Chapter 4
Is the helper always happy? Platform-based domestic cleaning in Denmark

Platform work is often associated with the food couriers we see in the streets wearing pink, blue or orange clothes and bags, who deliver take away to customers, often by bike or scooter and at a rapid pace. However, this form of work has also made its entry into other areas of the labour market that are invisible compared to food delivery. One of these areas is domestic cleaning, that is, cleaning in private households. In Denmark several platform companies selling cleaning services through digital platforms have emerged in recent years. They differ from existing cleaning companies in that they do not see themselves as cleaning companies as such but merely as an online platform that connects cleaners with customers based on an automated matching algorithm. In this chapter I explore working environment challenges that may arise with such digitalized work arrangements. I am particularly interested in how platform companies in domestic cleaning use algorithms to manage workers, and how this form of control influences the workers’ occupational safety and health.

This chapter first describes important characteristics in the Danish cleaning industry as a background for understanding app-based domestic cleaning. I then present the case I have studied, namely a Danish domestic cleaning platform called Happy Helper, my methods and the theoretical and analytical frameworks used. I conclude that cleaners working through digital platforms are exposed to occupational safety and health risks. Some of these risks are similar to the risks in the traditional cleaning industry, but some of them are connected specifically to the business model of the platform company.

Cleaning is carried out in many different places and areas of society including private homes and private firms, as well as public workplaces. In general, the cleaning industry is characterized by many small, often local, cleaning companies (self-employed people and businesses with few employees) as well as large companies (Rasmussen et al., 2016: 28). As was emphasized in the chapter on domestic cleaning in Norway, it is relatively easy to establish a business in the cleaning industry in Denmark. Cleaners often have low or little formal education. While it is possible to work as an unskilled cleaner, formal education exists for work as a service assistant, house assistant, cleaning technician etc. According to a 2022 analysis from the employer organization Dansk Erhverv, less than two percent of those working in cleaning and window cleaning in Denmark in 2021 had formal cleaning education (Dansk Erhverv, 2022: 9). Around one third had primary school as their highest educational level and another 30 percent had vocational education. Cleaning staff with a formal cleaning education are more common in hospitals, nursing homes and day care institutions compared to other areas of the service sector (Dansk Erhverv, 2022: 9). Women and persons with a non-Danish background are overrepresented in the sector, in particular women with a non-Danish background (Dansk Erhverv, 2022: 6–9). Because the barriers to entry are low, this type of work functions as an entrance to the labour market for foreigners in Denmark (Refslund, 2014). Furthermore, part-time employment is widespread (Dansk Erhverv, 2022: 19) and the industry often lacks labour power (Dansk Erhverv, 2022).

When exploring the Danish cleaning industry, it is important to distinguish between cleaning in the public sector and in the private sector, where the conditions differ. In the public sector (state, regional and municipal workplaces), cleaning staff have traditionally been employed directly at the workplace and covered by collective agreements. These collective agreements regulate wages and working conditions for the cleaners. However, in recent years public authorities and municipalities have increasingly outsourced cleaning at public workplaces such as nursing homes and schools to private providers. When cleaning is outsourced, wages and working conditions may change or worsen because the cleaning staff are not covered by the collective agreements that were applicable when they were directly employed in the public sector. For instance, a recent Danish study compared a group of employees who switched from public to private employment due to government outsourcing with a similar group that did not experience outsourcing and found that outsourcing had a negative effect on employee income over time (Petersen et al., 2021).

Cleaning in the private sector involves cleaning at private companies and domestic cleaning. In private companies, cleaners can also be employed directly at the workplace, or the company can buy cleaning from a cleaning company. In the private sector in Denmark, collective agreement coverage is not 100 percent like in the public sector, but around 75 percent (Appel, 2020), and the private cleaning sector is known for a lower-than-average collective agreement coverage. Collective agreement coverage is difficult to measure precisely, but according to a 2013 study by Andersen and Felbo-Kolding, 59 percent of firms within the private cleaning sector were covered by collective agreements, and cleaning, together with agriculture and the hotel and restaurant industries, is the industry with the lowest number of collective agreements (Andersen and Felbo-Kolding, 2013: 121–123). Wages can be low for cleaners working without a collective agreement (Madsen, 2015; Refslund, 2015). The industry is also characterized by low union density (Ibsen et al., 2015) and widespread undeclared work (Korsby 2011), but the extent of it is difficult to estimate precisely. Undeclared work in the cleaning industry is more likely to take place within domestic cleaning and when subcontractors are used for cleaning tasks (Korsby, 2011: 17). Furthermore, there have been examples of illegal work in the industry (Korsby, 2011: 17). Therefore, cleaning, especially in the unregulated part of the private sector, seems to be characterized by insecure working conditions.

