Yannick Fondeur^[1]

Digital services in France have long been dominated, and still remain largely so today, by a very specific type of company operating as a labour market intermediary (Fondeur, 2013)^[2]. Called SSII (‘Sociétés de Services en Ingénierie Informatique’), or ESN (‘Entreprises de Services du Numérique’), these ‘IT Services Companies’ behave like employment agencies, invoicing their clients for the provision of the digital professionals they employ. As the linchpin of the French IT employment system since the 1970s, they have played a decisive role in its employment dynamics and have thus long been an obligatory route for young graduates, a form of entrance door into IT employment (Fondeur and Sauviat, 2003). They also acted as intermediaries for clients: once contracted by the latter, they subcontract service providers, including an increasing number of freelancers, to carry out assignments.

In recent years, new players have been trying to challenge the key position of the SSIIs in the digital services market, a large part of which is characterised by a very low capital intensity (mainly intellectual services) and by not concentrating the workforce in a single location (working at the client’s premises, or from any location). These actors are of two types. Firstly, as in many other areas, digital platforms present themselves as vectors of ‘disruption’, offering digital workers the opportunity to work through them as freelancers rather than as part of the IT services industry. Secondly —and this is the subject of this contribution—, the sector is marked by the multiplication of forms of self-managed organisations bringing together autonomous professionals, under various statuses, claiming to work and do business differently.

This desire to be different is part of the general context of a new demand for autonomy on the part of qualified workers in the service sector (Bureau and Corsani, 2014). In the digital professions, it is aimed more specifically at three targets, with a different intensity depending on the case: traditional wage employment, SSIIs, and freelancing platforms. It is the construction of this ‘alternative’ that I am trying to grasp here, through an in-depth but provisional analysis (see box) of a group of freelancers offering digital services essentially oriented towards the Web. Compared to other types of alternative companies in the sector which I have had the opportunity to study more superficially (Fondeur, 2020), this collective has the relatively rare characteristic of displaying the ambition to ‘scale up’ and thus compete with the large SSIIs and freelancing platforms. Born from the meeting of two developers in a Parisian coworking space and based on the founding principle of freedom at work, it has grown through networks of relations to include several hundred members spread throughout France (as well as in a few locations abroad). In doing so, the collective found itself confronted with the difficulty of transforming its very organic and continuously deliberate organisation into a genuine mutualised platform with a governance adapted to its new dimension.

Box: Research field and methodology

1. From two to eight developers in a coworking space

The story of this collective begins in the 19th district of Paris, in a workspace emblematic of the first age of coworking, where this ‘movement’ was largely part of the digital economy and claimed the notion of ‘community’. In February 2014, two developers met there. Both had studied engineering, had rather generalist profiles in web technologies, and just had failed experiences in creating start-ups, which led them to launch themselves as independent service providers. One of them, whom we will call Aurélien, says:

I worked in a start-up in Brazil, which I tried to replicate in France: I went broke, it didn’t work. […] After that, I started to work as an independent service provider […]. I had been creating websites on my own for three months, so as make a bit of money. And he [laughing]… he came to see me and said: ‘I hear you have too much work, and I’m looking for work, so…’. […] That’s how we met. So we knew each other a little bit, but that’s how we started working together. (Interview, October 2017)

Aurélien, barely 26 at the time, had already ‘done quite a bit of travelling’ working abroad, with varied experiences outside IT. His future partner, whom we shall call Sébastien, is thirty-one years old and has a more traditional engineering background, with substantial experience in SSIIs. He had a particular interest in alternative business models:

He had studied a lot of new forms of company organisation, and he felt that there was a tendency towards the horizontal company, which was a model he had studied. Particularly models like Gore-Tex, Valve, which are companies that have…, that have basically done away with the very idea of a manager, and are saying to people, ‘Let’s divide you into teams that make sense, and let’s all move forward together in a way that..’. They organise themselves as they go along without anyone thinking about the organisation chart and trying to make it work. (Aurélien, interview, October 2017)

From the very beginning, these themes are a central topic of discussion for the two developers.

The two basic ingredients of the future collective are already present at this stage: the need to network in order to work as freelancers and the desire to work and do business differently.

