broadly, how the households acquired, stored, prepared, ate, and disposed of foods. Ethnographic methods and a combination of various data collecting techniques, both on and offline, were key to generating detailed data on this and achieving a form of co-presence (see also, Tanja and Eli 2021).

The fieldwork was carried out by the author and a post-doc² over a period of four months in 2019 – between the 18th June and the 20th September. Using our personal network, social media, and referrals from participants, we recruited fifteen households for the study. The participants either used or had recently used a meal box service. Apart from this criterion, we sought variation regarding meal box schemes used, family composition, educational levels, occupation, and diet and food preferences (see Table 1). For practical reasons, participants all resided in urban areas in the southern part of Sweden. The resulting sample shows some variation regarding occupation and educational level; the households participating are mostly Swedish middle-class families or couples living in urban areas. This is by no means a representative sample (nor is it intended to be) but does align well with consumers targeted by meal box services and with the reported typical purchaser of online food (Digital mathandeln 2018).

Table 1. Study participants appear also in C. Fuentes and Samsioe (2021).

Participants (Pseudonyms)	Type of household	Number of household members	Occupation	Dietary preferences	Retailer, service provider former/current
Anders	Family	3 (1 child 16 years old)	Reseracher and teacher, University	Try out vegetarian & LCHF	Linas Matkasse (Flexitarian box) & City Gross (Family box)/Linas Matkasse (Flexitarian box)
Nicole & Emil	Family	4 (2 children 11 and 8 years old)	Shop manager/ Football coach	Everyday meals	Mathem (Family box)
Linda & Markus	Couple	2 (no children)	Teacher, high school	Keeping to Weight watchers diet	City Gross (Weight watchers box)
Susanna	Couple	2 (no children)	Reseracher and teacher, University	Vegetarian, Vegan and anti-Inflammatory foods	Årstiderna (Vegan food box) /Årstiderna (Fruit box)
Anna & Niklas	Family	5 (3 children 18, 16 and 10 yeras old)	Administrator, education/ Sustainability manager	Everyday meals, lactose intolerance	Linas Matkasse (Everyday classic box) & MatHem (Family box)/Linas Matkasse (Everyday classic box)
Sofie	Family	4 (2 children 9 and 7 yeras old)	Administrator, healthcare	Try out vegetarian	City Gross (Inspirational flexitarian box)/ ICA (Inspirational box)
Hanna	Family	4 (2 children 5 and 3 years old)	Payroll consultant	Everyday meals	City Gross (Family box)/ ICA (Family Box)
Lisa	Family	4 (2 children 6 and 4 years old)	CEO Start up incubator	Try out vegetarian	City Gross (Family box)/ Linas Matkasse (Flexitarian box)/ Linas Matkasse (Everyday classic box)
Johanna	Family	4 (2 children 4 and 2 years old)	VP Employee Sucess	Try out vegetarian	MatHem (Family box)/Linas Matkasse (Flexitarian box)/ICA (Vegetarian box)
Lotta	Family	4 (2 children 8 and 4 years old)	Education manager, folk high school	Everyday meals and try out Weight watchers diet	ICA (Healthy box)/ City Gross (Weight watchers box)
Cajsa	Family	5 (3 children 11, 9 and 6 years old)	Employment officer	Try out vegetarian	ICA (Family box)
Anne	Couple	2 (no children)	Retired	Everyday meals	City Gross (Everyday classic box) & ICA (Family box)/City Gross (Everyday classic box)
Tommy & Louise	Couple	2 (no children)	Employment officer/ Unit manager healthcare	Try out vegetarian	Coop (Healthy box)/ City Gross (Weight watchers box)
Josefine	Family	3 (1 child 17 years old)	Administrator, personnel	Everyday meals	Linas Matkasse (Everyday classic box)
Isabelle	Family	4 (2 children 8 and 2 years old)	Custumer service manager	Everyday meals	Linas Matkasse (Family Box) & ICA (Family box) & City Gross (Family box)

The ethnographic interviews conducted were focused on understanding a set of practices and their meanings (Spradley 1979). That is, when designing and conducting them we were interested in generating both accounts of practices – i.e. how the participants purchased, stored, prepared, ate, and disposed of food – and the meanings ascribed to these food-related practices by the participants. The interviews were designed to be informal and semi-structured, guided by ‘grand tour’ questions which were complemented with follow-up questions. As part of our methodological approach, the interviews were conducted in the participants’ homes, either in or near their kitchens. The interviews lasted between 60 and 90 min and were audio recorded and transcribed in full. While the interviews were typically conducted with those mainly responsible for ordering and managing the meal boxes, it was not uncommon for other household members to join in. The interviews covered both the multiple actions connected to the meal boxes (e.g. interactions with websites/app used to order the meal boxes, the unpacking of the meal boxes, how the meal box food was prepared, eaten, and disposed of) and the households general food practices (shopping, storing, preparation, eating and disposal of their food).

