Organising fragmented labour: the case of migrant workers at Helpling in Berlin

Abstract

Research on collective organisation of migrant platform workers is mostly concerned with ride-hailing and delivery logistics, where explicit forms of collective action have been visible. This paper addresses the issue of the self-organisation of migrant workers on cleaning platforms through a case study of Helpling cleaners in Berlin. In the paper, we ask why the attempts for organising workers have failed to scale up beyond the informal exchange of information and tactics among the workers. Our article argues that the spatial dispersion of work across the city, lack of occupational identity, and the legal framework of work, make worker organising difficult for cleaners in Berlin. Still, these factors do not lead to an absence of collective practices. Helpling workers in our study gather in online groups, can receive help and exchange in a community centre, and have been in touch with political groups. Based on the case study, the article discusses potentials and hurdles for the development of collective counter-power.

Introduction

The topic of labour conflicts on digital platforms has received much attention by researchers in recent years. While discussions about work on digital platforms initially focused on the new regimes of control on platform business models (Shapiro 2018; Veen et al 2020), numerous organising movements as well as studies on this, have raised attention on how such forms of control are challenged (Animento et al 2017; Heiland 2020). In industries such as delivery logistics and ride-hailing, these organising movements have even led to a comeback of union issues in the public sphere (Vandaele 2020; Trappmann et al 2020). Nonetheless, attention to platform work has been mostly limited to the industries mentioned above, where publicly visible conflicts have taken place. In fact, the diversity of platform-mediated forms of work is much higher and extends to forms of online-based platform work (Mechanical Turk, Upwork, and social media content creation) as well as to place-based platform work in domestic cleaning or care (Wallis 2021; Glatt 2022; Ticona and Mateescu 2018). Organising also took place in some of these areas, albeit to a lesser degree and in their own forms (Golušin 2021; Niebler and Kern 2020). In much of the earlier research on platform work, the role of migration as a key factor was also often omitted (Van Doorn and Vijay 2021; Altenried 2021).

This paper addresses the question of the collective resources and organisation of migrant workers on the cleaning platform, Helpling. Based on qualitative research in Berlin, we explore the question of why the attempts of organising on cleaning platforms have failed to scale up and to go beyond the informal exchange of information and tactics among the workers. Our article argues that the spatial dispersion of individualised work across the city, conducted in private households, the absence of shared socialisation because of platform mediation, and its legal framework make organising labour on platforms like Helpling difficult. We argue that this does not lead to an absence of organising and social struggles, but instead to a change of their forms. Workers on Helpling in Berlin organise in informal mutual aid groups, with the help of a counselling centre and through migrant political groups. Central to all forms of organisation we observed is that they emerged from particular migrant communities and not through social bonds formed in the workplace. These circumstances give rise to both potentials and hurdles for the development of collective counter-power, which are also relevant for the analysis of other forms of platform work, especially in the care sector. The development and structure of our argument builds on and relates to earlier work we have published on the topic (Niebler et al 2023a) and presents an engagement of our findings with existing and current research on workers organisation and platform labour. ¹

The article starts with an overview of the phenomenon and the role of cleaning companies in the platform economy and explains the hurdles to organising in platform-mediated labour. After a brief presentation of methods and approach, the article describes the company and business model of Helpling, as well as the specific situation of platform cleaners working in Berlin. We then present three forms of collective (self-)organising of Helpling workers, which we then discuss analytically in a concluding section.

Platforms, cleaning, and collective action

Cleaning platforms are part of the platform economy that has emerged in the past decade, and which has launched new forms of labour intermediation through digital technology, regulatory loopholes, and high concentration of capital investment (Graham and Woodcock 2020; Cooiman 2021). The business model of labour platforms is to provide a pool of (mostly self-employed) workers to clients or customers, with the company both exercising digital control over the workers and retaining often very high commission fees (Kenney and Zysman 2020; Altenried et al 2020). In addition to business models for delivery services, taxi services or short-term rentals, companies for cleaning services in private households have also emerged in this context (Bor 2021; Gruszka et al 2022; Tandon and Rathi 2022). Like other service providers, cleaning platforms offer quick solutions for household and everyday tasks for individuals and households that experience a shortage of time for these tasks in their everyday live. Existing research has debated the increasing difficulties of households in coping with tasks falling into the field of reproductive activities as signs of a ‘crisis of social reproduction’ (Fraser 2016; Aulenbacher et al 2018; Altenried et al 2021). ² Theories describing a crisis of social reproduction argue that the financialised logic of contemporary capitalism has cut off the resources of working populations to perform unpaid care activities, a dynamic that threatens the very functioning of capitalist economies, which are fuelled by unpaid care work (Fraser 2016). Part of this process is an austerity-driven decline of public health infrastructures and lack of child or elderly care alongside higher costs for housing and living expenses (Mohandesi et al 2017, 65). Under such conditions, ‘care gaps’ (Kluzik 2022, 3) arise even for small everyday tasks in households, which for middle- or upper-class populations can be closed using digitally mediated and low-paid wage labour. Cleaning platforms offer low-income workers to private households, for example, to mop floors, clean toilets, or wash windows. The companies control the labour process through ratings and sanctions and manage the payment processes (Altenried and Niebler 2022; Gruszka et al 2022). By using self-employed workers and coordinating them through digital infrastructures, the company avoids fixed costs such as staff provision and management, thereby re-establishing modes of labour extraction that have been prevalent prior to the rise and fall of the standard employment relationship in wide parts of the world (Stanford 2017; Flanagan 2019). The often-high commission margins of the companies, lack of protection, and resulting precarity for the workers have been repeatedly problematised in recent years (Ticona and Mateescu 2018; Bor 2021; Floros and Jørgensen 2023).