In 2014-2015, Blein (2016) conducted an ethnographic study in this coworking space, which he named ‘Potemkine’ to anonymise it, and met our developers there a few months after their meeting. His contribution is doubly interesting: for his analysis of the role played by these spaces in the constitution of freelancers’ social networks, which partly based on the observation of the beginnings of the collective that interests us here, but also for the particular history of this collective, before our own observation. Here is how he describes the small group:

They are now a group of eight web developers who mostly work in the coworking space. [Sébastien] and [Aurélien]^[3] then decided to formalise the network by creating [during the summer of 2015] an SAS^[4] so that Developnet could serve as a showcase for them. The aim is for each of the associated developers to retain an independent status, but to use Developnet to contract with clients, with the SAS then benefiting from a higher turnover to gain credibility. (op. cit., p. 155)

Aurélien is described by Klein as the ‘network maker’. He lives next door to Potemkin, of which he is an almost daily user. ‘He therefore knows most of the regular coworkers well and keeps in touch with former coworkers with whom he is friends, with certain evenings regularly taking place between current and former coworkers’. He also takes part in training sessions organised by Potemkine.

Blein’s thesis is that such places are not only flexible workspaces, but also allow freelancers to access the social networks necessary for their activity. Thus, Potemkin, where information and reputations circulate rapidly, serves as an intermediary in exchanges between coworkers, creating conditions of trust for transactions. These transactions may be commercial or not; in the latter case, they are part of a gift/counter-gift logic (Mauss, 1925) and can be direct or mobilise a coworker’s external network. Blein crosses these two dimensions to construct a typology distinguishing four cases: direct co-contracting when coworkers team up on a common project for which they are paid; direct participation when a coworker freely helps another one on a project; indirect participation when a coworker freely recommends another one to his or her network, or sends information about an opportunity (or even both simultaneously); indirect co-contracting when this last type of action is not free of charge and gives rise to a commission (as with a business introducer).

During his stay at Potemkine, Blein observed mainly non-market exchanges between coworkers. And this is also the case for our freelance collective: ‘At Developnet, indirect participation […] in the coworking space was decisive to launch the two founders’ business’ (op. cit., p. 168). He illustrates his analysis with the following interview extract with Aurélien:

We’ve been recommended by people from [Potemkine] to other people. It really depends to what extent you think it comes from [Potemkine] or not. When it’s someone you met through someone from [Potemkine] who recommends you, is it still [Potemkine] or is it… because if you consider that it’s still [Potemkine], almost 100% of our business, well, not almost… maybe not 100%, more like 80%, but almost all of it comes from [Potemkine], directly or indirectly. Even if it’s people we just bumped into here one night.

Aurélien and Sébastien thus mobilised the extended Potemkine network according to a principle close to that of the ‘strength of weak ties’ (Granovetter, 1974) to harvest work opportunities that they gradually co-processed with other developers, replicating the mechanism at the origin of their professional encounter. Success soon led them to formalise their co-processing network in a company comprising recurrent working partners.

2. Extending the network: a well understood interest

Their collective had changed a lot when I met them in October 2017: ‘We have added more and more people; today there are a little more than 120 of us’, including ‘eight associates in all’, said Aurélien during our first meeting. The ‘historic’ members, and in particular Aurélien, still cowork on a regular basis at Potemkin. In fact, our meeting took place just a stone’s throw from there. The Potemkine network still remained a collective resource for a long time, even though it was less and less present. In particular, developers regularly found opportunities in the Potemkin coworkers’ Facebook group feed, even after the venue closed in spring 2018. But as the collective grew, it became itself a commercial force via the network of each of its members. This is their primary interest: this pooling of their respective networks gives them access to more projects.

There is also a second dimension: by becoming its own entity, the collective itself becomes visible on the market and credible to larger companies. This strategy has been present since the creation of the SAS, as noted by Blein (2016), but the development of the ‘brand’ is now the main concern of the small group of associates, who are multiplying initiatives to develop the collective’s reputation. As a result, a growing number of requests arrive via the website’s contact form, landing directly in a channel of the instant messaging system shared by —and open to— all members^[5].

Besides the quantity of opportunities, it is their quality that is targeted by the members of the collective. This is why the shared objective is being able to ‘charge more’, within a market where trust is the key and where assignment opportunities are therefore not subject to much competition; a configuration that the members of the collective readily contrast with access to the market through freelancing platforms ‘which drive prices down’.