The interviews were complemented with kitchen observations and digital walkthroughs. The kitchen observations typically started as the interviewer asked the participant for a ‘tour’ of the kitchen. While the objective was not a systematic inventory (Hebrok and Heidenstrøm 2019), the participants typically showed us and allowed us to photograph their cupboards, fridges, and freezers (see also, Evans 2012). More than 180 photographs were taken during the kitchen observations. The kitchen observations were key to generating material regarding the practical and material reorganization of food provisioning and consumption that resulted from the introduction of meal box schemes. These observations allowed us both to directly observe (part of) the materialities involved and enabled us to ask more detailed questions.

Furthermore, we also conducted researcher assisted digital walkthroughs with research participants (Light, Burgess, and Duguay 2018). Adapting this method to an interview setting, participants were asked to demonstrate how they used the digital interface when ordering, changing orders, or conducting other meal box related interactions. During this demonstration, the interviewer also asked questions regarding the various elements and functions of the websites/apps, prompting participants for both successful and unsuccessful examples of digital interactions. The walkthrough method interview generated detailed accounts of the interactions with the digital interface of the meal box providers. It also allowed us to observe how consumers logged in to the meal box websites/apps, how these were browsed and used, and to explore the difficulties consumers encountered when navigating these digital devices. These observations were documented on video.

Finally, we also conducted digital observations of the meal box websites and apps that the participants used or had used in the past. Here, focus was on understanding the scripts of these digital devices. The assumption made here is that these digital devices, as with other designed material artefacts, are inscribed with specific programmes of actions: a certain set of actions that the user is intended to perform (Akrich 1992). Researcher then ‘de-scribe’ these artefacts, describing the actions that the device prompts as well as the message used to frame them as meaningful (see, e.g. C. Fuentes and Sörum 2019). In total, six meal box providers were studied generating around 300 screenshots.

The material was coded and analysed by the author, guided by the constant comparative method inspired by grounded theory (Charmaz 2006) and using Nvivo. Guiding the analysis were two overarching questions: What consumer configurations can be read when studying the meal box platforms and their multiple marketing devices (i.e. what actions did these devices encourage and enable consumers to perform? How were these actions framed as meaningful to consumers?)? How do these consumer configurations shape food consumers’ identities and actions (i.e. what did consumers do, how did they framed their actions as meaningful, and how did this relate to the configurations of the platforms?)? The first question was answered mainly by analysing the digital observations of the websites, apps, and social media accounts, while the second question was addressed mainly by analysing the consumer interview transcripts, the videos of the digital walkthroughs, and photographs of the household taken by the researchers or provided by informants.

Using these questions, I developed a set of interlinked analytical categories that analyse the meal box platforms marketing work as well as how these platforms become part of or fail to become part of consumer households’ food practices. More specifically, an analysis is developed that centres the meal box platforms efforts and subsequent failure to produce stable ethical food ‘consumer’ arrangements. Below, I present the results of this analysis, using the empirical material to illustrate the analytical points. Pseudonyms have been used for the participants.

Unstable consumer arrangements

The analysis shows that while the digital food platforms of Meal box providers are socio-materially designed to configure stable ethical meal box food consumers, these efforts often fall short as the consumers in our study rarely stick to a single meal box provider or diet. In what follows, I will first analyse the meal box providers and their efforts to configure meal box consumers. This is followed by an analysis and explanation as to why these configurations failed produce the intended consumer arrangements.

Meal box platforms and the configuring of consumers

The meal box market originated in Sweden in 2007 (Moskin 2013) and has since grown exponentially in the Scandinavian countries and has spread globally to countries such as Germany, United Kingdom, USA, and Australia (Lindsten 2020). At the time of writing, in Sweden there are approximately 20 companies operating and the interest among consumers to shop from these companies has grown steadily and has experienced somewhat of a surge during the pandemic (HUI 2020).

The meal box market is made possible by the ongoing digitalization of organizations, their marketing, and consumers’ everyday practices. However, the meal box is more than digital platforms just interconnecting providers and consumers; meal box schemes consist of multiple components, of which only a few are digital. The meal boxes sold consist of food items accompanied by a set of recipes delivered to the home or picked up at service point. Typically, the food items are premeasured and at times even pre-sliced. The idea is to provide the household with everything needed to prepare the meal. Of course, some basic food items (such as salt, oil and so on) and kitchen appliances are assumed to be part of the kitchen infrastructure. The services are sold via a subscription and usually ordered weekly or bi-weekly. Although the meal box schemes are more than their digital components, digital platforms are key to this mode of food provisioning. The meal boxes are ordered through digital platforms – accessed through the websites interface or apps – and consumer can choose from a variety of boxes, such as the vegetarian meal box, the family meal box, or the budget meal box.

