Being a Deliveroo Rider: Practices of Platform Labor in Nijmegen and Berlin

Introduction

One Friday night in Berlin, I am out making deliveries when a double order takes me to a gated condominium on the edge of Friedrichshain. The weather has taken an unexpectedly sour turn and rain soaks through my jacket and collects on my glasses, obscuring my vision. Biking blind, wet, and frustrated, I drift off the curb and pop a tire. Without a spare tube, pump, or even a proper rain jacket, the work night comes to a premature end with a forty-minute walk home. The fees from the double order don’t even cover the cost of repairs. (Fieldnotes, 2019)

The last two decades have witnessed the emergence of the so-called platform economy (Srnicek 2017b). Economic conditions, including the 2008 financial crisis, and the widespread adoption of digital communication technologies set the stage for internet-based marketplaces that mediate connections between customers and services rendered by independent agents (De Stefano 2015; Schiek and Gideon 2018). According to Schmidt (2017, 3), “literally thousands” of different companies have sprung up with a commensurate degree of variation. Some, like Airbnb, allow users to monetize their existing housing, while others, such as Uber, connect workers to a task to be done at a specific location and time. These companies can be described as “lean platforms” for their strategy of outsourcing as many operational aspects as possible. This includes labor, as most workers are classified as “independent contractors” (Srnicek 2017b).

This paper focuses on a specific type of lean platform that has proliferated since the early 2010s: on-demand delivery companies. Services such as Caviar in North America, Deliveroo in Europe, and Etobee in Asia rely on the widespread availability of 4G smartphones and global navigation satellite system (GNSS) location-tracking technology, to match fleets of couriers with customers requesting local deliveries (Cant 2020). Some estimates suggest that nearly 10% of European workers participate in the platform economy (Huws et al. 2017), with on-demand delivery services accounting for a large portion of this activity in many cities (Dablanc et al. 2017). While there has been commendable research on the rise of the platform economy (Srnicek 2017b), including precarious labor arrangements, the legal and regulatory issues they raise (Goldkind and McNutt 2019; Huws et al. 2017; Schiek and Gideon 2018; Van Doorn 2017; Zwick 2018), and opaque algorithmic controls used to monitor the workers (Ivanova et al. 2018), less attention is given to the prosaic labor of actually being a delivery worker (Shapiro 2018a, 2018b). This includes the diverse, contingent, and often messy ways in which on-demand labor comes together on the ground. Our guiding research question therefore is: How are practices of on-demand delivery workers shaped by organizational, spatial and social factors in European cities?

Such a research question benefits from an (auto)ethnographic approach. We used a practice-oriented perspective (Jones and Murphy 2010; Reckwitz 2002) to explore the set of practices delivery workers carry out while performing platform labor in two European cities: Nijmegen (the Netherlands) and Berlin (Germany). We investigated these practices not only by interviewing thirteen platform workers in both cities, but also by conducting an analytic autoethnography (Anderson 2006), during which the first author worked as a Deliveroo Rider for nine months. Hence, he gained direct experiences regarding the meanings, emotional states, and actions this particular job entails, as illustrated in our opening vignette. This personal experience is valuable, as this study sought to gain a subtle understanding of on-demand delivery work, some of which may not easily emerge from other forms of data collection such as interviewing.

The research particularly focuses on the experiences of on-demand delivery workers contracted with Deliveroo, one of the more prominent platforms in the sector (Richardson 2019). Since its foundation in London in 2013, Deliveroo has achieved broad reach with operations in more than 200 cities across Europe, the Middle East, and Southeast Asia (Ghosh 2018). It can be considered a “lean platform,” which keeps its valuation and profit margins high by adopting a “hyper-outsourced” model (Srnicek 2017b, 76). This implies that Deliveroo does not directly employ the workers who carry out deliveries, nor does it make substantial investments in training, equipment, and other personnel. This distinction is one of the defining features of many similar organizations in the platform economy (Huws et al. 2017; Zwick 2018).

Further, we first examine theoretical discussions on labor practices in the platform economy, followed by a more detailed account of our autoethnographic research design. Subsequently, our findings and conclusions reveal the motley, contingent, and conditional ways in which on-demand labor comes together on the ground.