Compared to the latter, the collective also enables its members to form teams, united by bonds of trust built up through recurring collaborations, and thus to respond to client needs beyond the framework of a one-off task. This ability to quickly build teams allows them to approach the market of the traditional players in digital services for companies (agencies and SSIIs), a market whose access for freelancers is usually conditioned by these intermediaries, who act as ‘assemblers’ of skills and keep the hand on the commercial negotiation with the final client.

Little by little, this logic of building freelancer teams led the collective to diversify and to welcome digital professions other than those of development. The ‘devs’, who are very much in demand, still form the core of the collective, but complementary skills are often necessary in the projects they work on, whether it be, for example, graphic design, UI/UX design, or digital marketing. The collective can thus position itself as a single service provider, like a generalist web agency. Thanks to this diversification of profiles, it is also gradually becoming more feminine.

3. Staying close

While strongly encouraging this extension of the network, the SAS partners, which now form its bridgehead, are keen to maintain its human dimension by relying on physical and social proximity both in the recruitment of new members and in the operation of the collective. It is as if they were deploying on a larger scale the very organic model that was initially used at Potemkine.

As in the coworking space, the collective seeks to develop mutual aid and informal communication between its members. In the absence of a shared workplace, this is achieved through an instant messaging system^[6] in which a large number of channels are created to exchange and coordinate common projects, but also to allow all sorts of discussions as opportunities to socialise between members. Several dozen channels are dedicated to technologies or areas of expertise (WordPress, SEO, UX, etc.), to exchanges of useful information on the conditions of freelance activities (change of status, social protection, etc.), or to all sorts of discussions outside of work in an overtly joking atmosphere (the most active channel is aptly titled ‘#h-mess / Chatting and talking crap’). Between 2018 and 2019^[7] a total of more than 2,000 messages were exchanged on average each week, with a lot of smiley faces and a very friendly and benevolent tone. One of the favourite activities of the freelancers in the collective is to discuss everything and anything at any time of the day or night, which shows that this internal messaging system is an important place of sociability within this more and more physically disperse group. The group’s socio-demographic homogeneity, being concentrated in the 25-35 age group and in a population with higher education, naturally favours these remote exchanges.

Such homogeneity is partly linked to the specific professional field of the digital professions, which already brings together a population that largely corresponds to this profile, but it is also explained by the fact that entry into the collective is based on co-optation. The main recruitment channel consists of regular events where members are encouraged to bring in freelancers likely to join the collective. During my first meeting in October 2017 with three members (Aurélien and two other freelancers, whom we will call Laurent and Patrice here), this conviviality was presented as a very strong differentiating element as compared to the platforms:

Laurent: So, I would say that the real difference between Hopwork^[8] and us is that we are built first and foremost on social exchange; it’s really about…, it’s about people.

Patrice: Community.

Laurent: That’s it, community, that’s it, it’s primarily an informal community. They first built a tool, and today they’re trying to build… a community. […] They organise afterworks, things like that, but, for our part, we actually built ourselves around social events. I mean, the recruitment tool, our main recruitment tool, is the [cool drinks]^[9] : […] we meet on Thursday evenings for a beer, we discuss everything and anything… Truly anything! Because it’s not the time to talk about work anymore. And we recruit people based on these things, on feelings, on people, rather than on CVs.

Since 2016, a formal collective membership has been conditional on validation by three people who are already members. In concrete terms, a fairly classic case is that the potential candidate is invited to a ‘cool drink’ by one of the members, or that he or she goes there on his or her own after having been introduced via Meetup or the collective’s public account on Facebook. The few words exchanged over a beer with the members present may then be enough to enter the collective, provided that the latter find the applicant ‘cool’ (the word is systematically used when describing ‘good’ applicants for entry). One of the guarantor members is designated as his or her mentor, a function described in an internal ‘manual’:

[The collective] is not intended to be a company that brings in business for lonely self-employed people. We are calling for a close-knit collective that shares much more than services by rethinking the business world and its governance. Mutual aid, collective times (e.g., coworking days, [cool drinks], barbecues, trips, chat) are all opportunities to take part in the collective and to contribute to it. Your mentor should make sure that you are aware of these events and that you feel welcome.