In most areas of platform work, discussions about working conditions have raised the question of organising workers, the role of unions, as well as forms of organisational misbehaviour (Vandaele 2018; Heiland 2020; Altenried and Niebler 2022). Compared to workers employed in non-platform companies in the same sectors, workers of platform companies appear to experience high hurdles to organise and act together. Building on existing debates on this issue, we frame these hurdles here as forms of fragmentation of labour (Heiland 2020; Della Porta et al 2022; Niebler et al 2023a). Fragmentation can be understood as ‘an active politics of isolation designed to prevent the collective organisation of workers’ (Della Porta et al 2022, 222). In the case of platform work, we consider three dimensions of fragmentation as crucial: first, legal fragmentation of workers through forms of (pseudo) self-employment or atypical employment; second, technological fragmentation through practices of algorithmic management and data control; and third, spatial fragmentation through the diffusion of labour across space and the absence of concrete sites of socialisation (comp. also Altenried et al 2020). While none of these fragmentations are equally pronounced in every platform business model, in sum they limit conventional forms of workers’ countervailing power such as strikes or collective bargaining as well as the access to existing labour rights (Vandaele 2018; Basualdo et al 2021).

Despite these difficulties, the existence of collective action and organising on platforms is now a well-established fact and has shown that it is possible to challenge the fragmentations mentioned above. For the case of food delivery platforms, Tassinari and Maccarone (2020) have evaluated the potential and obstacles for emerging solidarity among workers. Based on two cases of successful organising in Italy and the UK, they identified several success factors such as the availability of meeting spaces for workers in the public urban sphere to socialise and nurture social relations, the development of workers’ common consciousness during protest actions, and the presence of organisations such as unions and political collectives supporting the riders’ protests (Tassinari and Maccarone 2020). While similar studies exist for the case of ride-hailing (Aslam and Woodcock 2020) or beauty platform work (Dhar and Thuppilikkat 2022), they do not exist for domestic cleaning platforms so far. This might stem from the sheer lack of publicly recorded cases but cannot only be attributed to this. The argument of our paper is that the lack of such cases does not just stem from the smaller size of the platform cleaning sector (comp. Haidinger et al 2022) but also from the fact that domestic cleaning platforms are particularly illustrative examples of the power of platforms in fragmenting their labour force. While instruments of technological fragmentation play a somewhat subordinate role in comparison to other sectors, the potential to atomise workers across space has far-reaching consequences. Platform cleaners at Helpling are not just legally separated as self-employed freelancers but are also not recognisable in the public, do not have work-related sites of socialisation, and (apart from some exceptions) have to deal with their clients individually. These factors, which we will describe in more depth below, make cleaning platforms a particularly difficult terrain for collective advocacy. Still, we argue that these conditions do not necessitate the absence of solidarity infrastructures or collective resources. Instead, they lead to more informal and mostly community-based practices of dissent among cleaning workers. We argue that the scientific debate on platform labour conflicts should also include such forms of organisation in their analysis if it wants to avoid a narrow focus on (mostly male dominated) sectors such as delivery logistics and mobility (comp. also Kampouri 2022, 15).

Methodological approach

Our research is based on an investigation of platform work carried out in seven European cities (Bojadžijev and Mezzadra 2020). The project conducted research on the labour process, social protection, and skill development of platform workers in an explorative manner. As part of the study, the platform company Helpling was analysed according to these issues in Berlin, Bologna, London, and Paris. This paper will focus on the results of the empirical research carried out in Berlin where a total of 13 qualitative, semi-structured interviews were conducted with workers between January and May 2020. Further, a focus group with cleaners and other platform workers was held in October 2020. Interviewees were recruited through classified advertisement websites, personal contacts, and community groups on social media. Throughout the research, we were also invited to two messenger groups for workers on cleaning platforms and other related activities, which we used to get in contact with interviewees. It was also possible to make use of snowballing in several cases. Interviews were conducted in Spanish, English, and German.

Generally, the recruitment of precariously paid cleaning platform workers is resource-intensive and presents a range of new challenges for researchers (Iphofen et al 2022; Ustek-Spilda et al 2022). In the case of Helpling, workers are on one side very visible, as their profiles are more or less publicly accessible through the application. On the other hand, workers are very much invisible in the physical space, particularly in the public space, even if their work is on site and not remote. For researchers willing to investigate labour conditions at Helpling, it is hardly possible to find workers in the city, as it is the case with delivery or ride-hailing workers. While these workers do move through the city, there are no places in which they concentrate, they are physically dispersed and moving alone (Kampouri 2022). Helpling workers do not wear a uniform, nor do they have any physical signs which can signal their belonging to the platform or to any type of domestic work. A much-discussed entry point for researchers concerns using the application itself (by creating a work request) to contact possible respondents. This contains a number of ethical risks concerning the informed consent of interviewees (given that workers might feel obliged to comply with an interview request), transparency of the research process, as well as the anonymity of participants (Ustek-Spilda et al 2022, 85). We therefore did not approach workers through (assumed) requests or any other means via the platform itself.

