Platformed professionalization: Labor, assets, and earning a livelihood through Airbnb

This paper analyses processes of professionalization on Airbnb in Berlin, exploring who is able to take part most successfully in urban value creation processes facilitated by short-term rental platforms. In doing so, it intervenes in debates on platform urbanism that focus on the role of digital platforms in reconfiguring urban governance and livelihoods. Combining a political economic approach and affordance theory, I conceptualize professionalization as a particular platform logic that benefits Airbnb and hosts who are able to take part, while reinforcing existing inequalities. Drawing on eight months of fieldwork in Berlin, I show how these professionalization practices differentially affect the strategies and practices of hosts, offering benefits to some while worsening the position of others who are unable or unwilling to professionalize. As such, professionalization processes produce inequalities and power asymmetries both on and off the platform, between hosts as well as between the platform owner and platform users. In a context where a growing number of city-dwellers rely on platforms to generate their livelihoods, such power shifts resulting from platform dynamics have a significant impact on who is able to benefit from platformization and thrive in a platform society.


Introduction
At Airbnb's 1 inception in 2008 its offerings ostensibly consisted of little more than a simple airbed and in some cases breakfast. Initial interpretations of platforms in the so-called 'sharing' or 'peer-to-peer economy' (Botsman, 2015) were positive and stressed that the low entry barriers allowed anyone to join the market. Yet, over the years Airbnb has vastly outgrown its initial status. The platform developed into the globally largest short-term rental (STR) provider with currently over 5 million active listings, including shared homes, 'serviced apartments,' (boutique) hotels, and private islands (Airbnb News, 2020).
Fierce and contentious discussion has emerged in both academic and public debates, highlighting how Airbnb's spectacular growth has come at a cost As cities have proven to be a particularly fertile breeding ground for platforms, the notion of 'platform urbanism' has been developed to conceptualize the reconfiguration of relations between technology, capital, and cities (Sadowski, 2020: 2). One of the main issues in cities is that Airbnb takes housing stock off the regular housing market, effectively reducing the availability of housing (Wachsmuth and Weisler, 2018). Cities have responded with new measures to regulate or even ban Airbnb rentals (Aguilera et al., 2021;Ferreri and Sanyal, 2018;Nieuwland and van Melik, 2018), but these measures have faced strong opposition from both local hosts and Airbnb's legal department (van Doorn, 2020). Meanwhile, as a 'policy entrepreneur' (van Doorn, 2020), Airbnb deliberately and explicitly tries to influence urban policy-making processes. In particular, the company has advocated for favourable regulations that allow homeowners and tenants to take part in 'home-sharing' practices through digital platforms.
Home-sharing is not all that happens on Airbnb though. Recent studies have shown that the capability to create value on Airbnb is unequally distributed, suggesting that 'professional hosts' and 'professionalization' have become increasingly important on the platform (Deboosere et al., 2019;Wachsmuth and Weisler, 2018). What a professional host is has been defined in terms of the number of listings they manage, the experience they have in managing these listings, and the type of listings and their year-round availability (Adamiak, 2018;Deboosere et al., 2019;Grisdale, 2021). Analyses of spatial and quantitative data on Airbnb listings and bookings have already provided much insight into the share of such professional hosts within the STR market, have demonstrated that professionalization is an important determinant of pricing and revenue (Deboosere et al., 2019), and have shown that in some cities 'ghost hotels' (large portfolios of private rooms or properties bought entirely for STR use) are proliferating on Airbnb (Wachsmuth and Weisler, 2018). Professional hosts, who represent an increasing share of the available listings and of the total platform revenues, are assumed to operate more like businesses than 'regular hosts. ' While scraped 2 quantitative and spatial data on Airbnb have resulted in useful insights, qualitative studies have thus far been scarce. It remains unclear, for instance, how professional hosts differ from 'regular hosts' or home-sharers, how they operate in practice, and what it takes to become a professional host. In fact, the definition of 'professionalization' on Airbnb itself remains unclear. When considering professionalization, it should also be borne in mind that this particularly benefits Airbnb as a company, harnessing its power over the platform and its users. In a society where platforms are emerging as a central institutional form, and which are (re-)organizing the relations between markets, citizens, and the state (Van Dijck et al., 2018;van Doorn, 2020), such shifts of platform power resulting from programmatic platform dynamics might have tremendous consequences regarding who is able to benefit from platformization and thrive in a platform economy.