Coworking days are organised several times a week and are announced to the whole collective, with a little phrase ritualised by Aurélien: ‘Feel welcome’. Several times a year, ‘colivings’ also take place, during which a few members (10 to 15 in general, but participation is open) share the same place for several days and alternate work sessions, social activities, and reflections on the future of the collective. Finally, once a year, an anniversary event of the same type is organised, but on a larger scale (several dozen members gather there). Everything is done to encourage meetings and exchanges between members outside of direct work collaborations, so as to create the closeness and conviviality considered essential to the functioning of the collective.

4. A common desire to work differently

The nature of these shared times is already in itself the expression of a desire to work differently. But, beyond that, the collective puts forward a complete alternative to the salaried work model as well as a new way of ‘doing business’. These principles have been formalised in two documents, both of which are available on the collective’s website (and as such are instruments of a claimed identity).

The ‘manifesto’ was drawn up in September 2015 at the collective’s very first coliving. This type of declarative public document is very common in the IT world: among the best known examples are the GNU Manifesto of 1985 and the Manifesto for Agile Software Development of 2001. More specifically, the drafting of a manifesto has become a classic for digital companies claiming an identity based on free work. However, the tone of the text of our freelance collective has very particular characteristics that are all the more important to note, that this document, drawn up at the very beginning of its history, when the collective only had about fifteen members, would subsequently constitute both a particularly effective recruitment tool for freelancers in the digital sector and a symbol of commitment for new members, who would have to sign it. It should also be noted that this text, which is shared via an online document that can be edited by all members (so that they can add their digital signature, but also suggest modifications), has only been marginally modified as the collective grew. It is probably one of the only things that is still entirely agreed upon today.

The text sets out eight ‘common values’, from which two main semantic fields emerge clearly on analysis. The first emphasises the ‘human’ character of the collective and of the work (‘our job is human before it is technical’). The word itself is repeated three times (in the form of a noun or adjective), and it is part of a very marked lexicon, whether with regard to the members (‘benevolent’, ‘supportive’, ‘good humour’, ‘joyful energy’, ‘mutual understanding’, ‘passions’, ‘extravagances’, etc.) or to the clients (‘friends’, ‘trust’, ‘listening’, etc.). The second register is that of freedom and self-realisation: the adjective ‘free’ is repeated twice, while the expression ‘we aspire to’ is repeated three times. What is striking about the passages in question is the equally free form of expression. The very first point of the manifesto sets the tone: ‘We aspire to have fun at work; we refuse to take part in any activity that would make us unhappy; the fulfilment of the members of the network is as important as the search for income’. Another passage points to wage employment as the enemy of this total freedom: ‘We want to feel free; wage employment is a form of servitude for us; [the collective] makes the members of the network autonomous and masters of their destiny’.

The other document that the collective puts forward on its website, the ‘vision’, explains in part why the ‘manifesto’ refers to wage employment as a negative reference to the sought-after “other way”. This second document, written in 2017^[10], proposes to place the collective in a broader and more ambitious framework: ‘beyond the values described in our manifesto, we aspire to a professional life in harmony with our vision of the world’; ‘this text presents our vision of the 21st century company’. It is clear from reading it that behind the virulent rejection of wage employment, what is targeted is a representation of the functioning of the traditional company combining, on the one hand, heteronomy and distance from decision-making centres and, on the other hand, disconnection from the aspirations of individuals and real work. A few expressions and passages illustrate this: ‘hierarchical administration’, ‘giant organisations’, ‘transforming autonomous human beings into pawns in the service of issues that have no meaning for them’, ‘an ever-widening gap between the management methods applied and the aspirations of the individuals who work there’, ‘by enlarging, centralising, industrialising, and disconnecting the task from its effect, the company has gradually cut itself off from what gave it meaning: it has become dehumanised’, ‘the obstacles of a conservative hierarchy’, etc.

Another dimension appears in the course of a sentence when the fact of ‘being able to approach and move away from the network in a fluid manner’ is mentioned. We understand that this freedom of commitment in relation to the organisation is opposed to a representation of the ‘wage-earner’ as locked into the company. ‘Not having a binding link obliges to build the collective on a common will to move forward together’, as it is specified a little further on. Observation of the functioning of the collective shows that this is not a binary freedom (entry/exit) but rather the possibility of situating on a very broad continuum both one’s involvement in the life of the collective and the insertion of one’s freelance activity in such framework.