Over the last ten years, several domestic cleaning platforms have emerged in Denmark. Some are centred around cleaning in private homes (like Hilfr and Happy Helper) while others offer a wide range of on-location tasks such as craft work, moving assistance, dog walking, babysitting and cleaning (for instance MePLoy and Care.com). Most of these platform companies are Danish-owned and, in the beginning, they primarily used a freelancer/independent contractor model in which they do not consider themselves an employer but merely a mediator between a customer who wants to buy a cleaning service and a provider who offers to do the cleaning task. This business model differs from traditional cleaning companies in the industry because these new companies do not see themselves as cleaning companies as such but merely as online platforms that connect cleaners with customers based on an automated match.

The platform company Hilfr has received the most public attention in Denmark. This company was the first cleaning platform to enter into a collective agreement with a union in 2018, which resulted in a lot of media coverage. Now, Hilfr has a combination model wherein the “Hilfr” (which is the term used for the cleaning provider) starts as a freelancer, sets their own prices and is paid per cleaning task. When the Hilfr has worked 100 hours through the platform, they are offered employment directly at the company and are covered by a collective agreement, which mandates an hourly wage, a minimum income of at least 152 Danish kroner (around 20 Euros), savings for pension and holiday payments and the right to sick pay. The concept is called “Super Hilfr”. The Hilfr is automatically transferred to the employment model unless she or he actively choses to continue with the freelance model (Hilfr n.d.). According to a Danish study of Hilfr, the Super Hilfr concept quickly turned out to be attractive. In 2019, more than a third of all cleaning tasks at the platform were carried out by Super Hilfrs, and in 2022 two thirds. However, only 70 people were employed as Super Hilfrs in 2022 (Ilsøe and Larsen, 2022: 74–75), which indicates that the number of people working through the platform is limited. From this study, we also know that the platform owners believe that this model has been an asset for them, among other things because it has helped promote their brand as a socially responsible platform company. However, they find the business model difficult to maintain because they are competing with other platform companies that use the freelancer model and because the sector is characterized by a lot of undeclared work (Ilsøe and Larsen, 2022: 75). 

In this article, I analyse Happy Helper, which is the largest platform company in the Danish domestic cleaning market. In addition to domestic cleaning services, Happy Helper also offers move-out cleaning and cleaning of smaller commercial offices. According to the Happy Helper website, 4,500 helpers are associated with the platform (as of March 2024), and they operate in all major cities in Denmark.

Happy Helper was founded in 2015 by a group of Danish entrepreneurs who, inspired by Uber’s business model and other American platform companies, noticed that domestic cleaning in Denmark was often unregulated and informal and saw a market in domestic cleaning. In the interviews, management told me that they believed that, with their business model, they could formalize some of the informal work in the industry. They established a freelance model, wherein the company presented itself as a digital platform connecting customers who need cleaning services with independent providers offering cleaning services based on an automated match. They named the service providers “helpers”. The management at Happy Helper argues that the advantage for the helpers when they use the platform instead of operating independently is that they do not have to find customers on their own but can connect with customers through the platform. Furthermore, both the helper and the household goods are covered by insurance. Moreover, the helpers have the possibility to contact a live support team in case anything happens while working at a customer’s house.

According to the management, Happy Helper had a reasonable turnover in the first couple of years and experienced a demand from both customers and cleaners who were interested in the concept. However, the company struggled during the Covid-19 pandemic because, as the management said in our interview, “people became afraid of letting people into their homes” and fewer cleaning providers were interested in working in private homes. At one point the company closed the access for new helpers to ensure that there was enough work for existing helpers. After the pandemic, the company became more economically stable despite having only half its previous turnover. However, in April 2024, Happy Helper announced that it had gone bankrupt (Pedersen, 2024). Shortly after, another platform company, HandyHand, bought the company and has stated that it intends to carry on the concept (Weis, 2024). Handy Hand is a Danish-owned platform company that offers a wide range of domestic work tasks (lawn care, furniture collection, dog walking, painting etc.). 