Instead of focusing on professional hosts as a category and professionalization as a given, I suggest that it is more fruitful to unpack processes of professionalization, and to conceptualize them as a distinct platform logic (Andersson Schwarz, 2017) that creates and maintains unequal power relations both between hosts themselves, and between platform owners and hosts. In this paper, I seek to contribute to our understanding of platform-mediated STR, and platformization in general, by providing an in-depth account of processes of professionalization on Airbnb, based on long-term qualitative research in Berlin. As I will show, some of these processes are sociotechnically assembled by the platform, while others are not facilitated or are even discouraged by the platform. Moreover, professionalization, as engineered by the platform, is constantly negotiated and contested by diverse hosts in different (regulatory) contexts, resulting in variegated outcomes. In practice, professionalization is not necessarily a successful or tenable strategy for many hosts; closer scrutiny of actually existing processes of professionalization shows a more complex and ambiguous picture of both failure and success.
In what follows, I first present an overview of the relevant literature on the professionalization of STR, followed by an explanation of how I conceptualize processes of professionalization as a particular platform logic. After a brief methodological discussion, I present my analysis of findings based on fieldwork in Berlin, focusing on the stories of three different hosts in the city, who represent divergent cases of precarious, privileged, disrupted, and platformed professionalization. I discuss what these cases can tell us about platform-mediated value creation and how these processes are engineered into the Airbnb platform, concluding with suggestions about what professionalization entails more broadly as platforms emerge as an increasingly present institutional form.

STR and processes of professionalization
In this section, I outline existing approaches to STR professionalization. I argue that a binary distinction between home-sharers and professional hosts is problematic. Instead, I suggest that it might be more fruitful to look at processes of professionalization as a particular platform logic.
What professional hosts are and what they do has been described in a series of studies using scraped Airbnb data. These studies generally construct indicators to measure professionalization by focusing on how the characteristics of hosts and rental properties affect listing revenue. The most frequently mentioned indicator is 'multi-listing. ' Li et al. (2019), for example, define "professional hosts as those who offer two or more unique units on Airbnb" (7). Similarly, Gibbs et al. (2018) designate hosts as professional "if they ha[ve] two or more active listings" (50), Dann et al. (2019) regard professional hosts as "those who offer more than one listing" (451), and Deboosere et al. (2019) find that hosts with 10 or more listings are even more 'successful,' i.e. they bring in more revenue per listing.
Other proposed indicators of professionalization have been the renting out of entire apartments (as opposed to rooms) and high yearly availability. For Ioannides et al. (2019), for example, "'professional Airbnbs' refers to entire homes/apartments, controlled by owners with multiple listings (two or more), available for over 60 days per year" (10-11).  expand the definition of professionalization to include what they call "commercial operators with 'ghost hotels'" (2). Deboosere et al. (2019) further expand these indicators by suggesting the inclusion of 'Superhost' status, 3 the number of reviews per listing, and whether Instant Booking is allowed. 4 Deboosere et al. (2019) conclude that "hosts who treat their listings as de facto hotels rather than opportunities for part-time 'home sharing' are considerably more successful in the Airbnb marketplace" (152-153). The success of professionalization is evidenced by Katsinas (2021) who shows that 17% of the annual STR revenue in Thessaloniki is earned by the top 1% hosts -21 in number.
While the distinction between 'home-sharers' and commercial operators thus remains of central concern, Kadi et al. (2019) argue that a binary understanding of these groups does not hold. Since definitions of professional hosting generally include entire apartments only, rooms rented on Airbnb would be categorized as home-sharing listings. However, based on empirical material from Vienna, Kadi and colleagues find that these rooms are in some cases also rented out using 'commercialized provision practices,' including year-round availability. They conclude that "commercial provision of Airbnb homes may thus not only be a strategy for large-scale landlords …, but also be pursued by small-scale entrepreneurs or amateur landlords" (ibid.: 17). The consensus seems to be that professionalization amounts to a spatial and temporal expansion of 'hosting': the rental of multiple, often entire, apartments that are available all year round with the ultimate aim of making as much profit as possible. 5 Limited attention has been paid to the question of how hosts achieve professionalization in practice. Grisdale (2021) and Cocola-Gant and Gago (2021) form two notable exceptions. Both suggest that intermediaries play an important role by offering management services for STR and advising potential investors. What are the conditions of possibility for professionalization and how do they affect inequalities in access to value creation? Moreover, existing accounts based on quantitative data are developed on the basis of the information that ends up on the platform. Excluded are the trajectories that lead to these spaces ending up on the platform; the stories of rentals that were taken off the platform; and potential hosts who were ultimately unable to join the platform.