Wage employment as a status is not the enemy when it is externalised and freed from subordination: a significant proportion of the collective members belong in parallel to Employment and Activity Cooperatives (Coopaname and Smart in particular), which enable them to ‘hack’ wage employment in such a way as to reconcile, as Demoustier (2006) summarises, ‘the autonomy of individual entrepreneurship with the dynamics and collective protection of wage employment’^[11]. The recourse to umbrella companies also exists, but is rarer.

The ‘manifesto’ and the ‘vision’ are therefore statements of a different way of working and doing business, around which the collective intends to build its identity. Direct observation of its day-to-day operations clearly shows that these are not merely declarative elements, and that they largely correspond to the aspirations and practices of its members. However, it also shows that these two documents do not exhaust all the dimensions of this desire to be different. In particular, the analysis of the project offers circulated within the collective and of the members’ reactions and comments^[12] shows the central importance of two very concrete criteria for the freelancers.

The first, which comes up almost systematically, is the possibility of working remotely. Full remote working is particularly valued, and often a prerequisite for accepting the assignment, especially for the growing number of non-Parisian developers. For freelancers based near the client, spending one day a week in the client’s premises is well accepted, but the assignment must be really interesting to go beyond that. Being free to choose where to work is therefore central to ‘the different way’. Conversely, the obligation to work at the employer’s premises is implicitly associated with a characteristic of ‘wage employment’.

The interest of the project is the second criterion that emerges from the discussions. The nature of the project and, for the developers, the technologies used come into play here. On the first dimension, the question of meaning is of course important: in particular, the collective has a large number of clients belonging to the social and solidarity economy (and often declares itself to belong to this field), but this seems to have as much to do with a market positioning as with an intrinsic vocation. More modestly, it is above all the ‘coolness’ of the project, but also of the client, that appears to be the determining factor. On the second dimension, taking pleasure in one’s work by using modern tools or, better still, having the opportunity to learn and experiment with new technologies (also with a view to maintaining one’s employability) is an absolutely central element for developers.

All this does not exclude the consideration of more traditional criteria in the choice of assignments. This is of course the case of remuneration, expressed as an average daily rate. But also, and above all, the duration and type of contract, with long assignments paid on a time basis being highly sought after, both for reasons of economic stability and for the possibility of an extended commitment to a project and to a client. These are elements that could be compared to ‘wage employment’ and which show that, though the desire to do things differently focuses on freedom at work, this does not exclude the search for certain dimensions of traditional employment.

5. A parsimonious mutualisation

In July 2014, when they were just starting to exchange ideas about the new form of ‘distributed’ business they were dreaming of, Sébastien wrote in an email to Aurélien:

The idea is at least to avoid employment contracts, so that all members are independent, and are therefore paid according to their activity. So we apply a kind of subsidiarity principle, letting everyone live their own life, but collectively managing everything that is better managed that way.

But what to share then? Sébastien’s message gives a few ideas: ‘a shared sales force, so that each person can concentrate on his or her job’, ‘simplified administrative management’, ‘guarantees for the client, so that the assignment can be carried out even if the developer is ineffectual, ‘financial security, for example through a personal reserve account’, ‘financial insurance, so as to be able to take on large contracts while limiting the risks’, and ‘surely a lot of things I haven’t thought of yet…’. All these elements are specific to a work activity carried out within an entrepreneurial framework, Sébastien specifying that ‘it is a question of providing the self-employed with everything that an SSII could provide if they were employees’, with the exception of the assurance of being paid between two projects, as is the case in SSIIs between two client contracts (people “on the bench”).

More generally, there is no question of building a mutualised protection against the hazards of the activity. And the position remains the same when, five and a half years later, this type of problem is addressed during the ‘agora call’, a regular video conference that serves as a public forum within the collective. Eric, a developer from Toulouse who joined the collective a year earlier, raises his case:

I worked on a contract basis for a client who didn’t pay me… Well, I worked for him in July, August, and September, and he still hasn’t paid me. And the question was, can it be a… how to say… one of the functionalities [of the collective] to prevent non-payment, or at least late payment by clients?