The Platform Economy

The term platform has become somewhat ubiquitous, referring to everything from multifaceted behemoths like Alphabet, Android, Facebook, Spotify, and Uber to a plethora of smaller, more specialized business often grouped under the rubric of the gig economy, sharing economy, or on-demand economy. It has been approached as a managerial discourse (Steinberg 2019), organizational form (Gawer 2014), or set of hardware and software infrastructures (Casilli and Posada 2019). Steinberg (2019, 1) hence points out that “almost anything can become a platform, if one merely calls it such.” Consequently, De Reuver et al. (2017) concede that while platforms are transforming nearly every industry, they are nonetheless an exceedingly difficult object of research due to the degree of variation they exhibit and the complex relationships they sustain with markets, institutions, and technologies.

However, despite its slipperiness, there are a few recurring characteristics of platforms that emerge in the literature, which also apply to Deliveroo. Bratton (2015, 42) provides the following definition of a platform: “a standards-based technical-economic system that simultaneously distributes interfaces through their remote coordination and centralizes their integrated control through that same coordination.” Although seemingly broad, this definition nonetheless captures a distinctive quality shared by all platforms: an overall structure that balances centralized control with a peripheral autonomy. This also holds true for Deliveroo, which remotely coordinates a distributed network of Riders through their centralized software. This is in contrast with previous labor arrangements, where delivery workers contracted as direct employees of a single business (Kinkade and Katovich 1997; Thompson 2015).

Further refining this definition, Gawer (2014) identifies two ways of thinking about platforms: either as technological architectures or as multi-sided markets. The former understands platforms as modular systems that allow different components to interact; the latter as interfaces for two or more parties to conduct a transaction. This market framework stresses the “network effect” aspect of platforms, as the more users they enroll, the more valuable they become to other users. This also applies to Deliveroo: the more restaurants that sign up with Deliveroo, the more useful it becomes to customers, and vice-versa.

There is a general consensus that this centralized control is one of the key ways platforms generate value for themselves and their investors (Andersson Schwarz 2017). Whether platforms are viewed as architecture or a marketplace, they often position themselves as neutral, simply empty vessels for hosting interactions that would otherwise be taking place (Gillespie 2017). However, the opposite is true, as platforms do not necessarily meet an existing need, but instead actively induce the exchanges they mediate, set the terms of the transactions, and monitor the activity to extract further value (Casilli and Posada 2019). This has been described as “platform control,” in which a platform maintains “exclusive control over the surface on which the exchange takes place” (Andersson Schwarz 2017, 381). In practice this means that platforms like Deliveroo can set the terms for their use; for instance, setting the terms for restaurants using the service or determining the fee Riders are paid for their services. By maintaining control of these transaction costs, Deliveroo is able to shape the marketplace and generate revenue. This model incentivizes a monopoly tendency, where platforms must aggressively compete to control a market in order to set its terms and reap the benefits of network effects (Casilli and Posada 2019).

Srnicek (2017b, 76) distinguishes five types of platforms: advertising platforms such as Google and Facebook; cloud platforms such as Amazon Web Services; industrial platforms like General Electric and Siemens; product platforms such as Spotify and Zipcar; and finally “lean platforms.” Lean platforms are distinctive for adhering to a “hyper-outsourced model,” whereby the company owns its software and data analytics, but attempts to outsource all other aspects associated with the service it provides. This final category best describes Deliveroo, which owns its software but relies on the assets and labor of independent contractors to carry out its actual delivery services (Healy, Nicholson, and Pekarek 2017).

These companies aim to stay as lean as possible, continually cutting labor costs and other responsibilities, while workers scramble to stay afloat (Scholz 2016). This reputation results in an established narrative that the “[platform] economy is a form of oppression in disguise” (Goldkind and McNutt 2019, 1) or “neoliberalism on steroids” (Martin 2016, 149). Platforms are said to exploit structural inequality by relying on both massive accumulations of venture capital (Srnicek 2017a) and large pools of underemployed workers to operate (Healy et al. 2017). Moreover, as De Stefano (2016, 8) discusses, the human cost of this pressure is often overlooked, as on-demand workers are often treated as merely instrumental extensions of their platforms, an undifferentiated mass that “could be expected to run as flawlessly and smoothly as a software or technological tool.” These inequalities result in what Malin and Chandler (2016) term “splintering precarity,” where the risks and rewards of the platform economy are unevenly distributed, with the workers employed as contractors shouldering much of the risks with little proportional compensation.