The analysis is based on a qualitative research approach that combines interviews with the company management and cleaners working through the platform with existing data sources, such as information from the Happy Helper website, company reports, news articles etc. I also got access to an online community for “helpers” and the support team. Approximately 800 helpers are associated with this community and in the online forum I could follow questions, especially from new helpers, about how the Happy Helper business model works and learn more about the communication and sharing of information between the support team and the cleaners. Access to this online community has served as a supplement to the interviews. Table 4.1 shows an overview of the data.

Table 4.1 Overview of data

In the period of April to June 2023 I conducted a total of five interviews, two with management and three with helpers working through the platform. The management was approached by email and the helpers were approached through the online community where the management allowed me to look for interview subjects. I made several posts in the online community but was only able to reach three cleaners. The fact that I was not able to reach more cleaners is a limitation of the study, but the cleaners that did sign up had experience working through the platform. Furthermore, I was able to get supplementary data from the online community, where more helpers engaged in discussions that also gave me valuable and relevant information despite not being interviews.

All interviews were conducted online. The interviewed helpers were all located in Copenhagen and had experiences with cleaning there. The interviews with the cleaners followed a semi-structured interview guide; I asked about their background and motivation for working through the platform, their pay and working conditions, their knowledge about algorithmic management, health and safety issues and representation/union involvement. The interview guide was inspired by the work characteristics of digitalized platform work identified by Ropponen et al. (2019; see also Jesnes and Rasmussen, Chapter 5 on food delivery in Denmark and Norway). During the interviews, I tried to learn more about job insecurity, time pressure, isolation, competition, harassment and unfair treatment, among other topics. The interviews with the management representatives covered the company’s history and business model, including algorithmic management. The interview with the head of the support team was centered around the communication between the company and the cleaners.

All interviews were transcribed, carefully read through and subsequently coded. I combined open and closed coding. In the closed coding, I was inspired by the work characteristics mentioned by Ropponen et al. (2019), but I also allowed for codes to emerge from the material. For instance, in this process I learned that waiting time and competition are not central work characteristics of domestic cleaners compared to food delivery couriers, but isolation seems to be more pronounced. Furthermore, the contact and communication with both the support team and customers seem to matter more for domestic cleaners compared to food delivery couriers.

To ensure anonymity, especially for the cleaners, I have chosen a strategy in the analysis where I refer to what has been said in the interviews rather than using quotes, and when I use quotes, I do not indicate which interview they are from. 

I begin the analysis with a description of the business model at Happy Helper, including the process from the time a cleaner decides to offer cleaning through the website through the booking and finally to when the cleaning is completed. This business model is important for understanding the second part of the analysis which deals with the central work characteristics and the working environment challenges associated with app-based cleaning in private households. My focus is on the following work characteristics: job and income insecurity, time pressure and overtime, the physical work environment, unfair treatment and isolation. 

Most of the cleaners working through the platform are foreigners. When browsing through the profiles on the website, we see that most cleaners have English profile descriptions and non-Danish backgrounds, which is in line with the general trend in the cleaning industry described earlier. Like in app-based food delivery (see Jesnes and Rasmussen, Chapter 5), the barriers to entry in the field are quite low. People who want to work for the cleaning platform can register on the website with personal information, a Danish bank account and phone number and a clean criminal record. They watch an introduction video, complete a quiz and attend an online onboarding meeting with the company. According to the management, these meetings are used to check whether the cleaner has the skills to communicate with clients and is punctual. However, no previous experience with cleaning is needed. When the helper is approved by the company, they mark on a calendar when they are available to work. Because they are freelancers and not employees, they also set their own hourly rate, which is visible on their profile on the website/app and can be viewed by customers during the booking process. Although cleaners set their own prices, the company provides some guidance regarding reasonable pricing. According to the management, this guidance is designed to help cleaners set prices that are neither too high nor too low, as well as to help them be transparent with customers regarding what they can expect from the cleaner in terms of experience and quality. The platform has therefore developed a set of categories (“new helper”, “standard helper”, “premium helper” and “pro helper”), and when the helper decides on a certain hourly price, she or he is automatically placed in one of these categories, which is then also visible on their profile. A new helper has the lowest hourly price (around DKK 145/EUR 19) while a standard helper costs a little more (DKK 175/​EUR 23) and premium helper costs the most (DKK 235/​EUR 31). The pro helper category is the most expensive and is reserved for professional cleaning companies that bring their own cleaning supplies, which are also allowed to operate through the website. On the website, the company states that the new helper category is for new cleaners that do not have any reviews, which is why they have the lowest hourly rate. Customers are encouraged to help these cleaners by making a list of work tasks, because they are less experienced, and evaluating them after the cleaning. A standard helper “can do any cleaning task and is often the best choice for a domestic cleaning” while premium helpers are the ones with the most experience and the best ratings, which is why their hourly rates are higher and the customer can expect a higher quality cleaning (Happy Helper, n.d.). New cleaners cannot choose the premium category as a starting point (Happy Helper, n.d.). 