In the analysis below, I approach these meal box schemes as market devices. A meal box scheme as a market device is made up of a heterogeneous network of artefacts and people. It consists of not only the digital interface (website and/or app) but also the boxes and bags in which the food is delivered, the recipes that come with the food, the food packaging, the food itself, the system of trucks, staff, warehouses, and food preparation areas that all work together to provide the meal box service. Consequently, when I refer to market devices, it is these heterogeneous arrangements I am referring to.

These market devices, I will contend, are designed to configure ethical food consumers of different sorts. They are intended to materially and semiotically arrange consumers to align with their offers. The focus of the analysis is on the six different companies used by the participants in our study: Linas Matkasse, Årstiderna, Coop, ICA, City Gross, and MatHem.

Configuring ethical food consumers

As has already been hinted above, much of the marketing of the meal box schemes configured meal box consumers as ethical consumers; consumers that, in one way or another, explicitly consider the interest of distant or absent others when consuming (Barnett et al. 2005). This involved the ethicalization of food consumption as notions of what was the right or wrong way of consuming food were brought to the fore (C. Fuentes and Sörum 2019).

In some cases, being an ethical food consumer meant avoiding or at least reducing food waste. Meal box providers accounted for the measure they took to make sure that the meals are planned as to reduce food waste to the minimum. Not surprisingly, the eating of ecological food was also presented as a form of ethical food consumption. Coop, for example, offered an ecological meal bag, containing ‘100% ecological food’ that was ‘good for you and the environment.’

At other times, being an ethical food consumer was instead framed as someone who eats Swedish and local foods. Here, phrases like ‘only locally produced produce’ or ‘meat from Swedish farmers’ were key. In other consumer configurations, being an ethical food consumer involved instead eating a vegetarian diet. Meal boxes tailored to vegetarians, flexitarians or vegan consumers were common among the meal box companies in the study.

At times it what was excluded rather than what was included that made a meal box service ‘ethical.’ On their website, Årstiderna wrote a statement declaring:

Carefully selected

What we choose to include in our recipes is just as important as what we choose to exclude. Owing to the fact that all our meal boxes are 100% ecological, you will never find genetically modified vegetables, synthetic sweeteners, taste enhancers or most E-numbers. (www.arstiderna.com, 07-08-2019)

In this example, we see a commitment to ‘natural’ as well as ecological food. By avoiding GMO and other ‘synthetic’ ingredients, Årstiderna makes sure that their products are natural in expectation of an ethical food consumer who will value this. Often, the various ethical registers are combined:

Always in Lina’s

Our meal boxes focus to a large extent on natural ingredients. Cooking from scratch gives you better control over what you eat as well as the opportunity to learn about ingredients, seasonings and sauces. The ingredients in the meal box are measured to fit with what is needed to cook the weekly menu. This way we save the planet’s resources!

Vegetables are an important part of a sound and varied diet and for that reason fresh vegetables make up a large part of our meal boxes. We select vegetables by season, when possible, but also import goods to be able to offer variation during the winter season.

We use the World Wildlife Fund seafood guide and never select species that are red listed by the WWF. Every week you also find ecological products in the food box. (www.linasmatkasse.se, 17-08-2019)

Here, we can see that this ethical food consumer configuration involves multiple ethicalities connected to various concerns. In different ways, the market devices of the meal box providers worked towards the ethicalization of food consumption, bringing to the fore the moral dimensions of food consumption and making these an explicit, value adding quality. However, these processes did not produce a single version of ethical consumption. The meal box platforms were not ethically coherent; rather, they drew on multiple ethicalities and made it possible for various ethical consumers to take form.

Configuring various food consumers

The meal box consumers envisioned by the meal box providers were not only, and sometimes not predominantly, ethical consumers. Common to all the meal box services is that they configure various types of food consumers. That is, rather than enacting a more generic ‘everybody’ consumer configuration (Lieshout 2001) or, conversely, focusing narrowly on one specific consumer configuration, the meal box providers configured multiple food consumers simultaneously.