While Airbnb rentals are closely tied to the housing market, in this paper I focus on the labor involved in hosting. It is clear that professionalization increases listing revenue, but it is unclear how processes of professionalization transform the various types of labor involved in hosting, and perhaps require new forms of labor. After all, hosting on Airbnb does not just require a home or space to rent out; it involves emotional and affective labor, from both hosts and neighborhood residents (Knaus, 2018;Roelofsen, 2018;Spangler, 2020), as well as physical (e.g. cleaning) and digital labor. In order to understand who is able to benefit from platforms and who is excluded, it is important to see which types of waged or unwaged labor are (de)valued and which types can be automated and taken over by algorithms (cf. Fields and Rogers, 2021).
To develop a more processual and multifaceted understanding of professionalization, I suggest looking at processes of professionalization as a particular platform logic. The concept of platform logic is derived from Andersson Schwarz (2017), who conceptualizes it as "a twofold logic of micro-level technocentric control and macro-level geopolitical domination, while at the same time having a range of generative outcomes, arising between these two levels" (376). In studying platforms, he suggests, micro-level perspectives on platform practices should be combined with macro-level perspectives on the political economy of platform owners, "simultaneously acknowledging the technical capacity of unyielding local control and its consequential concentrations of global dominance by a handful of corporate actors" (ibid.: 378, emphasis in original).
I depart from Andersson Schwartz's notion of 'unyielding local control' on a micro level and instead take a more dynamic, non-deterministic approach to platforms' power over users. 'Hosting' on Airbnb is always mediated by the platform and as such affects how someone can host Platforms "set the stage for actions to unfold" (Bratton, 2015: 47, emphasis in original) without fully determining what these actions should look like. In this sense, affordance theory comes in useful to account for the technical aspects of the platform in relation to its users. Affordance theory is an approach derived from Science and Technology Studies that aims to theorize technologies and their users in conjunction. Accordingly, affordances are defined as "the 'multifaceted relational structure[s]'… between an object/technology and the use that enables or constrains potential behavioural outcomes in a particular context" (Evans et al., 2017: 36). Thus, they are more than technical features per se; rather, affordances emerge in the relation between a technical artifact and its user (Bucher and Helmond, 2018). Depending on the user's capabilities, knowledge, and resources, affordances can take multiple forms and strengths: artifacts might allow, request, encourage, demand, discourage, or refuse specific behavior in particular contexts (Davis and Chouinard, 2016). Consequently, empirical inquiries should go beyond a mere description of particular tools and features and focus on the conditions of affordance: "for whom and under what circumstances" do technological objects afford? (Davis, 2020: 103). If we follow the analogy of platforms setting 'the stage for actions to unfold,' then affordance theory might help us to understand what the stage looks like, who is allowed to enter the stage and under what circumstances, and what roles are they able to play.
In the case of Airbnb, platform logics hint at the interplay between technical features and tools, such as the Superhost status discussed above, and the political-economic interests of Airbnb as a company. By meeting particular targets, hosts can attain Superhost status, resulting in particular rights and privileges being bestowed on them. While hosts are not obliged to meet these targets, these rewards clearly encourage hosts to work to meet them. Meeting these targets directly benefits Airbnb, since the Superhost status results in higher nightly rates for hosts and higher service fees transferred to Airbnb. Similarly, I suggest that professionalization processes are partially engineered by the Airbnb platform and tied to its business model, though they are also always shaped by differentially situated users embedded in local spatio-historical contexts.

Methodology
To explore how professionalization actually takes place on Airbnb, data was gathered during eight months of fieldwork in Berlin. Berlin is an excellent site for exploring this, as the conditions are favorable for both 'home-sharers' and 'professional hosts.' On the one hand, Berlin's population has a relatively low income compared to the rest of Germany, and even more so compared to other European capitals (Bernt et al., 2014). Within the city, large income inequalities and high unemployment rates exist, making home-sharing potentially attractive to large shares of the population. On the other hand, housing is under severe pressure, with demand for housing vastly outnumbering supply. As a popular destination for both travelers and longer-term residents from both Germany and abroad (Novy, 2018), rents have increased drastically over the past decades. The rent gap in the city (Smith, 1987) is large due to relatively low housing costs in the past, strong population growth, and the recent influx of a large amount of investment capital (Elias, 2017). Even though measures are currently being taken to release pressure on the rental market, 6 Berlin is still a popular place to invest in real estate (Collinson, 2018), potentially making it an attractive market for professional hosting as well. At the same time, regulation of the STR market is strong. STR of entire homes was banned citywide between 2016 and 2018. While STR was allowed again from 2018, strict conditions apply and large fines of up to €500,000 were introduced for offenders (Paul, 2018). This makes Berlin a good test case for examining how professionalization takes place under conditions of strict regulation.