Beside the hourly price, customers are also charged a service fee that ranges from 15 to 35 percent of the hourly price depending on how often the customer uses the platform. If a customer uses the platform weekly, for example, the fee is 15 percent, but for a single cleaning, the fee is 35 percent (Happy Helper, n.d.). Happy Helper previously set the minimum rate at 120 Danish kroner (16 Euros). In 2020, however, the Danish Competition and Consumer Authority stated that minimum prices are not allowed when using the freelance model because it can limit the competition between freelancers (Konkurrence og Forbrugerstyrelsen, 2020). Following this decision, the company developed its new concept with categories.

When customers book a cleaning through Happy Helper, they enter the platform’s website and type in the time and date for the cleaning, the physical location and how many square metres need to be cleaned. Based on the area registered, an algorithm calculates a duration for the booking. This is similar to the Norwegian case of Vaskehjelp discussed in Chapter 3. The customer is then directed to a page where they can choose between available helpers and are provided information on their hourly prices, the helper category they belong to and reviews/ratings from previous cleaning jobs. The customer then chooses a helper, and a booking request is sent. The helper can accept or reject the requested booking. The booking must be accepted within 24 hours. If not, it will be offered to another cleaner. The guidelines published on the website state that the company is allowed to “deactivate” a cleaner’s profile if they do not respond to requests (Happy Helper, n.d.), although they do not specify how many requests they can decline before being deactivated. Furthermore, the website states that a cleaner’s profile can also be deactivated if they arrange cleaning with customers from the platform outside the platform (Happy Helper, n.d.).

When a booking is accepted, the helper becomes responsible for communicating with the customer, which can be done through the app. I have not been able to determine whether cleaners can communicate with customers by phone, but the guidelines on the website encourage cleaners to arrive in advance of their appointment to talk with the customer about their expectations for the cleaning, which suggests that most communication happens at the start of the appointment (Happy Helper, n.d.). On the day of the booking, just before the booking starts, the helper checks in on the app. This is important for the cleaner to get payment. The cleaner checks out when the cleaning is done. Compared to food delivery couriers (see Jesnes and Rasmussen, Chapter 5), cleaners generally have more contact with the platform/support team, which was confirmed in the interviews. For instance, if a cleaning takes more or less time than what was agreed upon, the cleaner must notify the support team, which must then correct the payment. Live support can also assist in the case of a dispute. According to the interviews, management also perceived the live support team as an important service for the helpers.

In terms of cleaning supplies, the customer must provide supplies for the helper. The website lists the cleaning supplies and equipment that customers should have (Happy Helper, n.d.). However, the interviewees explain that they often bring their own in case something is not available the house, or if they prefer using their own supplies (for instance special gloves).

All the cleaners interviewed worked part-time through the platform for between five and 20 hours per week. Two of them also worked other jobs to have a level of sufficient income. They explained that they worked for the platform because it was difficult to get a job in Denmark with their educational background. They all had a non-Danish background. I observed the same tendency on the website when looking through the profile descriptions, where several cleaners mentioned that they had completed higher education but did not currently work in their field of study. The three interviewees were not students, but many cleaners indicate on their profile description that they are students and I therefore assumed that some of the cleaners used this type of work as a part-time job in conjunction with their studies. 

The three interviewees explained that they had also chosen this type of work because they liked the flexibility the platforms offer. The appeal of flexibility is also noted in other studies, for instance in Jesnes and Oppegaard (2023), who studied food delivery couriers and Uber drivers in Norway. However, they were all aware of the insecurity associated with this type of job where they do not get paid for anything but the cleaning task and they all considered this a disadvantage.

All interviewed cleaners were categorized as premium helpers, meaning that they are more experienced cleaners, have better reviews and that they have set their hourly prices higher than other cleaners. They explained that they decide the prices on their own, but once they have decided on a price, they are automatically put into one of the categories, as one interviewee explained: “Well, I kind of fell into that category after I raised my hourly price” (female cleaner). This category is also shown on their profile on the website. One interviewee explained: “So, they don’t really decide my hourly price. They do try to influence it a lot though”. This is different from traditional cleaning companies that decide prices on their own.