The collection of our data took place both before and after the outbreak of the COVID-19 pandemic in Germany, which allowed us to observe a variety of vulnerabilities and challenges (sudden loss of income, social protection, and health hazards) faced by workers during this abrupt moment of crisis. To complement the findings, we also conducted interviews with stakeholders and representatives of the cleaning industry in Berlin and Germany as well as with a company representative in Berlin at the Helpling headquarters. The interview material was analysed in accordance with the framework of Qualitative Content Analysis by Kuckartz (2014), which provides a reliable and transparent way of working both deductively and inductively with qualitative interview material (Kuckartz 2014, 181). The analysis followed a six-step process: (1) data preparation and initial read-through, (2) building of main categories based on interview questions and key issues (labour process, social protection, and skills), (3) coding of the material with main categories via QDA software, (4) development of the code frame by forming subcategories inductively from the material, (5) category-based analysis via QDA software, and (6) reporting and documentation in the form of internal reports and working papers. In many of the interviews, the biographical journey from the workers’ country of origin to Europe or Germany was a topic that became an important part of the analysis. During the project, preliminary and final results were discussed at policy workshops with trade unions, workers, and representatives from politics and public administration.

The case study: the platform Helpling in Berlin

Helpling was founded in 2014 in Germany and is today the most dominant firm in the sector of platform cleaning services in Europe. After its foundation, the company was able to raise investment capital with a business model idea oriented at existing start-ups like Homejoy or Handy in the United States (Auchard 2015; Koutsimpogiorgos et al 2023, 9). The company operates in 11 countries, most of them in Europe, and Germany appears to be the biggest market. The company has also tried to set foot in countries like Canada, Brazil, and Australia but was not successful. While investment rounds in recent years were not very successful, the company claimed profitability in 2020 (Deutsche Startups 2020).

On a formal level, Helpling operates as an intermediary between private households and self-employed cleaners. However, the company has developed very different terms and conditions, depending on national legislation (Koutsimpogiorgos et al 2023). In Switzerland, the company employs cleaners through regular labour contracts (Niebler et al 2023a, 254). In Germany, the company also started offering a small number of permanent cleaners through a subcontractor, although the size of the workforce is unclear (Niebler et al 2023a). Self-employed workers can start at Helpling without previous experience and receive at most a short introduction, for which they often must pay themselves. There are several problems for workers: the wages paid, once the fees and transport costs are deduced, are often below the minimum wage set by the law (Altenried and Niebler 2022). Migrant workers especially are often unaware of their tax obligations as self-employed workers. In addition, cleaners are often heavily dependent on customer reviews on the app, as well as subject to threats of financial penalties from the company. In Germany, the company has already been criticised for these working conditions and has taken part in a concerted dialogue with the Federal Ministry of Work and Social Affairs (Bundesministerium für Arbeit und (BMAS) 2022), although no big changes seem to have resulted from this. On a European level, the currently discussed proposals for an EU Directive for platform work have raised discussion whether work at Helpling will be declared as false self-employment in the coming years (Henning 2022; Gruszka et al 2023).

Berlin is Helpling’s headquarter city with around 140 office employees, as well as the company’s largest location (Deutsche Startups 2020). In a 2019 sample, this included about 300–400 workers available daily in the Berlin area (Bor 2021). In the last decade, Berlin has become both an investment location for venture capital and a recruitment pool for many workers from Germany and abroad. Migrant workers without access to the primary labour market are of great importance to the platform economy. With companies such as Zalando, Gorillas, or Delivery Hero, the city has become an experimental field for platform work in recent years. Since its foundation, Helpling has bought out two competitors in Germany (BookaTiger and CallJeffrey). Recently, the Danish start-up HappyHelper has also entered the market. Domestic care platforms such as Careship and the European offshoot of its competitor Care.com are also based in Berlin. Symbolically and economically, the city is therefore highly relevant for Helpling and other companies in the platform economy.

Parallel to corporate developments, Berlin has also become a site of collective mobilisations against tech companies. A campaign led by the grassroots union FAU against the delivery service companies Deliveroo and Lieferando, the successful protests against a Google campus in Berlin Kreuzberg, and a wave of wildcat strikes at the company Gorillas have led to established structures of counter-power by workers in the city (Labournet 2017; Bronowicka and Ivanova 2021; Orth 2022; Scholz 2022). These structures complement the generally high level of social conflict in the city, ranging from strikes at hospitals to organising tenants against large housing companies (Scholz 2022). Given the strategic importance of Berlin for the platform economy and for Helpling in particular, as well as the high level of social conflict in the city (and against technology companies in particular), the question arises as to how this mixed situation might affect the present and future of the company and its labour force.