To finance the scheme, he proposes increasing the percentage taken by the structure on each project carried out under its brand^[13], which would allow the collective to ‘turn into a kind of intermediary that takes up more space than expected’. The responses he received from the other members present were all negative, using cost arguments (‘[the collective] will not be able to take money out of its pocket every time a client does not pay’) and calling for individual responsibility. The response of Maxime, a developer from Lyon who joined the collective two years earlier, is particularly clear in this respect:

I don’t think the network was created for that […]. For everyone, the best thing is to get paid regularly, and not to have many unpaid bills from the client. That is to say, to try to make payments at least once a month. Especially if it’s under a contract, it’s completely fair! And for me, it’s really a way of avoiding to have to spend three months without pay! In other words, at some point you have to put the brakes on: when you really put yourself in danger, stop doing it, at some point!

Nor is there any question of the mutualised functions giving rise to salaries, not only so as not to contravene the dogma laid down in the ‘manifesto’, but also because the idea prevails that the resilience of the organisation presupposes not entrusting too many things to a ‘permanent’ person on whom it would then become dependent. This is what Patrice said during my first meeting with the collective: ‘If [the collective] had an employee who was in charge of something very specific internally, and from one day to the next he is no longer there, it could jeopardise the network in a way, and that is not what we want’ (interview, October 2017). Thus, the ‘sales force’ that Sébastien put forward in his prefiguration did not lead to the creation of a dedicated team either: it is expected that the members’ networks and the reputation of the ‘brand’ will generate the incoming flow of customers.

The elements that are pooled are very concrete and effectively limited to the essential ‘tools’: the services of a lawyer and an accountancy firm, professional liability insurance, and server rental. Together with the use of the ‘brand’, this package of services constitutes a kind of integrated business scheme, a platform in the original sense of the word. The financing of this parsimonious pooling is not based on a membership fee but on commissions deducted from the profits made by using this platform. Until recently, it was the SAS initially created that owned the brand, subscribed to the service package, invoiced the clients, and therefore deducted the commission, which amounted to 5%, from the amounts paid.

6. A continuously deliberating heterogeneous organisation

This company is also the bridgehead of the collective’s heterogeneous governance structure. Sébastien describes it as a conventional company type that is operated as a cooperative:

We have rather special statutes: you are only allowed to have one share per person, paid at the nominal value, you don’t get dividends, and… the capital is open, it’s open variable capital. So the idea is really to say that anyone who wants to get involved –the share price is €1,000, so there is a small entry barrier to make sure that people commit themselves– but the idea is to bring in partners, in a cooperative form. We’re not a cooperative because French law is standardised on what a cooperative is, so we don’t fit into the boxes. (interview, March 2019)

As described by Aurélien, however, partnership is not a trivial operation:

It’s a long process to become a partner: […] from the moment you say you want to be partner and the moment you actually are and you have your share of the business, it’s a year, a year and a half, generally. […] We tell people who want to become partners [that] they must have worked with the network for at least twelve months. And then you need the unanimity of the existing partners to become partner. […] You have to be senior in your activity. (interview, February 2019)

This probably explains, in part, why the number of partners in the company never exceeded ten or so during my observation period, while at the same time the collective grew from 139 to 643 members. Another important factor is the responsibility linked to this status: all the projects carried out must be validated by a partner, who must then ensure that it is carried out correctly, without entering into a subordinate relationship with the members of the working team.

The small size of the bridgehead company limits the working possibilities within the collective: ‘Once you’ve burnt yourself out with all [the partners], well, you can’t work anymore!’, Aurélien confessed to me during our first meeting (interview, October 2017). It also constitutes a clearly identified obstacle to its development: ‘Our big bottleneck today is that we lack partners’, he also conceded at the time. Finally, of course, it formally concentrates power in the hands of a few people. On this last point, however, observation of the functioning of the collective shows that this configuration is counterbalanced by a set of practices that closely combine egalitarianism, transparency, and inclusion.