All three interviewees explained that they do not decide the length of their working time on their own. The time spent cleaning for a particular customer depends on what the client has registered on the website, but the minimum amount of time for one client is 2.5 hours, which is equivalent to a home of approximately 60 square metres. Hence, when accepting bookings, cleaners normally accept the number of hours that is calculated based on what the client registers when they make a booking request. Sometimes they only have one booking a day and sometimes they accept more than one booking depending on the length of each booking.

Compared to the food couriers at Wolt, for instance (see Jesnes and Rasmussen, Chapter 5), platform workers in domestic cleaning seem to have more predictable working hours or working days because they mark when they are open to accept bookings in advance on the calendar in the mobile application. This procedure resembles the shifts at Foodora and Just Eat, but without the peak time. However, all of the cleaners experienced cancellations from time to time, so they could not always rely on having work when they expected to. Furthermore, they are not compensated if a client cancels an appointment. This is also different from the traditional cleaning industry where cleaning companies normally charge the customer a fee if they cancel an appointment or want to make changes close to the day that was originally agreed upon. One of the interviewees, who had been working through the platform for a while, told me that there used to be a cancellation fee, but that it was deemed a breach of Danish consumer law and the company therefore stopped collecting it. Her experience was that “basically there is more risk for the helpers now. We show up, ring the bell and no one opens” (female cleaner). The cleaners are not compensated for their transportation expenses, either, if they are on the way when the cancellation is made. All of them emphasized that they had regular customers who booked their services often. This gave them a sense of stability because they knew some of their working hours in advance. According to the Happy Helper management, there are incentives for customers to book cleanings on a regular basis because the service fee depends on how often they book cleaners through the website. If they order just one cleaning, they pay a certain service fee but when they make more bookings (for instance once a week), the fee lowers. The management claimed that this was done to ensure that cleaners can get more regular bookings.

In the case of the food couriers in Denmark and Norway discussed in one of the previous chapters, waiting time was a significant work characteristic, but in domestic cleaning, time pressure and working more than what is agreed upon are more of an issue (see also Huseby, Chapter 3 on cleaning in Norway). All the interviewees talked about time pressure, especially in the beginning when they were less experienced. The issue is that the algorithm that calculates the cleaning time does not consider anything but the size of the home (the number of square metres registered on the website), but the cleaning can take a longer time if the house is really messy, if there are two bathrooms, a large kitchen, windows that need cleaning and so on. The management at Happy Helper is aware of this challenge – “So, it’s always a little difficult to take the nuances into account” (Management interview) – but they claimed that they always encourage cleaners and customers to “enter into a dialogue about the tasks and priorities that exist so that it becomes a good experience for both parties” (Management interview) and if the helper can see that he or she cannot finish on time, they must ask the customer to pay for additional time. Management also encourages helpers to contact the live support team if they are engaged in this type of negotiation with the customer and they need assistance, and they do mediate in these matters from time to time. The helpers confirmed that they regularly use the support team.

Two of the interviewees shared that they negotiate with customers if they experience that there is a mismatch between the hours they have been booked for and what they are expected to clean and that they normally find a solution, whether it is being paid for more time or not cleaning everything that the client wanted. However, one interviewee explained that during her first years on the platform, she tried to go into these negotiations but found them too difficult and therefore she has made it a habit to check the size of the house in the Danish building and housing register and if the size of the house does not match what the customer has registered, then she will not accept the booking:

I would just end up going and talking to the person face to face, and they are like “oh but it is not so dirty. We keep it tidy”, and I think well, you have splattered and greased kitchen walls, but I can’t say these things right in your face, right? After a while trying to communicate with these people, I thought it was a waste of time, because it was very rarely I could make them see the light, so you know, when you get a booking, you get a notification on the app and either have to accept or reject it. I would get a booking. I would see the address and look it up in the system and ok, something doesn’t match here, and I would just reject it. I wouldn’t even go into that back and forth because it is fruitless. (female cleaner)

A related issue is that the cleaners depend on good reviews because new clients will read these reviews on the website when they book a helper. One interviewee explained: “You have the massive review pressure, because if you don’t get a good review, you don’t get more jobs. Even though you don’t have enough time, because you have to get a good review” (female cleaner). She is often booked for too few hours because the algorithm’s calculation is too simple, but because she finds it difficult to negotiate with customers and because she is dependent on good reviews to get more bookings, she often works more than what is agreed upon.