The “Actually Existing” Platform Economy

Despite the considerable pressures on platform workers, they do retain agency in navigating these constraints. Both Rosenblat and Stark (2016) and Lee et al. (2015) point out how Uber and Lyft drivers use online forums to share information to better understand how algorithmic management operates. This allows workers to develop varied strategies to succeed with these constraints, from knowing when to ignore nudges and surge pricing to strategies for obtaining higher customer evaluations. Shapiro (2018a, 2966) shows how decision-making on the job also includes a degree of intuition and affective sense-making. This so-called qualculation suggests “contradictions between the ways that companies model worker decision-making and how workers themselves weigh non-monetary information in their quotidian choices,” which can undermine even the most sophisticated algorithmic control. Sun (2019) indicates that on-demand delivery workers in China have developed strategies with the explicit goal of “gaming” their platforms. These workarounds, dubbed “labor algorithms,” included tricking a platform’s location tracking or collaborating hand-offs with other workers to generate better statistics.

Workers also organize both inside and outside traditional trade unions (Cant 2020). In the case of Deliveroo and other delivery platforms, examples of such organizing include union-affiliated guilds such as the Collectif des coursieres/Koeriers Kollectief in Belgium and the Union Freie Arbeiterinnen- und Arbeiter (FAU) in Berlin. Thus, while platforms are designed and managed as sophisticated systems of control, there are still “spheres of autonomy” available to those laboring within them (Ivanova et al. 2018).

Yet, this autonomy is not only restricted by scripted algorithms. As Richardson (2019) points out, platform delivery is not a smooth operation of supply meeting demand, but is a complex arrangement of actors who come together under shifting and contingent conditions. We follow her call to “decentr[e] interfaces and algorithms” and “examine other material entities involved in platform arrangements” (Richardson 2019, 3) by shifting the focus from the machinations of the platforms themselves to the practices of their workers, to uncover—to borrow a phrase from Brenner and Theodore (2002)—the “actually existing” on-demand economy. In making this distinction, we emphasize the importance of examining the “contextual embeddedness” of this large-scale economic system. To put it another way: If “platforms are what platforms do,” what are platforms actually doing? That is, rather than describing how platform labor is technically arranged, we examine how it is actually carried out different contexts. We dub this nuanced, grounded account of platform work “actually existing” platform labor.

In order to describe how platform labor “actually exists,” this study uses a practice-oriented framework (Jones and Murphy 2010; Reckwitz 2002). According to Jones and Murphy (2010, 376), this approach rests on the idea “that in order to understand higher-order [. . .] economic and social outcomes [. . .] it is necessary to first closely observe and understand the micro-social activities (i.e., practices) carried out and performed by people living, laboring, and creating in the everyday economy.” Reckwitz (2002, 249) defines practices as routinized behaviors consisting of “forms of bodily activities, forms of mental activities, ‘things’ and their use, a background knowledge in the form of understanding, know-how, states of emotion and motivational knowledge.” Thus, a practice is an action that enrolls a number of heterogeneous elements into the creation of a social phenomenon. It is “a ‘type’ of behaving and understanding that appears at different locales and at different points of time and is carried out by different body/minds” (Reckwitz 2002, 250).

In this conception, a single person can “carry” a multitude of practices, and each practice can contain many discrete actions. Jones and Murphy (2010) distinguish four non-exhaustive categories of practices: (a) perceptions (conscious and unconscious intentionalities, desires, discourses, and symbols); (b) performances (actions, social interactions, and activity entwined with non-human materials and technology); (c) patterns (roles, norms, rules, and conventions); and (d) power (repressive or dominating forces, identity positionalities such as class, gender, and race).

By distilling individual accounts of on-demand delivery workers into such sets of practices, and examining the contextually specific factors that inform those practices, as well as the different outcomes they produce, we will be able to understand how platform labor “actually exists” in the world.

Research Design

In order to investigate the practices of platform labor, the first author, worked as a Deliveroo delivery worker (or Rider, in company parlance) for a nine-month period, from August 2018 to February 2019 in Nijmegen and from February to April 2019 in Berlin. This type of direct participation is common practice in labor studies, with researchers taking on work as, for example, industrial tuna fishermen (Orbach 1977), Tampa Bay table dancers (Ronai and Ellis 1989), pizza delivery drivers (Kinkade and Katovich 1997; Thompson 2015), traditional bicycle messengers (Fincham 2006; Kidder 2005), and on-demand delivery workers for Caviar (Shapiro 2018a).