What first strikes the observer is the impressive transparency of the organisation and its daily life. This is achieved first through the instant messaging system already mentioned, to which all members have access and which is their main tool for coordination and deliberation. The organisation is in fact in perpetual debate and is built in real time, as each situation arises. Of course, the chat is not the only place where this construction takes place: it also does so during synchronous collective times, such as the co-living or the regularly organised video-conferences (in particular, the ‘agora calls’, which I mentioned above). These spaces are open to all and are systematically announced. Another important pillar is the Google Drive, which is open to all and contains all the organisation’s documents. Among other things, there are ‘manuals’, which include principles and ‘good practices’, available in the form of shared files open for writing and scattered with comments from members suggesting modifications. This cloud also contains all the collective’s archives in free access, and in particular a large number of meeting minutes and recordings of internal video conferences.

The feeling that emerges when observing the functioning of the collective is that, in the words of one member, ‘it’s always debating’ and that ‘nothing is really settled’. The concern for the transparency and traceability of debates is both linked to the desire to implement a truly ‘liberated’ organisation and to the need to coordinate from a distance, as face-to-face collective time is limited by nature. This mode of organisational construction, based on horizontal, unrestricted, and transparent remote interaction, is reminiscent of the way in which the development of open-source software has been described^[14]. As in that world, ‘open contribution’ is explicitly a basic principle of the collective.

7. Trying to ‘scale up’

Early on, the founders of the collective, and more particularly Sébastien, had the ambition to build a large-scale network based on this alternative business model. With their characteristic sense of humour, they created a channel called ‘World Conquest’ as soon as the instant messaging system was set up, described as follows: ‘Here, we do as Pinky and the Brain do, we talk about the strategy to conquer the world’. There are many passionate discussions about initiatives that could help the collective to spread beyond its borders and thus contribute to its development.

Indeed, the ‘vision’ drafted during 2017 clearly sets the goal of ‘scaling up’. The collective ‘already offers us a fulfilling and effective working life; we think that’s not enough. A larger size will multiply our strengths through the network effect’, it says. This last concept was introduced by Sébastien, who studied the economy of digital platforms and came to the conclusion that it was necessary to mobilise the same mechanisms to propose an alternative, thus confronting a classic start-up problem: ‘scalability’. But the nature of the collective obviously makes the equation more complex: as the collective expands geographically, the question of maintaining the physical, social, and decision-making proximity that has given it its identity and cohesion becomes more and more acute.

The path indicated in the ‘vision’ is that of autonomous local ‘cells’, which will allow ‘to grow without recentralising’ and thus ‘to build a solid and powerful enterprise, over which we keep control and in which it is good to live’. The credo is that ‘to maximise contributions, the network must be a gathering of small, fully autonomous collectives that choose to equip themselves with common tools that allow them to have strength in numbers when necessary’. The ‘vision’ can thus be read as a strategy for ‘scaling up’ in line with the ‘manifesto’ principles.

Two initial branches were created in 2017 in the West of France, but they were closely linked to the historical Parisian structure and their creation was more the result of a combination of circumstances than a strategy. As of mid-2018, efforts were made to create ‘cells’ in all major French cities. The heterogeneity of each chosen statute is the first expression of their autonomy. Today, two thirds of them are SASs, but there are also associations, a rising form often chosen by more recent entities. At the same time, the collective is changing the way it refers to itself, increasingly favouring the terms ‘network of collectives’ or ‘federation’.

8. The challenge of establishing a multi-level horizontal organisation a posteriori

Until now, as we have seen above, the governance of the collective combined a bridgehead in the form of a ‘classic’ company, with some ‘cooperative’ specificities, and a set of very egalitarian collaborative practices that put the organisation under constant deliberation. With the extension of the network and the swarming of autonomous ‘cells’, the need was quickly felt to give the Parisian company the status of a ‘cell’ among the others.

The first stage of the transformation undertaken to adapt the organisation to the new issues began at the end of 2018, when the position of ‘brand captain’ was introduced to allow members who are not partners of the Parisian company to be given responsibility for projects carried out by the collective platform. The ‘manual’ uploaded in mid-2019 to the shared Google Drive to formalise this new position states in a preamble: ‘A small clarification: you are a partner at the cell level, you are a captain at the brand level’. However, the ‘captains’ must be able to legally commit the structure that supports their supervised projects, which implies becoming partners. In other words, this new status essentially formalises the possibility for partners of local ‘cells’ to use the ‘brand’. It also allows them to receive the inbound flow of customer requests sent from the collective’s website, which is not published in the chat room, but only sent to the ‘captains’ e-mail list. The ‘captains’ are co-opted from among the members who are considered ‘senior’ in their profession and receive 5% of the project team’s income as part of their function. At the end of 2020, there were only about twenty for 643 members, and, on the internal chat, their lack was regularly felt as there were many incoming projects which unsuccessfully called for a captain.