All the cleaners I interviewed explained that the work is physically demanding. One interviewee had back problems because the vacuum cleaners are not always adapted to his size and another described that she is not able to work the same hours as she did in the beginning because of physical pain. One had gotten chemicals on his hands because the gloves he used would often break. 

Even though the customers must provide cleaners with the proper cleaning supplies and equipment, the interviewees reported a lack of proper cleaning supplies from time to time, which posed a risk to their occupational health and safety. This was also the case for the Norwegian platform Vaskehjelp discussed in the previous chapter (Huseby, Chapter 3). The interviewees from Happy Helper told me that sometimes customers do not have supplies, like oven cleaner or limescale cleaner or even a vacuum cleaner. One interviewee described how she used a broom and wet wipes to clean a floor because the vacuum cleaner was broken. Not only do incidents like these worsen the physical work environment, they can also lead to disputes with customers. One interviewee told us that he normally offers to order the cleaning supplies through a grocery delivery app if he learns that the customer does not have the right supplies, but one time a customer refused and asked him to clean with the supplies that were in the house. Afterwards the client complained about the cleaning and wanted a refund. The interviewee was asked by the support team to document that he had done the cleaning properly but he did not want to use time documenting his work so he told the support team that they could refund the money to the client. The company ended up paying some of the expenses and the interviewee thought that this was a way of showing their support for him. While this interviewee did not say it directly, this example can be interpreted as a case of unfair treatment.

Two of the cleaners that I interviewed found the work to be lonely because the cleaning tasks are normally performed alone. Sometimes customers can book two cleaners at the same time, but because they are so busy, they often do not interact. This interviewee had experience from app-based food delivery, for which he had an employment contract, and compared the two jobs; he thought that the relationship between food couriers was more collegial. He wished for another job with more interaction. Another cleaner emphasized a more generalized feeling of being invisible:

You are invisible not only for the client, but also for society. I am in this unregulated grey area. It has also eroded my self-esteem, like mentally it is awful. Also, because you are literally not interacting with anyone. You are just scrubbing and just. Everybody is at work. You don’t meet anyone. (female cleaner)

On a related note, the interviewees all expressed concern about working alone in other people’s private homes. Two of them said that they preferred business customers because they found it intimidating to be working in people’s homes, and the female interviewee felt especially vulnerable because of her gender. A similar point is made in a recently published Danish study about working environment challenges for young people working for digital platforms (Nielsen et al., 2024). Nielsen et al. emphasize that cleaners feel insecure about working in private homes because platform companies do not check who the customers are and customers can therefore hide behind digital anonymity (Nielsen et al., 2024: 66–67). The point about females feeling vulnerable because of their gender is also present in their study, where the female interviewees link their insecurity to their position as females working in low status jobs (Nielsen et al., 2024: 67).

The management was aware of the challenge with isolation, and they explained that they have established different online communities for helpers, where they can talk with each other as well as with the support team because they know that the work can be lonely and there is a need for connection. One example is the online community on Facebook, which I got access to.

One of the interviewees also talked about union membership and representation. He had been to meetings organized by a union because he also worked as a food courier, but he believes that it is difficult to mobilize and organize domestic cleaning platform workers because they perform the work alone and have no colleagues to talk to. A study on app-based domestic cleaning in Germany had a similar finding (Niebler and Animento, 2023). This perception was confirmed by the management at Happy Helper, who also reflected on the lack of unionization and mobilization in the sector. The manager said he wouldn’t mind if helpers started mobilizing and making more demands because he would see it as a sign that this type of work is valuable, and he would like to enter into such a dialogue. However, this has not happened in his experience because the group is fragmented, cleaners rarely speak to each other and nobody dares to speak up. They have tried to arrange meetings where helpers could speak more freely but it has proved to be difficult. 

In this chapter, I have investigated working environment challenges for cleaners working through the cleaning platform Happy Helper, which is one of the largest platform companies in the domestic cleaning sector in Denmark. In addition, I have focused on understanding how the company uses algorithms to manage workers and how this affect their occupational health and safety.

Like the chapter on food delivery couriers in Denmark and Norway (Jesnes and Rasmussen, Chapter 5), I have used the analytical framework from Ropponen et al. (2019) to guide my empirical analysis. Ropponen et al. have identified several job characteristics that apply to workers in digital work arrangements that can threaten their occupational health and safety, and these characteristics seemed like a fruitful point of departure for understanding my empirical data in the cleaning case. But – like in the food delivery analysis – I have also been aware of the empirical specificities of my particular case and data, which can add to the framework of Ropponen et al. (2019). 