Helpling workers in Berlin are mostly young and have recently migrated to Germany. In our research, most workers we interviewed had arrived in Berlin from Latin American countries such as Argentina, Chile, and Peru via temporary visas, and in particular with Work & Holiday programmes. Most of our interviewees were highly educated and had a BA or MA academic degrees. On first sight this seems surprising, because it neither resembles the dominant composition of the migrant labour force in Germany (stemming from Eastern Europe, Southern Europe, or MENA countries) nor their educational levels (which are mostly lower). While in our sample this group might be overrepresented, an analysis of profiles on the application as well as existing literature about the situation in Berlin and other cities supports its relevance also on a broader level (comp. Floros and Jørgensen 2023; Orth 2023). Research also shows that the migration regimes regarding young migrants coming to Berlin interact with platform work, as platforms offer an easy and unbureaucratic way to start working right away after moving to the city (Animento 2023; Orth 2023). Furthermore, most cleaners in our sample are male, while the company claimed a roughly 50/50 distribution in an interview with us. Even the latter distribution would be quite different from the domestic cleaning sector in Germany generally, which is widely dominated by female cleaners.

What makes Helpling attractive to migrant workers in Berlin? Flexibility of working hours has been an aspect often advertised by the company and in fact appears to be an aspect that draws some workers to the job initially. At the same time, many respondents also refuted these arguments, stating that flexibility mostly existed for customers and that changes in the relatively fixed, two-weekly time slots due to other obligations or sickness were sanctioned by formal requirements and sometimes fines. A more obvious advantage for our respondents was that the company, at least during the time of our research, did not ask for the kind of paperwork other companies would ask for (comp. also Van Doorn 2023, 166). Many workers indicated that they did not have to provide a housing registration, something that is demanded by most employers in Germany in order to register the worker as proof of their legal status. The issue of registration has proven to be a chokepoint for many newly arrived workers. One respondent told us about his rejections at other jobs for being not able to present a registration:

In three months I moved about 7–8 times. […] I lived in like four hostels and moved [to] many apartments. It was really really tough. So that means not having the Anmeldung. […] even I was working in a really nice coffee shop in Charlottenburg but I had to quit because I had no paperwork. I always was saying: ‘Okay, another month, another month’ and then it was just like: ‘You know what, I don’t know how much longer it’s going to take, so bye bye’, you know. So that’s another good reason because they [Helpling] don’t really check my Anmeldung or anything. [Helpling BE 9]

Technically, a registration appears not necessary for workers to enter employment (EU Equal Treatment Office 2023). Still, the statement emphasises that the issue of registration is of high importance for many newcomers, even in low-paid service jobs, and that ways to circumvent this chokepoint are needed for them in practice. Migrant associations state that the lack of available and affordable housing in the city makes it ‘practically impossible to register their address in public offices’ (Bloque Latinoamericano Berlín 2023), which then turns into ‘a vicious circle of precariousness that consumes migrant life in Berlin. Without ‘Anmeldung’, new citizens remain excluded from the most essential urban goods and services and, above all, from the possibility of access to formal employment’ (Bloque Latinoamericano Berlín 2023). Previous research on the access of migrants to labour and housing showed that dynamics of ‘differential inclusion’ shape the process of registering in Berlin (Animento 2021). The concept of ‘differential inclusion’ helps to clarify the diversification of social trajectories of migrants once they arrive in the new place (Mezzadra and Neilson 2013). In Berlin, migrants of higher social status are better able to ‘buy’ their access to the city, by paying disproportionate rent prices. In contrast, migrants from poor families or with precarious residency status have to engage in highly precarious and non-qualified work in order to cope with their insecure situation and find accommodation (Animento 2021). By offering easy access to labour without paperwork and without the necessity to speak German, Helpling has become a somewhat attractive opportunity (or depending on the perspective, a last resort) for work and income. Advice on this can be found in Facebook groups such as ‘Trabajos para hispanos en Berlin’ or ‘Españoles y latinos en Berlín’, where workers also discuss taxation and self-employment related to the job. The motivation of Helpling to work with individuals who are not eligible to work has also shown in (unsuccessful) company initiatives to grant work possibilities for asylum seekers (comp. Koutsimpogiorgos et al 2023). Effectively, Helpling appears to fill in a void created by various problems in the city’s public services as well as its housing and labour market.

Facing fragmentation: hurdles to collective action

Based on our interviews and analysis, our research could identify three main hurdles to organise collectively: the socio-spatial dimension of the work, the lack of identification with the activity, and third, the absence of legal organising frameworks due to the status of the workers as self-employed (comp. also Niebler et al 2023a). All three will be described in the following and also contrasted to other forms of platform work or service work.

Work at Helpling is characterised by the spatial diffusion of the workers across the urban space, who, unlike delivery service bike riders or cab drivers, do not have access to shared places of socialisation or signs of recognition in public space. As a result, even workers with several years of experience in the company do not meet colleagues at any time. At the same time, the work at Helpling is characterised by specific relationships of recognition and loyalty, especially the personal (and often confidential) relationship with clients in their private spaces. In contrast to the exchange with other workers, the focus here is on the relationship with private clients. Although sometimes workers offer their cleaning services to clients with no intermediation from the company, thus sabotaging the company, this bypassing does not give rise to a collective capacity for action on the part of workers (Altenried and Niebler 2022). Of course, some of these hurdles exist to some extent in service work more generally, for instance, in the form of subcontracting service workers and their frequent change of working site. Still, common locations, meeting points, and a common legal frame usually exist. As Pannini (2023) has shown in a study about a cleaning worker strike at British universities, even subcontracted service workers are able to make use of spatial proximity, the legal umbrella of their subcompany, and the regular locations of their work in corporate or public buildings.