During the fieldwork period, I conducted 35 in-depth interviews with past and current Airbnb hosts, home-sharers and multi-listers, and with companies offering services catered to Airbnb hosts. This method was chosen in order to gain insight into respondents' relations to platforms, and how they reflect upon them (Rinaldo and Guhin, 2019). Interviews were held in either English or German and focused on respondents' backgrounds, their motivations to use the Airbnb platform, how their income from Airbnb related to further sources of income, and their views on 'good hosting.' While a topic list was used to ensure that these topics were touched upon, the interviews were loosely structured and open-ended to leave space for unexpected topics and participants' own concerns. In addition to the interviews, I participated in gatherings of Airbnb hosts, neighborhood events, and town hall meetings. To assure anonymity, pseudonyms are used and some details that might allow for the identification of respondents, including occupation and neighborhood, have been changed or withheld. 7 Further data was gathered through a variant of the 'walkthrough method' in order to get acquainted with the features and functionalities of the Airbnb platform. This method is "a way of engaging directly with an app's interface to examine its technological mechanisms and embedded cultural references to understand how it guides users and shapes their experiences" (Light et al., 2018: 882). I used it to get a sense of the Airbnb platform's affordances by directly interacting with the platform, both using a mobile device and through the web interface. Airbnb's online Help Center for hosts, 8 as well as the Community Center, 9 were consulted to get a better understanding of how particular functionalities work and how they are perceived by hosts.

Modes of professionalization
This section presents three stories of diverse hosts in Berlin who took part in one or more processes of professionalization. The three individual trajectories detailed here exemplify broader patterns and show some of the most diverse outcomes. The first trajectory, focusing on the stories of Irena and Thomas, involves contrasting forms of precarious and privileged professionalization: Irena has been able to professionalize as a host, but under conditions of high risk and with insecure and uncertain outcomes, while Thomas, who is not involved in the daily work of hosting, is able to benefit from previously accumulated capital. As rental housing has long been the dominant tenure and home-ownership rates are low in Berlin, Thomas' mode of privileged professionalization is unattainable to most. Correspondingly, my respondents were mostly tenants in situations akin to Irena's: without the necessary permission from their landlords, hosting is risky or even impossible for them. The second trajectory looks at a case of disrupted professionalization: Elli initially sought to make a living by renting a few apartments, but was unable to continue her business due to changes in her personal circumstances and local regulations. Elli presents as a unique case that exemplifies that property alone is not enough. The final trajectory focuses on Johannes, whose company acts as an intermediary in support of real estate owners. By minimizing asset ownership, collecting data, and using software to manage his business, he mirrors the strategies of lean platforms, which are characterized by their "attempt to reduce their ownership of assets to a minimum and … profit by reducing costs as much as possible" (Srnicek, 2017: 49-50). A small minority of my respondents were able to professionalize like Johannes, but his case exemplifies processes of professionalization that are unique to digital platforms. Together, these three trajectories highlight how professionalization should be understood as a set of dynamic and multi-faceted processes.
Precarious and privileged professionalization I first met Thomas at an event for Airbnb hosts. After we had chatted for a short while, I asked if I could interview him about being an Airbnb host He immediately introduced me to Irena, saying that she was the one who actually "does the work." Their contrasting trajectories and situations represent cases of precarious and privileged professionalization. As I will show below, central to this contrast are the resources they were able to bring to the platform, the work they had to do, and the present and future returns they could expect.
Both Irena and Thomas were from a neighborhood in former East Berlin. Irena worked as a photographer and described her life as "almost not structured, no regular things. Just Airbnb, that makes really a regular thing." She was receiving social security, an amount of €700 a month after rent, and had an irregular income from her work as a photographer. In 2014, she started renting out a room in her rental apartment through Airbnb. This generated a fair additional income but was very risky: if she were to get caught illegally subletting, she would not only receive a fine but would also lose her social security and her apartment. Besides hosting guests in her home, Irena did most of the hosting work for Thomas and several other property owners, including cleaning, communication, and guest check-ins. For Irena, things were going well at that moment, but she was not where she wanted to be. She returned to her aspirations regularly: You know, it is just so beautiful, what the people are writing about the experience they have with me. So I'm really good in that, I'm really, really good in that. So that's the reason why I'd love to have more space, to be the queen of my house … and welcoming the people, and give them just a good feeling. I'm really good in that. I think if I would do it more in the professional way, even more professional, having a bigger space and all that, I could create an income just out of that, and I could leave my social money for example.