First, I found that job and income insecurity were issues for the cleaners because of the freelancer model, where workers are paid per task. As in the food delivery case, the cleaners are torn between the flexibility of the work, which they like, and the insecurity in terms of income. Compared to workers in the food delivery industry, workers in the cleaning industry may be even more exposed to low earnings because customers can cancel their bookings, in which case they are not compensated; when bookings are normally for a number of hours, the loss of income is significant.

Second, I found that waiting time was not a significant work characteristic of domestic cleaners, but time pressure and working overtime were, especially for new cleaners. Furthermore, time pressure and working overtime seem to be connected to the way the company uses algorithms to manage the workers and to the functioning of the business model, in which customer reviews are central. The algorithm calculates a cleaning time, which is often too short, and cleaners find it difficult to negotiate with the customers about adding time because of the difference in their status; thus, cleaners risk getting trapped in a vicious circle whereby they continue to work more than what is agreed upon because they are dependent on good reviews in order to get more bookings. 

Third, I found that cleaning is also a physically demanding job, which can affect the occupational health and safety of workers. In the chapter on the Norwegian cleaning platform (Huseby, Chapter 3), the issue was mainly the danger of the supplies used, which was not brought up as an issue in this case. Here, it was more about physical strain and pain as the result of using cleaning equipment that was not adapted and about disputes when the cleaning was not done properly because the customer didn’t have the right equipment or supplies. In this regard, I also noticed that the cleaners did not bother arguing with customers who were not satisfied with the cleaning. In such cases, cleaners may lose income if customers do not want to pay for a cleaning that they are not satisfied with. For several of the interviewees, there seemed to be an awareness of their low status that made them hesitant to make demands on the customer (to have the right equipment and supplies or to negotiate about the time needed to do the job properly), keeping them in an unhealthy work situation.

Lastly, I found that competition was not an issue, as Ropponen et al. (2019) found, but isolation was. I found that the interviewees felt isolated and insecure about working alone in people’s private homes and some had a more generalized feeling of being invisible. Furthermore, cleaners as a group do not seem to be able to mobilize in the same way as food couriers which keeps them in a position where they are not able to secure better working conditions.

All in all, my empirical data shows that cleaners working in app-based domestic cleaning in Denmark are exposed to significant safety and health challenges. The work is economically insecure, physically demanding and can be mentally stressful, like we also saw in the case of food couriers (Jesnes and Rasmussen, Chapter 5), but isolation and a feeling of being insecure and invisible were more prevalent among cleaners. Some of the challenges that cleaners in app-based domestic cleaning face are similar to those faced by cleaners in the traditional cleaning industry but others appear to be connected to the fact that the service is managed through a digital platform, especially the vicious circle described before wherein cleaners continue to work under inferior conditions because of the way the algorithm and the review system function.

Andersen, S. K., & Felbo-Kolding, J. (2013). Danske virksomheders brug af østeuropæisk arbejdskraft. FAOS.

Appel, M. (2020). I Danmark er de fleste dækket af overenskomst. DA Analyse. Accessed from https://www.da.dk/politik-og-analyser/overenskomst-og-arbejdsret/2018/hoej-overenskomstdaekning-i-danmark/, 27 May 2024.

Dansk Erhverv. (2022). Tendenser og udvikling: Arbejdskraft i rengørings- og vinduespoleringsbranchen. Analyse. Dansk Erhverv. Accessed from https://www.danskerhverv.dk/siteassets/mediafolder/dokumenter/01-analyser/analysenotater-2022/arbejdskraftanalyse-i-rengorings--og-vinduespoleringsbranchen.pdf, 27 May 2024.

Jesnes, K., & Oppegaard, S. M. N. (2023). Plattformmediert gigarbeid i Norge. Fleksibilitet, uforutsigbarhet og ulikhet. In T. Fløtten, H. C. Kavli, & S. C. Trygstad (Eds.) Ulikhetens drivere og dilemmaer, pp. 137–153. Universitetsforlaget. DOI: https://doi.org/10.18261/9788215065403-23-08.