To structure our investigation of the Rider community, we choose an (auto)ethnographic approach in the tradition of Anderson (2006). Thus, while working as a Rider, we endeavored to become completely immersed in the Rider community, becoming a “complete member researcher” of the social world we were investigating. This entailed not only doing delivery work for approximately ten hours each week but also spending time socializing with other Riders, participating in Rider Whatsapp group chats, and attending labor organizing meetings with other platform workers. Moreover, this approach forced us to engage with all the institutional structures that scaffold the platform economy, including the practice of signing up with the company, registering as an independent contractor, and navigating the corresponding tax system. Detailed fieldnotes were written at the end of each delivery shift or attended activity, which were later coded for analysis in Atlas.ti.

A key benefit of autoethnography is that it allows access to subjective aspects of the performed labor through “an intense sensory immersion” (Jonas 2012, 648), particularly regarding the more tactile aspects of the job. Working as an on-demand delivery person—especially one using a bicycle for transportation—is by nature physical work. Larsen (2014, 59) points out that this type of mobility “is an embodied, affective and emotional practice involving specific, societal body techniques,” and that it “relies upon a set of corporeal, cognitive, social and imaginative resources and skills.” Consequently, engaging in and reflecting on one’s bodily interactions with mundane materials such as bicycles, roads, and doorbells provides insight into the actual practices Riders carry out. Yet, importantly, this work also relies on digital technology, specifically, a smartphone and the Rider app, which occupy a central position within the labor process (Ivanova et al. 2018). Research has shown that such apps “affect the body on a variety of habitual, unreflected upon and non-discursive registers” and can have powerful effects on a user’s response and behavior (Ash et al. 2018, 167). As Hughes and Mee (2019) argue in their study on smartphone wayfinding, an autoethnographic approach to these interactions is appropriate and useful as it encourages direct and self-conscious engagement with these digital tools. Thus, actually carrying out the duties of a Rider allows us to simultaneously engage with every aspect of this multi-faceted work, from the minute interplay of fingers, eyes, algorithms, and smartphone screen to the dramatic gestures of swerving through rush hour traffic. Moreover, bringing together these diverse elements as a fluid, unified experience presents a novel combination of digital (Atay 2020) and mobile autoethnography (Vannini and Scott 2020).

This intimate engagement with the labor process provides ample fodder for a descriptively rich account of delivery work. And, of course, a vivid, narrative handling of the subject that “refuses to abstract and explain” surely has merit (Ellis and Bochner 2000, 744). However, as our goal is to develop a more granular understanding of the larger phenomenon of platform labor, a purely evocative approach would narrow our research’s aperture too much (Stahlke Wall 2016). In contrast, Anderson’s formulation of a more analytic autoethnography provides guideposts to effectively connect our subjective to broader theoretical concerns.

Thus, to further broaden our analytical lens beyond an evocative “N of one,” we supplemented our personal data with “dialogue with informants beyond the self” in the form of semi-structured interviews with 13 fellow Riders and countless casual conversations in the field. Through working shifts, Timko encountered many other Riders, which eased the process of recruiting interview participants. He consistently identified himself as a both a researcher and a Rider, distributing handouts describing our project while out on deliveries. Like Fincham (2006), we found that visibly engaging the difficult labor of delivery also provided a degree of “authenticity” and “credibility” with other Riders. Yet, unlike Fincham (2006), Timko relied on money earned through delivery work to cover costs of living. This shared financial motivation is a crucial distinction, as it increased his “insider status” as an actual worker rather than merely a curious academic (Vail 2001).

We interviewed 13 Riders in total, seven working in Nijmegen and six in Berlin. Interviews lasted around one hour and were conducted in public spaces such as cafés and libraries located in each Rider’s delivery area. Interviewees were recruited via three methods: (a) soliciting participants via Rider group chats hosted on Whatsapp, (b) direct recruitment of Riders encountered on the streets, and (c) through snowball sampling where interviewees suggested friends and acquaintances. The resulting pool of participants included Riders reflecting a diverse range of social intersections: two are female, eleven male; five are Dutch citizens, seven are workers on student visas or immigrants from within and outside the EU; and ages ranged from 17 years to 36 years. As Nijmegen has only several dozen Riders, this sample is fairly representative, unlike Berlin that potentially has hundreds of Riders. Semi-structured interviews were conducted, including questions on the subject’s motivations of beginning and continuing delivery work, the routines and habits they follow to accomplish the job, the feelings and perceptions they experience, and moments of rupture which stand out to participants. The interviews were audiotaped, transcribed, and coded for analysis in Atlas.ti. To respect the respondents’ anonymity, we use aliases.