The way Irena took care of guests was highly valued, as evidenced by their reviews and her Superhost status, but she was not able to capture this value in monetary terms. Taking the next step, to host in "the professional way, even more professional," was unattainable for her, as she did not have access to a space where she could do so on her own terms. Irena expressed her frustration about this: "I just want to use my potential just for myself. I don't want to provide it for others." When I asked Thomas if he considered himself to be an Airbnb host, he hesitated and responded that he was more of a property manager (Objektverwalter); all the services (Dienstleistung) he "gives out to others" such as Irena. He owned a small construction company, a few dozen regular apartments in Berlin, and two STR apartments that he rented out on Airbnb. He had been able to finance the latter by taking out loans, using the apartments he already owned as collateral. He regarded this as an investment in the "middle to long term," with the plan of refinancing with his own capital and getting rid of the loans. In the long run, he hoped to be able to not only benefit from the future value increases of his property, but also to turn them into assets (Birch and Muniesa, 2020) that would ultimately deliver him a stream of "passive income." This income is not completely passive though: what Thomas mentioned that Irena "does the work" it remained implicit that she works for Thomas. Still, Irena did not want to complain too much and felt like she also had to take into account "the perspective of the owner of the flat, because he says he's not making enough money." Thus, a somewhat ambiguous relation exists between Thomas and Irena, which is discursively and technically mediated by the Airbnb platform.
Discursively, this relation pivots on the notions of 'host' and 'hosting'. While these suggest a singular activity undertaken by a singular person, 'host' might refer to users with different roles: 'primary hosts,' 'listing admins or owners,' and 'co-hosts' (Airbnb Help Center, n.d.). 10 Irena, as a co-host, "help[s] listing owners [like Thomas] take care of their home and guests" (ibid.). While the 'co-' in co-hosting and the framing of work as 'help' suggests that the relation between them is collaborative, an unequal power relation is obscured: the crucial difference between co-hosts and listing owners is that the latter have access to the payout from Airbnb, while a co-host does not. Irena would immediately lose her income as a co-host if Thomas would stop renting out his STR apartments.
Furthermore, the relation was socio-technically facilitated by the platform as Irena did not receive instructions from Thomas but from the Airbnb app. As she explained, in the app she "can see all the check-ins, the check-outs. Here, I always double check that I have everything under control." The platform gave Irena a sense of agency over her work, even though it still guided her work by showing the things (e.g. check-ins and check-outs) that she had to work on. Even though Irena still worked for Thomas, he was left out of the picture and possible tensions between them were ameliorated as her work was "not only mediated, but governed, by smartphones, digital platforms, and apps" (Fields and Rogers, 2021: 3). The platform mediated an ambiguous relation between asset owner and hosting laborer by creating a particular form of distributed agency (Richardson, 2020), giving the work of the laborers an entrepreneurial character while still governing the work the latter performs. In this sense, Airbnb is no longer only a home rental platform; it also becomes a platform for managing the labor required to rent out these homes. This affords property owners to outsource hosting labor, scale up their business and manage multiple STR apartments at a distance, while laborers have to resort to poorly renumerated and precarious type of informal, 'improvised entrepreneurialism' (Rossi and Wang, 2020).

Disrupted professionalization
Even for hosts who are able to bring assets to the platform, the improvised entrepreneurialism afforded by Airbnb might remain precarious, as Elli's story shows. As an artist without a steady income, Elli moved to Berlin from the Netherlands in the early 2000s attracted by much lower costs of living and housing. Using their savings, Elli and her then husband had been able to buy a home. Soon they were also renting a studio apartment next door, as well as another apartment. Elli started to rent out both as 'guest houses' around 2009, before Airbnb came to town. When Elli's father bought an apartment in the same neighborhood in order to be able to visit her in Berlin, Elli also started to rent this out to tourists when it was vacant. This fitted very well with Elli's situation: It seemed to me even very natural to do that, because I had space, I had time, and I was pretty motivated also to host guests. I was also myself not a Berliner, but I came from outside, so I would also be like, yes, nice to host guests. It was just a great possibility. It was a good opportunity for me to generate income.
Even though Elli was renting out three apartments, she "didn't want to bring it into the business-oriented direction, but more [a] kind of occupation and on a very human scale." Like Irena, she preferred the hospitality aspect above optimizing hosting as a business and preferred to keep it personal. She always met guests in person to hand over the keys and provided them with handmade guidebooks for the neighborhood. When a friend told Elli about Airbnb, she decided to create listings there as well, in addition to her own website that she had relied on before. Renting out through Airbnb made her work easier and resulted in more income, but Elli remained somewhat doubtful: It was very easy through Airbnb. It generated a lot of money. But maybe it would have been nice to keep it really this kind of personal touch and not go into this big scale business.
Her doubts about Airbnb's influence on her business might have resulted from how the STR market developed in Berlin after she started using Airbnb. As the number of listings grew exponentially, the city of Berlin started to develop policies to regulate STR. This eventually led to a ban on STR of entire apartments, which came into effect in May 2016. In the meantime, Elli had divorced from her husband. She kept one of the two guest houses and was also still renting out her father's apartment. She applied for a permit to continue her business but was denied. She was allowed to rent out the furnished apartment, but only for periods of at least one month. This was much less lucrative and did not result in a sufficient income.