Admittedly, all the meal box providers configured a general meal box consumer: a consumer able and willing to engage in this new mode of food provisioning. Consumers were taught how to order, what the benefits of meal boxes were, and how the deliveries work. There were also detailed instructions on how to cook the meals consumers received. However, beyond this general configuration, the meal box providers also configured consumers in more specific ways. Meal box consumers were configured as family consumers and offered specific meal boxes with easy to cook meals that would suit children’s palates. Conversely, they could also be configured as consumers interested in more adventurous foods in need of ‘inspirational meal boxes’:

Lina’s Inspirational Meal box suits you who spend a lot of time in the kitchen and have a great interest in cooking. In an ordinary week, you will be able to be creative and learn anything from frying tempura, pickling vegetables, to making dumplings. With this, your regular menu will get an exciting dose of influence form different food cultures – so get ready for wonderful taste explosions and a lot of food inspiration! (linasmatkasse.se, 07-08-2019)

In this example, we see that the intended user of the service is someone with a great interest in cooking as well as someone who wants to be inspired by different ‘food cultures’ and desires ‘taste explosions.’ Finally, meal box consumers were also configured as ‘dieting’ consumers, and offered meal boxes in line with WeightWatchers programmes (called health boxes), or, for those with food allergies or intolerances, lactose and/or gluten free meal boxes. Over time, the range of possible diets and thus possible Meal box consumer configurations expanded from a few alternatives to multiple possible meal boxes each with its own intended type of consumer in mind. The meal box platforms and their marketing work envisioned multiple versions of meal box consumers.

Configuring convenient food consumers

Meal box consumers were also configured as convenient consumers; consumers focused on making their everyday life less complicated, saving time, and reducing their household work. These meal box schemes promised to enable and sell a new form of food provisioning. The service offered by them involved changing the way consumers acquired, appropriated, appreciated, and disposed of food. At the centre of this offer was the promise of convenience (C. Fuentes and Samsioe 2021).

This offer has several components. First and most important is a promise of work transfer, where the idea is that by subscribing to the meal box service, consumers will save time and effort (see also, Hertz and Halkier 2017). The meal box scheme not only takes over the task of planning a varied, tasty, nutritional, and ethical weekly menu but also takes over the task of food shopping. It makes cooking easier by providing both detailed step-by-step instructions and, in some cases, by delivering pre-sliced food items. Meal boxes marketed as easy to cook were common as were ‘15 min’ meal boxes guaranteeing that the cooking would not be laborious. However, while offering convenience by taking over some of the tasks involved in planning, shopping, and cooking food, the meal box service also anticipated that using their services would require work from consumers. Because of this, their offer of convenience came with two additions.

Meal box services also promised the convenient use of the service. The meal box providers expected that the introduction of their service would come with coordination challenges and require consumer work. Consequently, work was put into making ordering easy and automated. Typically, the websites and apps were designed to make subscribing as effortless as possible. The customer entered some basic information such as name, address, and means of payment and then chose a meal box subscription. From that point on the order is automated and no more work is required from consumers unless they want to make changes. ‘Sign up, choose, and enjoy’ is, for example, the slogan of Linas Matkasse.

Finally, the meal box providers also anticipated that using their service would require considerable coordination to fit into consumers’ often busy everyday lives. Because of this, the meal box services offered were easy to order and had a great deal of flexibility built into them. Consumers could (and our material suggest that they often did) use the website or app to change their order, choose different ingredients or even different meals. In some cases, they could also use the digital interface to adapt the time or mode of delivery and even suspend the subscription. This flexibility is thoroughly built into digital interface, which consumers can find when first reading about meal boxes while signing up to a subscription and on the Q&A pages. Therefore, the meal box consumer was envisioned as a consumer in need of convenience.

Configuring economical food consumers

Meal box consumers were also configured as economical consumers. Through special offers and discounts, the meal box providers worked forcefully to push the economical consumer configuration, much of which was aimed at attracting new customers. In addition to more targeted and temporary ‘new customer’ campaigns, most meal box platforms offered, as a default, a new customer deal on their websites or through their apps. Any visitor on their site or app was offered a limited time offer when signing up to a subscription. Typical offers included a discount on the monthly fee or a ‘two weeks for free’ offer. Also common were promotional gifts, such as ‘order now and get a free thermos’ (Linas Matkasse, or gift cards). While all six meal box providers in the study engaging in this mode of consumer configuration, some meal box providers, such as City Gross, also marketed themselves as low-cost alternatives. On their digital interfaces, low prices and discounts were particularly dominant. Meal box consumers were, by all meal box schemes, configured as economical consumers: consumers focused on being cost-effective and/or interested in a bargain.

Configuring loyal food consumers

Finally, the meal box providers also worked to configure loyal consumers. The meal box services are subscription services, and the platforms were designed to, in various ways, configure consumers to be loyal to continue the service.

The log-in procedure of the platforms is an illustrative example. On the website or apps, consumers had a login and a profile which they could customize. Here, they would enter preferences and payment and delivery information. As previously mentioned, you could typically use the digital interface to add changes to the order or preferences regarding delivery. While it was designed to be effortless, it still came with a lock-in mechanism as switching to another service would force consumers to learn to use a new digital interface and re-enter all the required information. Furthermore, like all subscription services, the underlying logic is that of an automated subscription. While all the meal box services allowed consumers to suspend or terminate their subscription, the default script of the platforms was that of automated reorder. That is, if consumers did not take any action, the subscription continued.