The decision to investigate two sites where Deliveroo operates stems from the research goal of illuminating how the specifics of platform work vary based on contingent factors such as location. Nijmegen was selected out of convenience, as this is the city where Timko, the first author, was living and working at the time of research. Berlin emerged as a case through our collaboration with a European Research Council-funded research program investigating amongst others the platform economy in Berlin. Fortuitously, a number of factors validate the choice of these two locations as appropriate research sites for this comparative work. First, sites in two different countries allow for a comparison of how differences in organizational structures and legal frameworks affect work on the ground (Mair and Reischauer 2017). Second, each city has a different temporal relationship with Deliveroo: Berlin was an early market with operations beginning in January 2015 (and ending after our fieldwork was finished in August 2019), while the company only arrived in Nijmegen in the summer of 2018. Additionally, there are a host of morphological, infrastructural, and demographic differences between the two cities: Berlin is a national capital of 4 million and one of the most diverse cities in Europe, Nijmegen is a provincial city of 170,000, 96% of which are Dutch. Berlin is a sprawling, multipolar metropolis with uneven bike infrastructure, Nijmegen is relatively centralized and has won multiple awards for being cycle-friendly. This rich array of differences will be referred to and expanded on throughout the analysis.

Practices of the Actually Existing Platform Economy

Our research revealed six commonly observed practices, which can be divided into three categories: (a) onboarding the platform (i.e., platform migration and bandwagoning), (b) making the Roo community (i.e, distancing and socialising), and (c) performing deliveries (i.e., anticipating and active idling). As it is always possible to identify more practices, this list is not comprehensive, nor does each individual Rider carry all these practices (cf. Reckwitz 2002).

Conclusions and Discussion

This study sought to uncover the practices that underlie on-demand platform labor and the organizational, social, and spatial factors that shape them. Combining autoethnography with in-depth interviews with Riders in two cities provided a unique and effective avenue for accessing these practices. Regularly working as an on-demand Rider ourselves allowed for direct interaction with the conditions and constraints of the work. Navigating these people, technologies, infrastructures, and emotions involved in app-based delivery work required Timko to carry platform practices himself. Reflexively analyzing this personal process allows nuanced insight into how and why certain practices emerge. Bringing such individual insights into conversation with the experiences and testimony of other Riders enriches these findings, granting a comprehensive view into the diversity of practices at play.

We distinguished six practices: platform migration, bandwagoning, distancing, socializing, anticipation, and active idling. We also came across other interesting practices performed not prior or during delivery work, but largely after or on top of it, such as the organizing capacity of Riders to oppose Deliveroo’s management, which are discussed in more detail elsewhere (Timko 2019). This illustrates that our findings represent just a selection of the many practices carried by on-demand platform workers. Our list of practices is not exhaustive, nor does every individual carry all of them. Instead, every Rider is a “unique crossing point” (Reckwitz 2002, 256), bringing together an assortment of practices, discretely or in coordinated patterns, at specific times and places. They do this not as unthinking automatons, nor as purely rational, unencumbered agents; they are self-aware and reflexive, but also enabled and constrained by contingent factors that make up the world in which they work. While this form of labor is inextricably linked to digital technology, it does not exist in a “virtual” realm, “a separate, perhaps disembodied, dimension of spatial experience” (Kinsley 2014, 376). Rather, it indeed is a “fluid configuration of restaurants, riders, digital devices, vehicles and so on” (Richardson 2019, 7) that is enacted in thousands of small ways across and within a city.

In this way, on-demand platform work should be considered inherently situated; circumscribed within the context in which it takes place and entangled with a diverse array of elements including objects, infrastructure, institutions, and discourses, as well as the minds and bodies of those who carry the practices themselves. By comparing Nijmegen and Berlin, we encountered real differences in how the platform economy exists. Riders socialize and interact in both cities, but the nature of these encounters is different, with a strong tendency toward cohering in small city Nijmegen and more siloed socializing practices in Berlin. Each city is making different demands on Riders’ anticipation practices. The world-class cycling infrastructure and calm streets of Nijmegen mean Riders have to use less precautions, while Berlin’s patchwork cycling network and hectic traffic can make food delivery a strenuous experience that needs to be carefully managed.