When I last spoke with Elli, she was receiving social security and had just moved to a subsidized apartment a few blocks away from her old home. As she had just had a baby, the studio apartment that she owned and rented out was unsuitable for living in with a child. Now that she was renting out the studio for longer periods, she had lost her pleasure in hosting: I totally enjoyed hosting on Airbnb, like I really, really liked it. I really liked maintaining the flats and I really liked getting people from everywhere and really liked hosting them.
Elli's story shows the intricacies and complexity of processes of professionalization. She used to manage multiple listings that she had rented with the aim of subletting them on the STR market. Thus, she might have been regarded as a professional host who would have been able to upscale her business. She was not fully convinced about professionalization when it would have required her to take a distance from the hosting work by outsourcing cleaning for example. Her preferred mode of professionalization had become impossible due to the huge increase in listings on Airbnb and the subsequent changes in the local regulations. Her story also shows that having property or a space to rent out does not guarantee success in a context of changing regulations. While Thomas had to sell one of the multiple properties in his possession, Elli had been unable to circumvent the regulations as she was much more personally attached to the apartments she rented out. While she did have access to property, she was unable to treat them as capital as they were tied up in personal relations. This meant that her professionalization was disrupted.

Platformed professionalization
Johannes did not have real estate or financial capital at his disposal, like Irena and in contrast to Thomas, but was able to accumulate and exploit another type of capital: data (Sadowski, 2019). He took part in a novel type of platformed professionalization that is enabled by typical platformization strategies and platform affordances including datafication, upscaling, and minimizing asset ownership.
In return for a 25% cut of the rental income generated, Johannes' company offered complete hosting services to property owners. For about 60 of its 200 customers in Berlin, they provided full management, including the creation of the listing, vetting and receiving guests, undertaking small maintenance chores in apartments, and cleaning once guests had left. The remaining 140 customers managed their own listings on the platform and called in Johannes only for cleaning and check-in. While for archetypical 'home-sharing' hosts, there is a clear limit to the number of guests they can receive in a year, intermediaries like Johannes are able to benefit from economies of scale: … we can do our best to keep the cleaning quality, and we have way more experience [in] how to handle situations. It's a difference if it's like one single host [who] can't do all the stuff or solve a problem quickly. And we have 40 guests a week, [a homesharing host] might have three guests a week. Like Thomas, Johannes was able to professionalize by scaling up; not by acquiring more apartments, but by serving more customers and gathering experience in a high pace. Working as an intermediary meant that he could operate similarly to 'lean platforms' (Srnicek, 2017) by minimizing asset-ownership and profiting by cutting down on costs. Besides benefitting from economies of scale this also helped him develop strategies to maximize revenue, for example by strategically navigating local regulations. One strategy to do so stood out, namely the combination of STR with monthly or longer-term stays. The latter are generally not included in the 90-day limit for STR in Berlin and Johannes made use of this by strategically picking periods with high demand for highly profitable STR, while the remainder of the year he rented out apartments for monthly or longer stays. The value of Johannes' experience was also evident from the consulting that he provided to property owners who were working on setting up their own STR apartment or business.
To cut down on costs within his company, Johannes removed as much friction as possible from listing management. The Airbnb platform affords this through an Application Programming Interface (API), a crucial element for platforms as they turn a website "into a platform that others can build on" (Helmond, 2015: 4). Airbnb's API allows users to access and edit data in Airbnb's database in an automated and programmable fashion, but it was not accessible directly for Johannes: "they don't give API access unless you're a … big PMS". PMSs or Property Management Systems are cloud-based software services that connect to the API and facilitate the centralized management of large numbers of listings.
But a platform simply to manage listings was not enough for Johannes. He not only used API-connected algorithmic pricing tools, a type of software that aggregates and analyses market data to determine optimal pricing and maximize revenue, "what I [also] need is better coordination with my staff, for the cleaning." Thus, he was working on a more extensive platform: I'm building a platform where we have our client data, all the data about all the apartments, because it's important that we have all the details. It starts with easy stuff, like where to find it in the building, but it gets to where are the cleaning supplies, […] how is it to be done with the sheets, if we have three guests coming.
[…] The time-tracking, so we have up-to-date information about how long each cleaning took, that should also get in there.