Konkurrence og Forbrugerstyrelsen. (2020). Minimumspris på Happy Helpers platform. Konkurrencerådsafgørelse den 26. august 2020. Konkurrence og Forburgerstyrelsen. Accessed from https://www.kfst.dk/media/vi2gzmje/20200826-minimumspris-p%C3%A5-happy-helpers-platform.pdf, 27 May 2024.

Korsby, T. M. (2011). Menneskehandel i rengøringsbranchen? En kvalitativ undersøgelse af migrations- og arbejdsvilkårene for en gruppe migrantarbejdere i Danmark. Center mod menneskehandel.

Larsen, T. P., & Mailand, M. (2014). Bargaining for social rights – national report Denmark. FAOS Research Note 142. FAOS.

Hilfr (No date). Hilfr. Privat renhøringshjælp, nemt, godt og billigt. Accessed from https://hilfr.dk/, 27 May 2024.

Ibsen, C. L., Due, J., & Madsen, J. S. (2015). Fald i organisationsgrad... igen, igen. FAOS. Accessed from https://faos.ku.dk/pdf/temasider/Fald_i_organisatinsgrad_igen__igen.pdf/Fald_i_organisatinsgrad_igen__igen.pdf/Fald_i_organisatinsgrad_igen__igen.pdf, 27 May 2024.

Ilsøe, A., & Larsen, T.P (2022). Digitale arbejdsmarkeder. Fra jobs til opgaver. DJØF Forlag.

Madsen, P. K. (2015). Labour market flexibility in the Danish service sector, same old, same old, but more different. In W. Eichhorst & P. Marx (Eds.) Non-standard employment in post-industrial labour markets. An occupational perspective, pp. 221–244. Edward Elgar Publishing.

Niebler, V., & Animento, S. (2023). Organising fragmented labour: The case of migrant workers at Helpling in Berlin. The Economic and Labour Relations Review, 34(4), 689–706. DOI: 10.1017/elr.2023.46.

Nielsen, M. L., Yung Nielsen, L., Christiansen, S. W., Christiansen, S. W., Clausen, T., & Dyreborg, J. (2024). Risiko, arbejdsmiljø og krænkende handlinger blandt unge med platformsmedieret arbejde. (RADAR 2). Det Nationale Forskningscenter for Arbejdsmiljø. Accessed from https://nfa.elsevierpure.com/ws/portalfiles/portal/43519804/Nielsen_ML_Risiko_arbejdsmilj_og_kr_nkende_handlinger_RADAR-II_2024.pdf, 27 May 2024.

Pedersen, E. S. (2024). 623 aktionærer taber alt i Happy Helper konkurs. Børsen. Accessed from https://borsen.dk/nyheder/investor/623-aktionaerer-taber-alt-i-happy-helper-konkurs, 18 May 2024.

Petersen, O. H., Andersen, L. B., Bhatti, Y., & Houlberg, K. (2022). Competition, ownership, and the impact of government outsourcing on employees. Journal of Public Administration Research and Theory, 32(2), 398–415. DOI: https://doi.org/10.1093/jopart/muab032.

Refslund, B. (2014). Intra-European labour migration and deteriorating employment relations in Danish cleaning and agriculture: Industrial relations under pressure from EU8/2 labour inflows? Economic and Industrial Democracy. DOI: https://doi.org/10.1177/0143831X14550421.

Refslund, B. (2015). Adjusting the Danish industrial relations system after Laval: Recalibration rather than erosion. Transfer: European Review of Labour and Research, 21(2), 247–251. DOI: https://doi.org/10.1177/1024258915575249.

Rasmussen, S., Refslund, B., Sørensen, O. H., & Larsen, T. (2016). Reducing precarious work in Denmark through social dialogue: The case of Denmark. Aalborg University. Accessed from https://vbn.aau.dk/ws/portalfiles/portal/244600846/Denmark_final_report.pdf, 18 May 2024.

Ropponen, A., Hakanen, J. J., Hasu, M., & Seppänen, L. (2019). Workers’ health, wellbeing, and safety in the digitalizing platform economy. In S. Poutanen, A. Kovalainen, & P. Rouvinen (Eds.) Digital work and the platform economy: Understanding tasks, skills and capabilities in the new era, pp. 56–73. Routledge. DOI: 10.4324/9780429467929-4.

Weis, E. (2024). Happy Helper går konkurs – men lever videre med nye ejere. Berlingske. Accessed from https://finans.dk/erhverv/ECE17029302/happy-helper-gaar-konkurs-men-lever-videre-med-nye-ejere/, 18 May 2024.