The loyal meal box consumer configuration is clearly in conflict with the convenient meal box configuration stressing flexibility. While never completely resolved, one can on the meal boxes websites and apps observe a balancing act in which efforts to offer flexibility are often counterbalanced with barriers of sorts to keep customers subscribing. This balancing act becomes clear when reading through the ‘product information’ page of meal box offers:

When you order a meal box, you do not sign up for additional purchases. The try-out-price however only applies to subscriptions. You can end your subscription whenever you see fit. This meal box must be booked 7 days before delivery. If you wish to cancel a meal box, this must be done 8 days before delivery. PLEASE OBSERVE, if you order a food box for a Sunday, you will receive next week’s food box (of the following week). (Mathem.se, 07-08-2019)

Here, Mathem stresses the flexibility of its offer while also making clear that the special offer (and discounted price) can only be obtained by committing to a subscription. They go on to stress that a subscription can be terminated at any time while at the same time making clear that any changes to the order (including cancellation) must be done 7–8 days in advance. Here, the meal box consumer is configured as both flexible and loyal.

Summary: multiple and conflicting consumer configurations of varying specificity

As shown above, considerable marketing work went into configuring meal box consumers. While we did not here have access to the organizational practices and negotiations that resulted in these consumer configurations, we learned about the multiple and sometimes conflicting consumer configurations enacted by these meal box providers by ‘reading’ the market devices they mobilize in their marketing work. The digital platforms on which consumers order, the social media accounts, the boxes, and bags in which the food is delivered, the recipes that come with the food, the food packaging, the food itself, the logistic and product system of trucks, staff, warehouses and food preparation areas all worked, directly or indirectly, to configure meal box consumers.

The analysis shows that multiple consumer configurations were at play here. Meal box consumers were typically configured as ethical food consumers, albeit ethical in different ways, depending on offer and provider. Furthermore, meal box consumers were configured as different in their focus and characteristics. They could be family consumers, interested in meals for adults and children alike; dieting consumers, interested in following a low-calorie or high meat diet; or single household consumers, interested in tasty and easy to make meals. Moreover, meal box consumers were configured as convenient, interested in saving time and outsourcing part of the work of food shopping and cooking, but also economical and cost-conscious. And, finally, consumers were configured as loyal, as different measures were put in place to produce consumers committed to a long-term relationship with their meal box provider.

The various meal box consumer configurations enacted also commonly conflicted. Being a loyal consumer and an economic consumer do not necessarily align. Furthermore, seeing the purchase of local food (such as meat) as ethical does necessarily align with the configuration of the ethical vegan consumer. These conflicts and discrepancies are left largely unresolved by meal box providers. Consequently, a smorgasbord of possible consumers rather than a coherent meal box consumer is offered.

Finally, the multiple and often conflicting meal box configurations enacted also varied in terms of specificity. While certain consumer configurations, such as vegan consumers or weight watcher consumers, were very specific, efforts to configure consumers as convenient or economic were less specific.

In what follows, it will be shown how these configuration efforts contributed to the instability of the meal box consumer arrangements.

Dynamic households and food consumption

What do the consumer configurations enacted by the meal box providers and their market devices accomplish? The study shows that the consumer configurations materialized through the meal box market devices did not always accomplish what they were designed to do: generate stable ethical meal box food consumers of different variations. That is to say, the efforts of the meal box market devices to shape the identities and actions of consumers did not fully succeed. While meal box consumers were generated, these consumer arrangements were highly unstable. This instability, I will argue, was the result of both the built-in dynamics of the meal box market and the dynamic nature of household food consumption. And it is this instability, I will further contend, that makes the issue of ethical food consumption problematic. Below I present and discuss the main reasons why the meal box consumer configurations did not generate stable consumer arrangements.

Changing diets, changing ethics

In our studies, the households continuously changed the diets they followed. A focus on local foods could be followed by a vegetarian period only to later go back to eating (Swedish) meat. It was not uncommon to develop an interest in new diets for health reasons (e.g. anti-inflammatory diet) or weight concerns (e.g. WeightWatchers or a Low-Calorie High-Fat diet). When choosing a diet, what mattered to consumers was not something constant but rather continuously shifting, shaped by public discourse, family and friends, family dynamics and more.

This fluid approach to food was not seen as problematic by the consumers themselves. Instead, the consumers in our study tended to approach diets as consumption projects (Watson and Shove 2008). They would become invested, read up on, and develop new food practices to fit the latest ‘diet project.’ With the project model, however, also came the idea that there was an end to the project. Built into the notion of the project is a finite time plan. Consumers commonly changed meal box providers if these failed to offer a meal box that supported the new diet.

Lisa:  Yes, that is actually one of the reasons why we had the City Gross meal box, because it only contains Swedish meat. I never buy anything else when I buy meat.