While Johannes' business, like lean platforms, minimizes ownership of most assets, assets in fact are of crucial importance. First, Johannes' business does "own the most important asset: the platform of software and data analytics" (Srnicek, 2017: 49-50). Using his background as a software developer, he developed a platform that formed a central place to gather information and data on all aspects of the company; data that he then used "to optimize [and] to automate a lot of things." Instead of relying on real estate or financial capital, Johannes' company relied on the accumulation and exploitation of 'data capital' (Sadowski, 2019). Second, Johannes is reliant on his clients' assets and he gets access to those since he is able to add value for the asset-owners. As Christophers suggests, under contemporary rentier capitalism the "ability to perform work of a certain perceived value vis-à-vis the asset" (2021: 13) is a well-rewarded skill.
Mastering this skill allowed Johannes to build his business and take part in a novel type of platformed professionalization, effectively combining upscaling, datafication and minimization of asset ownership. The availability of an API combined with his technical background afforded Johannes to collect data capital on the transactions and operations of his business, which in turn allowed him to analyze and further optimize business processes. This, in turn, was valuable to property owners who outsourced hosting to Johannes and his company so that they do not have to invest in this themselves. On the US housing market technological developments including smartphones, digital platforms and apps opened new avenues for financial accumulation, allowing landlords to aggregate ownership of geographically dispersed portfolios (Fields and Rogers, 2021), the technological affordances of the platform allow Johannes to professionalize hosting on Airbnb building on disaggregated ownership. While Airbnb's co-hosting functionality enabled Thomas to distance himself from hosting, the Airbnb API allowed Johannes to develop his own kind of platformed professionalization.

Engineering professionalization
While Airbnb likes to argue that it enables middle-class households to generate an extra income, this dynamic suggests that Airbnb also enables growing inequality between asset-less hosts and asset-owning hosts. The latter can leverage Airbnb and benefit from processes of professionalization to increase rent from their assets and perhaps increase their assets' value. Professionalization works for asset owners: multi-listing, renting out entire homes, and experience in hosting all increase hosting revenue, but in these processes 'hosting' is broken down into a lot of small pieces. This reinforces Semi and Tonetta's (2021) recent argument in this journal that analyses of STR suppliers should take social class into account. The unequal chances for hosts to professionalize suggest that particular consideration should be given to both classed employment differences and the unequal distribution of assets among hosts.
Should we regard this as a coincidental, unintended side-effect of what is otherwise a platform that gives everybody a fair chance? Recently introduced features, including the API (officially announced in 2017) and PMSs that connect to it, show that Airbnb increasingly functions as an 'operational platform,' "allow [ing] investors to outsource and automate many aspects of the rental and property management process" (Fields and Rogers, 2021, 4). First, the growth of Airbnb's Software Partners program 11 suggests that property managers who make use of API-mediated interfaces are an important target group for Airbnb. The total number of partners offering PMSs, grew from 149 in 2019 to 193 in 2020 (Airbnb.com, 2019a;Airbnb.com, 2020). By their nature, PMSs afford to a particular group of asset-rich property owners, not single-listing 'homesharing hosts'. Moreover, PMSs have been in widespread use in the hotel industry since the 1980s (Collins, 1988). It is for such players in the hospitality industry, including asset-rich property managers looking to outsource 'hosting,' that the API affords. Indeed, Airbnb started accepting 'boutique hotels' to the platform in March 2018 (Airbnb Newsroom, 2018).
Second, similar functionalities have now been made available to a larger group of users through a set of 'Protools' available in the Airbnb Dashboard. Introduced in 2019, these tools are supposed to "make it easier than ever to manage listings and reservations at scale" (Airbnb.com, 2019b). Besides multi-listing functionalities, Protools include 'Teams,' a set of features that expands the functionality of the co-hosting system by affording property owners and managers to give multiple team members distinct permissions to access specific listing and booking details. Cleaning staff, for example, can be granted permission to see arrival and departure dates, but not guest details.
Like PMSs, Protools afford for property managers by facilitating an increasing division of labor, separating the management and daily operations of STRs. On the one hand, the platform affords managers to shift their attention to the optimization and automation of processes on the platform. What cannot be automated, on the other hand, are the daily operations that require someone to be present to provide physical and affective labor. Thus, the platform affords Johannes to professionalize by tracking and optimizing the cleaning labor at a distance, while Irena was unable to scale up or outsource the labor involved in cleaning and receiving guests. As such, the unequal opportunities to professionalize on Airbnb seem to further reinforce the inequalities generated by the capacity of capital and wealth to generate larger income streams than labor (Adkins et al., 2021;Piketty, 2018).