According to our analysis, a second barrier to collective organising is the barely existing identification of the workers with the activity. This phenomenon is common in the field of low-paid service work but differs in Helpling at least from other forms of platform work. Even for workers who have been working through the platform for years, the job at Helpling means a temporary job with which they do not identify. Often the job helps to finance a desired other activity (such as a language course, an education, or a trip). When asked about the possibility of organising, an interviewee who has been working for Helpling for more than 3 years explains that to him this,

…seems like a losing calculation. I [am] most of my time pretending that I’m not in this job, so if I were to start organising, I would have to accept that I’m still in this job. So, it’s also psychologically problematic. [Help_BE 12]

This non-identification may be linked not only to the lack of socialisation opportunities for workers explained above but also to the stigma of cleaning and household activities in general. The predominantly highly qualified interviewees in our study do not associate their self-image with that of a cleaner. Although a distancing can also be observed among workers for delivery services and in the cab sector, at least a (sometimes resistant) appropriation of subjectification narratives can also be observed there – for example, the self-designation as a ‘rider’ in the delivery service sector, which builds on bike-messenger culture or the perception of essential work in the ride-hailing industry (Gregory 2021: 318, Pirone et al 2020). Within our sample, such a positive identification was lacking.

An additional obstacle compared to other forms of platform work lies in the self-employment of workers, which has become an exception in the German platform economy. In the ride-hailing and delivery industry, a mix of collective action by workers and pressure by incumbent industry stakeholders and regulators made it possible that most platform workers are today in fact employees of a company or sub-company (Niebler et al 2023a). Formal employment grants workers’ rights to co-determination through works councils, a tool that at least in the food delivery sector has served as a catalyst for organising movements from below. Such tools, as well as other networking possibilities that come along with the employment status, are not available to (most) Helpling cleaners. This makes legal avenues to collective action or co-determination nearly impossible and means that workers are mainly bound to informal strategies.

Countering fragmentation: collective resources and counter-power

As we have seen, workers on Helpling lack some of the resources for organising that are open to other platform workers (networking in public space, mobilising a shared identity, and forming works councils). Nevertheless, they find ways to collectively navigate their risks and intervene against company strategies. Three forms could be identified from our research in Berlin: informal exchange and mutual help through messenger groups; advice and support through an information centre for precarious migrants; and organising through local political groups. In Berlin, all these practices are intricately linked to membership in Spanish-speaking or Latin American online and offline communities, which also make up a significant proportion of platform workers at Helpling, as explained above. The role of these groups, which therefore plays a major role in terms of collective resources, is briefly discussed in the following. In the second part of this section, the findings will be discussed in relation to current research on migration, platforms, and collective action.

An important collective resource on Helpling in Berlin are the exchange groups of workers via messenger applications like WhatsApp or Telegram. Because work on platforms is rarely linked to physical workplaces or locations, these messenger groups often play a central role in the exchange and collective socialisation of workers. Since Helpling’s market entry, several informal groups have been established in Berlin, where workers help each other with concerns and transfer knowledge, warn each other about bad or risky clients, and express solidarity with each other when experiencing suffering or problems at work. One interviewee describes how the groups work as follows:

There are WhatsApp groups with many people from Uruguay, Chile, Argentina, people from Latin America, we have contact with everyone. […] So that we can gain experience and often in the WhatsApp group you can see in the morning, ‘I have this problem’, ‘Can you help me’, and we all try to help each other. [Help_BE_6]

A key aspect of helping each other is to warn about problematic or harmful clients, who stood out for setting incorrect working hours, giving poor evaluations, or carrying out inappropriate treatment. Through the groups, workers also provide each other with mutual encouragement or consult with each other about ending the intermediation through Helpling by taking the work off the platform in cooperation with the clients. Some workers tried to take clients off the platform strategically to build their own client base. Many aspects of these groups resemble what has been described as online ‘whisper networks’ (Komarraju 2023), as a term for informal communication networks among marginalised groups to warn each other about potential danger and harm, often created to prevent sexual harassment (Komarraju 2023, 94). Based on interviews with beauty gig workers in the city of Hyderabat in India, Komarraju describes how workers use WhatsApp groups to exchange with each other about unreliable or dangerous customers, about filing complaints, and on how to deal with updates on the app that change their working conditions. Komararaju sees whisper networks through Whatsapp groups as essential tools ‘to avoid a domino effect on the physical, mental and financial well-being’ (Komarraju 2023, 95) of workers in her study. Although not created for instrumental purposes, conversations in these groups even triggered collective protests against a company in one case and subsequent lawsuits (Komarraju 2023, 92). In Berlin, Whatsapp groups serve similar purposes but occasionally also delve into more political terrain. The fact that one of the groups we encountered was called ‘Helpling Union’ emphasises this aspect.