Emma: Yes and was that the reason you changed to Lina’s?

Lisa:  Yes, it was. We grew a bit tired of the other one. But there we also had the ordinary or the family box … my father and his wife have ordered Lina’s and I think the tried the vegetarian too because they wanted to get more inspiration to cook vegetarian food.

Emma: Why is it a good thing that it is vegetarian?

Lisa:  Well, I guess it’s mostly some form of sustainability thinking in my own life. It would be good if we ate more vegetarian and reduce our impact on the planet.

As we can see in the interview excerpt above, consumers could switch meal box providers to accommodate new ethical concerns – in this case, vegetarian over Swedish meat – but also switch something simple to experience something new. Here, influenced by an example set by a family member – Lisa’s father – and also because of boredom, Emma and Lisa decided to try something new. This suggests that the dynamics of household food consumption are driven by multiple, interconnecting aspects: concern for the environment, the desire for novelty, and other issues intersect and shape households’ food consumption.

Changing household routines

While food consumption is highly routinized, food routines change, as both the discussion above on diets and previous research has shown (Samsioe and Fuentes 2021). In the households in our study, food consumption routines changed for several reasons.

For example, the meal box subscription was typically ‘suspended’ during the summer period. In Sweden, as in other countries, summer is the time for barbecues. This seasonal and social food practice did not align with the meal box offering leading to a break in routine; that is, a break in the recurring performance of this practice. These breaks commonly also led to consumers leaving the meal box subscription altogether or them taking the opportunity to try out a new meal box provider. Similarly, meal box subscriptions were also suspended during vacations and the Christmas holidays. Seasonal changes in food practice drove consumers to switch meal box providers.

Yes, it is. We felt that it was better not to have any [meal box], and after that we have not gotten around to ordering it again, and it was summer and so we didn’t think it was worth starting the subscription since we do so many other things. But now we have said that we will start up again and for the same reasons as before, to get inspiration and get the food delivered home. (Lotta)

Food routines could also change as a result of changes in the schedule of a household. A change in working hours, pick-up hours, or training could lead the household to re-organizing their food practices. This in turn could make the ordering of a specific meal box – with specific delivery times and conditions – impossible, leading the household to look for alternatives.

Finally, consumers also found the meal box schemes too routinizing. It was not uncommon to suspend the subscription every other week or to suspend it for a period for no other reason than to go back to more flexible cooking. This however often involved following their previously established routines in which the households alternated between a limited roster of familiar dishes.

Chasing offers

Finally, the households in our study also commonly changed meal box providers prompted by discounts, less expensive meal box services, and special offers of different kinds. Many of the households we spoke to had both tried meal boxes for a period of time and switched meal box providers as a result of an incentive: free trial weeks or general discounts were often mentioned by consumers. Meal box consumers were indeed cost-conscious, and many also found the meal box services expensive, which could also involve households trying to adjust to changing economic restrictions:

Yes, you get a welcome offering and so, yes, we were not 100% satisfied with City Gross. It took too much time since it was so cumbersome. And then the “Bamse” box was even more expensive, and we did not have a lot of money since we were both on parental leave. So, well then, we got a special offer from ICA and we were rather satisfied with that and we tried the easy box one week and then we were even more [satisfied]. (Hanna)

In this example, we see that because both parents are on parental leave, their economic situation is more restricted, leading them to search for a more affordable alternative. In all these examples, meal box consumers also act as economical consumers and, as a result of this, switch between providers to gain economic advantages. Therefore, the economical meal box consumer is not a loyal meal box consumer.

Restless households and restless markets

As the preceding analysis has shown, the dynamics of household food consumption make it difficult for the meal box consumer configurations to generate a stable consumer arrangement. Households changed meal box providers and offers to accommodate changes in diets, ethics, routines, or to get more economical offers. While the reasons for these changes are multiple and nuanced, I will discuss four interrelated destabilizing mechanisms. These mechanisms were not solely the result of market or household dynamics but a combination of the two. These destabilizing mechanisms developed and gained force in the interplay between the household and the work of the meal box providers’ market devices.

To begin with, the continuously changing discourses around food and ethics/sustainability seemed to contribute to the instable character of the Meal box market. As illustrated above, consumers continuously changed what they considered ethical or what ethical issues they prioritized. In part, this seemed to be connected to the changing and multifaceted food ethics/sustainability discourse. Influenced by something they had read, heard, or were told by friends and family, our participants would take on a new food project, guided by a new set of food ethics. While these changes were connected to broad societal discourses, it was also clear that they were further amplified by the work of the Meal box providers’ market devices. By continuously offering new variations of the meal box service, thereby marketing various diets, they encouraged consumers to switch behaviour. As shown above, the meal box platforms worked to promote and enable multiple diets, as well as a general interest in food and cooking. Consumers were configured as ‘ethical,’ ‘healthy,’ and ‘dieting’ in various ways. Multiple variations of the meal boxes are developed to try and cater to the continuously changing diet preferences of food consumers. In the process, the meal box providers also end up reinforcing the continuous changing of diets and ethicalities they seek to cater to.