While still central to most accounts of Airbnb, the figure of the 'host' and the work of 'hosting' is increasingly being broken down into separate types of labor that can or cannot be outsourced, automated, and optimized. 12 Some aspects of hosting labor have most likely always been outsourced (e.g. when neighbors hand over keys while a host is on holiday), but Airbnb's efforts to facilitate these practices by integrating them into the platform suggest that Airbnb is willing to further decentralize the figure of the host. Moreover, Airbnb promises guests personal, unique experiences, but simultaneously stimulates hosts to de-personalize 'hosting' and to regard it as an optimizable business process that generates a reliable and steady income stream. Ideally, labor is eliminated altogether: the Host Assist program 13 allows smart lock providers and other companies to integrate their services on the Airbnb platform, allowing guests to enter their holiday apartments without meeting anyone in person. To be eligible for business travelers, such a self-check-in option is required by Airbnb. Such changes help property owners to run an STR without having to be physically present, ultimately transforming hosting into ghosting.

Conclusion
As platforms emerge as an increasingly present institutional form (van Doorn, 2020), which are (re)organizing the relations between markets, citizens, and the state, the distinction between those who are able to create value using platforms and those who are not has important consequences for the inequalities that emerge within a platform society. Professionalization has been shown to be a successful strategy on platforms such as Airbnb, and is taking an increasing share of the revenue made. As this paper has asked, however, what are the conditions of possibility for participating in professionalization?
The paper started by developing a multi-faceted and processual understanding of processes of professionalization. As a particular type of platform logic, these are partially engineered into the Airbnb platform and tied to Airbnb's business model, though they are also always shaped by differentially situated users embedded in local spatio-historical contexts. How this plays out in practice has been explored through three case studies based on qualitative research in Berlin. The experiences of hosts in Berlin show that professionalization occurs in different shapes and forms and is not tenable for all hosts. Some hosts may be willing and able to take part in all or most of these professionalization processes, while others might aspire to do so but fail.
In some cases, professionalization turns out to be a precarious process. For such hosts, professionalization results in insecure income streams while they face the risk of prosecution if regulations change. Hosts who bring capital are able to generate a more reliable income stream due their flexibility to switch to other business models and ameliorate risks by reinvesting their capital if needed. This type of privileged professionalization allows property owners to not just generate an income but also to accumulate wealth, for example by benefitting from value appreciation of the apartments they rent out. Other hosts turned out to be unable to navigate the risks and change their strategies, leading to disrupted professionalization. This might occur, as in Elli's case, when an apartment that belongs to a family member or which has personal value in other ways is being rented out. While capital has typically enabled actors in all kinds of sectors to professionalize, Johannes' story exemplified a novel type of datafied, platformed professionalization. By acting in a similar way to 'lean platforms' (Srnicek, 2017), he scaled up his business while limiting asset ownershipexcept for data, collected through the Airbnb API on his own platform. This allowed him to benefit from economies of scale, to cut costs, and to optimize his remaining operations. As such, Johannes could offer a service to asset owners that at the same time allowed owners to increase income streams from their assets.
Professionalization is not unique to Airbnb and has emerged on other platforms as well. In the case of the ride-hailing platform Uber, Pollio (2019) has documented how some drivers in Johannesburg are dependent on 'middling men' who manage car fleets and lease a car to drivers. Such new actors have emerged informally and independently of the platform, though Uber has started to establish formal relationships with 'partners' that can help drivers to lease a car as part of their 'Vehicle Solutions program' or 'Fleet Match.' 14 Such professionalization processes drastically affect the valorization capacities of those who are able to take part in them, resulting in increasing inequality between platform users.
The research for this paper was undertaken before Airbnb rentals were severely hit by the Covid-19 crisis. While presenting a huge problem for hosts and platform laborers, the crisis promises to be a crucial period for research on platformization. In particular, it will reveal a lot about the durability of 'hosting'. As Roelofsen and Minca (2021) question, who will be able to continue renting out on Airbnb after the Covid-19 crisis? And how does the crisis differentially affect 'home-sharers' and professionalized hosts? listing, thus they are not necessarily the legal owner of the space. Similarly, in the public debate on regulating STR, policies such as the 'One Host, One Home' are founded on the idea that it is clear who or what the host is. 11. See https://www.airbnb.com/d/software-partners. 12. This is consequential for both scholarly work and public debates on STR and Airbnb. In academic studies, the notion of 'host' has been used as a building block for concepts such as the 'multi-lister. ' The ambiguity of what a host is means that the concept of multi-lister might both be too broad (including apartments that are managed but not owned by the listed host) or too narrow (excluding owners of multiple properties that use multiple accounts to manage them, or who have a middleman that lists them on Airbnb). Similarly, in the public debate on regulating STR, policies such as the 'One Host, One Home' are founded on the idea that it is clear who the host is. 13. See https://www.airbnb.com/help/article/1188/what-is-host-assist-and-how-do-i-sign-up 14. See https://help.uber.com/driving-and-delivering/article/how-can-i-find-a-vehicle.