A second form of resource for collective organising is a counselling centre in Berlin for Spanish-speaking migrants named Oficina Precaria. The Oficina Precaria is an initiative founded in 2015 that offers advice on labour law, tax matters, and bureaucratic tasks to Spanish speakers in Berlin (Ribés 2017). For this purpose, the initiative, which is part of the Spanish 15M movement, uses premises of a neighbourhood centre in Berlin’s Wedding district. One of the founders explains the concept of the initiative:

Oficina Precaria is not an association […]. We are a gathering of friends […] and we try to help Spanish speakers who do not speak German in solidarity and to inform them how to translate German bureaucracy. Actually, we are not advisors, we try the first level of orientation. […]And in this team we talk to very many people who work at Helpling. [Help_BE_6]

The concept of a centre for precarious workers has spread in the 2010s both in Spain itself and (temporarily) in several European cities such as London, Paris, or Prague (Ribés 2017). The activities in these centres are often carried out by volunteers, who are also recruited among the workers who previously looked for the consultation. The work of such centres is similar to the concept of Workers Centres, which are widespread in North America and reach out to migrant working populations with both counselling services and organising resources, often in consultation with trade unions (Fine 2006, Apostolidis 2019). In Berlin, Oficina Precaria assists workers with three main problems: first, when facing issues with the (algorithmic) system of the mobile application or its interface, as in the case of de-ranking or forced log-off; second, with bureaucratic hurdles related to their self-employment; and third, with a wide range of socio-economic issues connected to their arrival in Berlin (comp. Niebler et al 2023a ). Because Helpling sometimes sanctions its cleaners for misconduct against applicable labour law, contacts to lawyers or legal counselling can be arranged. Oficina Precaria helps with filling out paperwork, registering for self-employment, or preparing a tax return, which many cleaners often only learn about after they start working. Most importantly, the office offers all kind of advice for workers to navigate their arrival in Berlin and Germany, such as how to receive a housing registration (Anmeldung), registering a bank account, and settling fines for fare evasion, church taxes, or gastronomy certificates. It gives advice on finding housing, the conditions of accessing abortion, and maintains a register of Spanish-speaking doctors and psychotherapists in Berlin. In addition to providing advice, Oficina Precaria is also a place where workers can get in touch with colleagues or other migrant workers.

In a study of nine different workers centres across North America, Fine (2006) identifies three core functions of workers centres: first, service delivery such as legal assistance, language courses or rights education; second, advocacy activities such as political work and studies to improve legal conditions or strategic litigation; and third, organising activities, usually by fostering community-based organising and leadership initiatives to enable grassroots struggles (Fine 2006, 420). Often, such centres are professional or semi-professional institutions with a non-profit tax status, full-time staff, a board of directors as well as formal membership (Fine 2006, 443). Although Oficina Precaria is also involved in service delivery and organising activities, the initiative works on a more informal level and is volunteer run with little financial resources and no permanent facilities. Despite tensions, many workers centres in the United States work actively with unions or are even financed by them (Fine 2006, 458). While Oficina Precaria encourages union membership, describing it as ‘an improvable, but useful tool’ (Oficina Precaria Berlín 2018, own translation), they appear to not actively cooperate with unions. The group offers information on the German union system and lists both the risks and benefits of working with (official as well as anarcho-syndicalist) trade unions on their website. This ambivalence to and distance from unions reflects also the rather bad (or non-existent) experiences with unions many cleaning workers at Helpling as well as other recently arrived migrant (gig) workers have. As a constituency, they appear irrelevant for membership-oriented trade unions with the German Trade Union Federation (DGB).

In addition to networking through messenger groups and a counselling centre, in recent years some political groups have attempted to organise Helpling cleaners. Besides Oficina Precaria, groups such as Critical Workers Berlin and the Bloque Latino Americano have made efforts to support the networking and organising of workers in the city. One point of focus was the company’s actions during the COVID-19 pandemic, when a large share of Helpling cleaners saw their cleaning jobs on the platform suddenly cancelled or their health in severe danger through potential infection. The company’s lack of support for cleaners’ lost earnings and risk of infection crystallised in April 2020, when the company announced it would sell face masks for EUR 4 each in front of the company’s headquarters. In response, several groups organised a protest to give away free masks. For example, a Spanish-language call to action in Facebook groups called:

Helpling will be selling masks to cleaners and cleaning staff in the coming week from Tuesday to Thursday between 11 a.m. and 1 p.m. at their headquarters at Jägerstrasse 67 in Mitte. There will be people on the street against Helpling handing out free masks and asking for solidarity to protest Helpling abuse. If you work for Helpling and want to join protest actions […] you can get closer and get in touch. [Facebook group]

After this action, the company promptly stopped selling masks, thus preventing the spread of a larger protest. In addition to attempts at protest actions, information brochures in Spanish and English have been produced in recent years, which serve both as advice and for the political organisation of the workers. A ‘Workers’ Assembly Helpling’ also appeared as part of the May 1, 2021, anti-capitalist protests in Berlin (Bloque Latinoamericano Berlín 2021). Although the collectives also address workers and the public in German and English, the main focus remains on Spanish-speaking (and to a smaller extent, Portuguese-speaking) migrants. A Helpling worker who is also part of a political collective explains:

Our first step in building starts in our community. […] We start with the people we know, we are from Argentina, and people from Latin America have a lot of experience in our history with this. […]It’s easier for us to approach these people, and once we have organised a group of people, then the next step is: hey, how do we get in touch with workers of all nationalities? But we have to start with that first group. [Help_BE_7]

Although these organisations strive to expand their efforts to other groups in the medium term, this had not happened at the time of our research. More recent approaches have happened in the form of organising workshops geared at recently arrived migrant workers in the city (Lohana Berkins 2023; European Alternatives 2023), but it is unclear whether cleaning workers have been part of such events. Overall, this paints an ambivalent picture: on the one hand, the above-mentioned structures result in numerous benefits for the members of the communities themselves, who often have access to support in arriving in Berlin and in exchanging ideas about occupations, housing, and other issues in the first few days after their arrival. These structures can also catalyse collective action in broader conflict situations (such as during the COVID-19 crisis). Nonetheless, other workers on Helpling (such as those from Bangladesh, Poland, or Croatia) have little or no access to these resources or communication structures, leading to a stark contrast in access to (power) resources. Overall, despite these existing structures, no major mobilisations have formed around Helpling, such as those seen in other sectors of the platform economy.

Common messenger groups, community counselling services, and urban organising resources are three forms of collective resources that platform cleaning workers in Berlin have developed or could draw from, showing that such forms of work are far from incompatible with workers agency, collectivity, and resistance. Once in Berlin, migrants experience precarity both in their housing and labour condition. However, this precarity does not impact them in a homogenous way but rather through dynamics of differential inclusion based on race, class, and gender (Mezzadra and Neilson 2013). However, research on self-organisation of migrants, both in past (Bojadžijev 2012) and recent migration waves (see for instance, Alberti et al 2013; Berntsen 2016), shows that migrants are not passive receivers of processes of differential inclusion and racialisation. Such was observed by Floros and Jørgensen (2023) among young cleaning platform workers in Denmark. In our case, both the choice and the specific appropriation of cleaning platforms like Helpling by workers reflect deliberate choices and practices of adjustment, anticipation, and disobedience. From the perspective of migration studies, such processes have been described under the theme of ‘autonomy of migration’ (Bojadžijev and Karakayalı 2007; Mezzadra 2010), emphasising the agency of mobile populations in their migration trajectories and how their actions shape border regimes. Applied to platform labour, the gig economy becomes ‘a phenomenon simultaneously co-constructed by migrants’ agency and structural factors’ (Floros and Jørgensen 2023, 5), where both platforms and migrant workers are co-constituting how the platform operates (for instance, through a lack of bureaucratic hurdles but also lack of security and information). Some of our interviewees, who used the platform as a way to build an (off-platform) private net of clients for their own steady income, or as a stepping stone for a more permanent kind of visa, exemplify the possibilities that the platform holds for them. Others, who had started with the platform as a preliminary fix and then could not escape it properly, incorporate the more precarious side of it (comp. Floros and Jørgensen 2023).

Our findings also relate to discussions on subversion and organisational disobedience in the field of labour sociology, which have also proven useful to analyse conflict in the platform economy (comp. Heiland and Schaupp 2022; Altenried and Niebler 2022). While workers at Helpling have not staged explicit protests or strikes against the company, their mutual support and practices of dissent should not be deemed irrelevant. Based on the framework of labour process theory, Schaupp (2023, 185–206) proposes a conceptualisation of workers resistance that includes more subtle forms of dissent. His conceptualisation is described in the form of a four-layered pyramid: ‘practices of informal communication’ (Schaupp 2023, 193) such as casual talk or internal jokes are the most basic and commonly present layer, followed by more rare cases of ‘cultures of solidarity’ (Fantasia 1989) where workers act more deliberately in solidarity with each other. This is then followed by ‘organisational misbehaviour’ (Ackroyd and Thompson 1999), meaning actions by workers that are explicitly targeted against management or the company. The highest and most uncommon layer of informal workers resistance is called ‘self-organisation’ (Schaupp 2023, 193) and describes strategic coordination by worker collectives against the company. Reflecting our observations of Helpling workers in Berlin, most of these practices are present to some degree. Informal communication takes place in the form of the ‘whisper networks’ between workers. In contrast to fixed workplaces, where workers meet each other on a regular basis, such networking is an achievement in itself that sometimes requires considerable effort. ‘Cultures of solidarity’ were occasionally practised in the messenger groups of cleaners (encouraging each other after a bad shift and sharing advice on how to deal with problems). Cases of organisational misbehaviour were also present, for example, when workers exchanged how to (strategically) take clients off the platform without the company’s knowledge. Such practices receive support through counselling centres like Oficina Precaria or groups like Bloque Latinoamericano, which have led to more sustained efforts of organisational misbehaviour and prototypes of self-organisation such as during the protests against the company’s COVID-19 measures. The different layers of collective support and disobedience between Helpling workers and their communities can therefore be described as providing a ground for more sustained workplace resistance. Following Schaupp, it is not far-fetched to conclude that such practices usually happen ‘under the radar’ but serve as a basis for more explicit struggles. Still, the fluctuation of workers and the named existing hurdles (spatial diffusion, lack of identification, and legal fragmentation) remain as material limits to this and have to be taken into account.