Additionally, also key to understanding the changing food consumption practices of the households that participated in our study was the importance of food novelty. Almost all the households that participated in our study, regardless of family composition type of meal box used or other household characteristics, spoke of the importance of novelty. The continuous search for new food experiences was a central component of all households’ food consumption. New food experiences, for themselves or their family, seemed almost a moral imperative. This was also reinforced by the Meal box providers and their market devices which would continuously develop new offers to accommodate multiple consumers and the changing taste of its customers. As previously mentioned, this lead for the most part to an expansion in the number of variations offered over time. In catering to novelty, this was also reinforced as a mechanism, leading to more fluid consumer arrangements, and switching behaviour among meal box consumers.

The complexity involved in food consumption also added to the dynamics of the meal box market. In previous research, household food consumption has been shown to be shaped by and held in place by social conventions and social relations (family relations, social identities), materiality (domestic technologies and infrastructures), and temporal patterns (Dyen et al. 2018; Evans 2012). This complexity has often been offered as a partial explanation of its highly routinized character. Conversely, this study suggest that this complexity can also drive change (see also, C. Fuentes and Samsioe 2021). Changes in one area or practice will have ripple effects, leading to the reconfiguration of the entire household food practice nexus. Therefore, the interconnectedness of food practices – both to other food practices and, more generally, to other household practices – risks amplifying any changes made. The meal box providers attempt to cater to household complexity by offering flexibility in their service and through their digital interfaces. As shown above, the meal box services allowed consumers to use the app or website to suspend the subscription and to modify their order while also framing this flexibility as one of the selling points. The consumer configurations enacted by the meal box providers are loose configurations in the sense that they are less prescriptive and more open to user configuration. This openness is what enables flexibility. However, the flexibility offered and marketed commonly ended up contributing to the instability of the provider-household relationship by facilitating consumers to suspend their subscription in definitively while they tried out other options.

Finally, it was clear that the meal box consumers in our study also were economically minded: they expressed concern for what they perceived as expensive meal box services and were on the lookout for offers. The economic, cost-minded, self-interested consumer is a much broader societal construct, but it was here also reinforced by the meal box providers’ marketing. While this was done to attract and capture meal box consumers to get them to connect to and establish a relationship with the meal box provider, it often had the oppositive effect. As mentioned in the analysis of the marketing material, discount offers were common to all meal box providers in the study. They would also incentivise consumers with offers of ‘gifts’ or try out periods. In addition, when their customers suspended or terminated their subscription, the meal box providers would work to get them back by offering them discounts or other enticing offers. Thus, while meal box providers were indeed successful in configuring consumers to be economic actors, this rather unspecific configuration made it easier for competing meal box providers to attract consumers, generating instability in the meal box market.

Consequently, in our study, the restless household and restless market are interlocked in a mutually destabilizing relationship. The unravelling of consumer configurations is the result of forces both within and beyond the meal box market. The dynamic relationships that are the outcome of these processes make the project of producing stable ethical meal box food consumer arrangements impossible, if not at least very difficult.

Conclusions

While new, digitally enabled modes of food provisioning come with great promises for ethical and sustainable modes of food consumption, they often fail in their efforts to produce stable ethical food consumer arrangements. Efforts to recruit subscribers/users and establish stable consumer-platforms relationships often fall short. Although the development of digital food platforms offers new opportunities to reach consumers, encourage them to become ethical food consumers, and stabilize consumer-platform relationships through subscription and the automation of choice, this potential is seldom fulfilled. The question this paper has explored is: why is this so?

Starting from the position that all consumers, ethical or not, must be made, I argued that the success or failure of new digital food platforms is not a question of finding and targeting the ‘right’ group of consumers but is instead dependent on producing the appropriate set of consumers. This means that to understand the question of why digital food platforms seem to be failing in their efforts to produce stable and loyal ethical food consumers, one needs to understand the process of consumer making.

To accomplish this, I suggested drawing on the STS and market studies literature, and specifically the concept of user configuration (Woolgar 1998). Guided by this literature and drawing on an ethnographic study of 15 households that signed up to meal box schemes, I analysed the meal box providers and their market devices failed efforts to socio-materially configure stable meal box consumers. The analysis therefore discussed both the efforts to configure ethical food consumers and the resulting consumer arrangements.