These discrepancies between Nijmegen and Berlin demonstrate that the texture of the actually existing platform economy is very dependent on context. Different cities with their particular infrastructures, social compositions, and economic landscapes foster different styles and frequency of practices. Through routinization and repetition, these practices then compound into a contingent, place-specific version of this phenomenon. Over time, these differences could widen, leading to a profoundly fractured set of experiences for on-demand workers in different locations, which raises the question of how platform labor looks in different, non-European contexts.

Yet, different patterns of practices and outcomes are not only place-specific but also based on who is performing the labor. While platforms both require (Casilli and Posada 2019) and expect workers (De Stefano 2016) to provide a degree of standardized labor, our findings show that there is significant variation in the practices carried by on-demand workers. Just as with spatial context, examining such discontinuities between individuals can shed light on how practices produce different impacts and outcomes (Jones and Murphy 2010, 382). As shown, there are identifiable patterns to who decided to take part in on-demand labor, such as platform migration and bandwagoning, resulting in most of our respondents being migrants and students.

However, Riders are more than just migrants and students; they have multiple, shifting, and sometimes simultaneous identities, with age, race, gender, and abilities intersecting. Riders have different physical abilities, social responsibilities, and financial resources with which to contend when executing their duties. For instance, while younger or fitter Riders may be better suited to endure the job’s physical strains; older, more established Riders might have more financial resources to invest in tools and equipment. While some Riders may find excitement in the sometimes-dangerous work-environment, Riders with family or care duties may have to invest more in anticipating and avoiding collisions. Previous research notes this danger endows delivery work with “a certain type of machismo” that genders the occupation and can, but not always, limit broader participation (Kidder 2012, 65). As these examples illustrate, the actually existing platform economy does not evenly distribute its risks and rewards.

The practice of being a Deliveroo Rider is thus always a conditional process responsive to the surrounding environment and conditions; any given Rider may routinely engage in an evolving constellation of practices depending on the situation. Additionally, each practice can build on another, for example Riders engaging in platform migration to finance travel while also socializing into independent social groups at the arrival location. All these interconnected and routine practices are ways of behaving, understanding, desiring, moving, and using compound and accrete. Through time and repetition, they shape the actually existing platform economy (Reckwitz 2002), which is not unitary or homogeneous but malleable and contingent.

When carried out individually, most of the discussed practices are beneficial for workers in the short term; however, in aggregate they perpetuate the “lean” platform structure and its associated issues (Srnicek 2017b). The true benefits of the bandwagoning practice redound not to the individual delivery worker but to Deliveroo, as it increases the platform’s network effect since it is more likely to carry out any given order (cf. Srnicek 2017a). Moreover, amassing a pool of labor puts platforms at an advantage, as makes it easier to lower payments while still keeping enough workers willing to take orders (Healy et al. 2017; Wentrup, Nakamura, and Ström 2019). Similarly, individual workers benefit from anticipation practices, but Deliveroo reaps the greatest reward. The platform creates conditions where the burdens and risks of delivery work are transferred to those on the street (Zwick 2018), allowing them to keep operating with extremely low expenditures as workers cover the costs (cf. Goldkind and McNutt 2019; Trnka and Trundle 2014).

Platform labor is thus highly contextual and precarious, and subject to constantly shifting conditions to which workers must continually adapt. The fact that Deliveroo ceased operations in Germany in August 2019, shortly after we ended our fieldwork, is very telling. A new Finnish platform, Wolt, entered the market instead. This illustrates how quickly the entire platform ecosystem can shift over the course of just a few months. The outbreak of the COVID-19 coronavirus also demonstrates the job’s fluctuating conditions and dangers. Researchers have identified platform workers as a key at-risk group, as their labor is deemed “essential” yet it continually exposes workers to potential infection (Robinson et al. 2020). This compounds Malin and Chandler’s (2016) concept of “splintering precarity,” where only a privileged set of workers stand to benefit from new labor arrangements, while others are left to shoulder the downsides. The new threat posed by COVID-19 requires platform workers once again cultivate novel practices to maintain the already lopsided cost-benefit trade-offs of the job. It is imperative that future studies examine not only how platform companies institutionally adjust to this development (Katta et al. 2020) but also how workers actually managed to make do.