The strategies we have described here do not happen out of context but should be interpreted against the background of lacking initiative by legislators, the state, and trade unions. Both Germany’s federal government and the Senate Administration of Berlin have been in a public communications process with Helpling and other gig economy firms, which has even led to policy plans on how to improve working conditions for platform workers in 2020 (Bundesministerium für Arbeit und Soziales (BMAS) 2020). However, despite several reports and studies on the company, no legislative process has been initiated almost a decade after the company’s foundation, making regulation of the sector hard to expect for workers or community groups. Little tangible progress has also been made by trade unions in Germany, who have organised large conferences on the future of work and funded publications on the working conditions of platform firms (including Helpling). On the one side, trade unions appear legally limited in supporting formally self-employed workers. At the same time, even existing lobbying capacities towards regulators seem to have been not exercised or have not been fruitful. A notable exception to this is the agency BEMA, which is financed by the German Trade Union Associations (DGB) and the Berlin Senate Administration and offers counselling to migrant workers in Berlin. However, the work of this agency is limited to counselling and educational work and has no decision-making capacities on a policy level. So, in terms of concrete conditions, not much appears to have changed for Helpling workers in the city since the arrival of Helpling a decade ago. While trade unions generally have difficulties operating in the gig economy (with notable exceptions), a strategy for domestic cleaning platforms appears to be missing in particular, a fact that likely also stems from the relative negligence of care worker struggles (comp. Baum and Carstensen 2022).

Conclusion

Working at the cleaning platform Helpling has become a source of income for recently arrived migrants in Berlin who lack access to the formal labour market. Cleaning work via platforms is similar to other forms of platform work, such as delivery and ride-hailing, but poses special hurdles for the collective organisation of workers. In the case of Helpling, three aspects come into play: first, a spatial fragmentation that goes hand in hand with a specific subjectification vis-à-vis the customers; second, an absence of identification with the work, which sometimes results from the stigmatisation of cleaning work; and third, the absence of a legal framework for the collective representation of interests. Nevertheless, platform-mediated cleaners in Berlin organise collectively in several ways: in common messenger groups for exchange, through a counselling centre as a source of information and contacts, and through political organisations for mobilisation. These forms of collective exchange, disobedience, and organising are mostly limited to Spanish-speaking or Latin American groups, which represent a significant portion of the worker population on Helpling. They are structurally linked to the hurdles we have outlined above, which remain challenging due to lacking initiative by both state actors and trade unions. The findings of our research relate to debates in critical migration studies, which emphasise the agency of mobile workers and its impacts, as well as to discussions within labour sociology and the labour process debate about the degrees and impacts of informal resistance strategies in contemporary workplaces.

The resources, forms of exchange, and organising efforts of Helpling workers in Berlin show that workers on cleaning platforms are not passive victims of the platform economy but also actively use the platform in their interest, thereby subverting the company and sometimes working actively against it. In connection with other structures in the city, not only they can exchange information about problems with customers or the threat of sanctions from the company, but they can also network politically. However, these networks in Berlin have so far been limited to certain groups and have not yet led to larger collective mobilisations against the company. One reason for this is the lack of tangibility of the company, which, unlike other companies, does not rely on handing out work equipment or work clothes and can thus avoid any physical contact with the workers. The described protest action in front of the Berlin Helpling headquarters presents a notable exception, in which the company becomes concretely tangible. As an exception, however, the incident seems to confirm a rule.

For the future, various scenarios seem conceivable. It is possible to imagine broader conflicts and mobilisations that expand the potential of already existing messenger and social media groups based on existing networks and through tactics of organising online. At the same time, the absence of any workplace socialisation raises the question of whether the company level is an appropriate basis for mobilising workers in this field. Federal and local government, as well as community actors, appear more tangible as targets of mobilisation here. Alternative and non-profit cleaning platform models have been developed in other regions, such as by the National Domestic Workers Alliance (NDWA) in the United States, which provide a platform for cleaners and nannies, and which collects social security benefits from clients on behalf of workers (Kampouri 2022, 15f.). Other examples include platform cooperations like Up&Go in New York, or a booking portal set up by the group Las Kellys, a movement that has politicised the working conditions in the industry and founded its own digital booking portal in 2021 (Bunders et al 2022; Alcalde-González et al 2021). Lastly, some trade unions have made achievements in the field, most notably the collective agreement between the cleaning platform Hilfr and the union 3F in Denmark (Jesnes et al 2019).

Moreover, conclusions for other forms of platform work can be drawn from the dynamics highlighted here, for example, for care platforms such as Careship or Betreut.de, which are confronted with similarly strong spatial fragmentation, or the ride-hailing industry, which is fragmented by subcontractors in Germany (Baum and Kufner 2021; Niebler et al 2023b). In the medium term, it remains to be seen how the legal situation regarding the employment status of cleaners will change in the wake of the EU Directive on platform work. A repeal of self-employment status, which is associated with many disadvantages, to that of an employee would by no means solve all problems, but it would visibly increase the possibilities of company-wide organisation.