The analysis shows that the failure to generate stable ethical food consumer arrangements was the result of both a restless market and restless households. The multiple, often conflicting, and, in some cases, unspecific consumer configurations connected to and intensified the households’ tendencies to change (sustainability) diets, alter routines, and chase offers, thereby destabilizing the meal box consumer arrangements. In this context, constructing stable and loyal ethical food consumer arrangements was difficult: ethical consumer configurations unravelled in the dynamic meal box market.

This instable situation is detrimental to the enabling of ethical food consumption. Because of the instable character of the ethical food consumer arrangements, consumers are only ethical meal box consumers some of the time. In addition, the existence and promotion of multiple and continuously changing food ethicalities complicates matters for consumers. It forces consumers to choose or balance between multiple concerns and various ‘others’ (for a similar argument, see M. Fuentes and Fuentes 2015). Furthermore, because the meal box market offers an increasing and continuously changing range of ‘ethicalities,’ resources are likely to be divided between these various sustainability initiatives. This fragmentation and instability are bound to have detrimental effects both for consumers and ‘upstream’ for all the actors involved in producing and guaranteeing a certain type of ethical food product (e.g. Swedish meat, ecological products).

To conclude, this paper contributes to previous research in at least three ways. First, by contributing to studies of digital failures in a thus far under-researched area (C. Fuentes 2019; Sörum 2020), the paper offers a specific approach to and explanation of why digital food platforms fail to deliver on their promise to produce ethical food consumers. The present paper showed that failures to produce consumers could be understood by examining both market the configurations of consumers and how these configurations managed to actually shape consumer arrangements. Taking this approach, it was shown that, in this case, the dynamics of households’ food consumption were not only enabled but also intensified by the work performed by the market devices of the meal box providers, leading to instability and the unravelling of the envisioned consumer arrangements.

Secondly, this paper contributes to the field of ethical consumption by advancing the small but growing area that examines, often from a socio-material perspective, the making of ethical consumers and enabling of ethical consumption (e.g. Chatzidakis, Maclaran, and Bradshaw 2012; C. Fuentes and Sörum 2019; Giesler and Veresiu 2014; Hoelscher and Chatzidakis 2021; Stigzelius 2017, 2018). In our study, it becomes clear that the construction of ethical consumers is complex and involved much more than infusing consumers with ‘ethicalities.’ Moreover, the analysis shows the fragility of these constructions. While market devices may be performative, it is evident that the dynamics of household consumption make it difficult to perform stable ethical consumers. The construction of stable ethical consumers can therefore not be expected in dynamic contexts.

Finally, and more broadly, this paper can be seen as a contribution to the field of studies that explores the intersection(s) of market making and consumers. As argued in the past, while market studies and economic sociology tend to ignore consumption, the sociology of consumption has in turn paid little attention to the role of markets and marketing in shaping consumption (C. Fuentes and Samsioe 2021; Hagberg 2016). By examining how the market devices of meal box providers configure consumers and the consumer arrangements that unfold as a result, we have offered a nuanced analysis of the ways market and consumption intersect and interact. In fact, the failure to produce stable ethical consumer arrangements could in this analysis only be understood by examining both the dynamics of household consumption and the marketing work performed by meal box platforms. The use of the concept of configuration offers both a way to understand the socio-material marketing work performed to construct consumers and, in this case, is also key to understanding the failed efforts to construct stable ethical food consumer arrangements. Considering consumer configurations and their subsequent shaping of consumption offers a way to bridge the divide between studies of marketing and markets and consumption studies.

Acknowledgements

A special thanks to Dr. Emma Samsioe for her participation in the study. I would also like to thank the special issue editors and the anonymous reviewers for their constructive comments. Finally, I am also grateful to Mia Fuentes and Reka Tölg who read and commented on a previous version of this text.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Financial support was received for the study and the preparation of the paper from the FORMAS financed project Alternative food Markets: Promoting new modes of food provisioning and consumption. Grant number: 2017–01604 and the FORMAS financed SUSFOOD2 ERA-NET project. Sustainable Food Platforms: Enabling food practices through socio-technical innovation (PLATEFORMS). Grant number: 2017–02096; Svenska Forskningsrådet Formas.

Notes on contributors

Christian Fuentes

Christian Fuentes is a Professor of Marketing and Consumption at the Department of Service Studies, Lund University. He does research on ethical consumption, alternative food markets, digital platforms and sustainable consumption, and the digitalization of consumption.

Notes

1 In the present paper, the terms consumer, user, and household are used. The term user is used when discussing the user studies literature. I use the term consumer when referring to a set of specific user configurations made in the commercial meal box setting or more broadly to the enactment/configuring of consumers by market devices. Household is instead used to refer to the households studied and their everyday (food) practices. The term household also suggests that when examining actual practices, the analytical category of interest is seldom an individual user/consumer but a set of interconnected household practices.

2 Dr Emma Samsioe conducted the interviews with the households as part of the PLATEFORM project.