The second important step in the organisation’s transformation is the creation of a specific governance structure for the brand. Also at the end of 2018, the process of creating a cooperative society (specifically a Société Coopérative d’Intérêt Collectif, SCIC) was launched. The SCIC is a legal company with cooperative statutes based on the ‘one member, one vote’ principle, with the advantage of allowing several categories of members, including legal entities, to be integrated into the governance. Despite the efforts of the project’s initiators to involve as many members as possible in the process, few of them actually participated in its development, not even by taking part in the online vote organised to specify its form. The SCIC was officially registered a year later, in October 2019, with a relatively small number of co-founders^[15] compared to the number of members of the collective and the value set for the subscription of a share (100 euros).

Tensions quickly arose over the prerogatives of the new structure and its economic model. The first stumbling block was the transfer of the brand to the SCIC, as the members of the historic Parisian company felt that they should be remunerated for the investments made over the past four years to develop it. The setting of the amount and conditions of this takeover gave rise to tense exchanges between the members of the collective, which had previously been marked by a consensual and benevolent atmosphere. There was also a second point of contention: the request of a loan to ensure the development of the brand but also to clear the debt of the SCIC towards its president, who had not been paid for her administrative work. These two issues pit the partners of the historical company, which were responsible for these initiatives, against some of the other ‘cells’, particularly the more recent ones, whose members were very wary of paying a posteriori for the work done on the brand and the provisional governance of the SCIC, and moreover under conditions that they perceived as being defined unilaterally by the Parisian ‘centre’ of the collective. Many of them also felt that their ‘cells’ were still too new and fragile to guarantee the planned loan, or that their statutes did not allow them to do so.

The economic model devised to finance future investments for the development of the ‘brand’ was not unanimous either. It consists in taking a 10% commission on projects arriving via the ‘brand’, which is considered as a ‘business introducer’ that has to be remunerated. Combined with that of the ‘cell’ and the ‘captain’, this commission affects the income generated in proportions that can be higher than those of freelancing platforms. Thus, in the spring of 2020, a future ‘captain’ from Lyon considered, in his ‘discovery report’ presented to the ‘agora call’, that it was ‘quite hard for a non-captain senior to see the added value of a project going through [the collective]: -10% minimum (-20% with the brand’s commission) rather than through his network (0%)’.

In 2020, in the midst of the first COVID-19 lockdown, tension was at its highest. The ‘agora calls’ followed one another, alternative projects were presented and, faced with a lack of consensus, the SCIC board resigned. After considering closing the structure down, a new team agreed at the last minute to take over and try to save it. This was the beginning of a period which is still ongoing at the time of writing (the end of 2020), wherein the collective tries to respond to the challenge of instituting a posteriori an organisation made up of individual actors (the ‘members’) and collectives (the ‘cells’) sharing a common platform with cooperative governance, while retaining the greatest possible freedom for freelancers. Marked by the health crisis, which forced members to remain distanced and led the collective to cancel the annual gathering that contributes to its sympathetic cohesion, this period is experienced as a deep internal crisis.

Conclusion

The story of this digital freelancers’ collective illustrates the power, in the digital business services sector, of the movement towards organisations that give individuals more freedom at work. This configuration must be analysed in the light of the conditions in this sector, where the capital intensity of the activity is often low, tensions on the labour market are high, and traditional IT services companies (SSIIs) are seen by many as a deterrent. Everything contributes to making freelance work attractive for digital professionals.

This story shows that, behind individuals’ appetite for freedom at work, collectives do not disappear but are, on the contrary, at the heart of new dynamics through which freelancers seek to fulfil their desire to work and do business differently. The fact that the story continues to be written with difficulty suggests that the regulation and governance of such collectives, when they take on a certain dimension, constitute a formidable challenge.

The collective took off initially because it was created by two charismatic individuals who instituted a principle of conviviality, benevolence, and peer-to-peer dialogue that corresponded to the aspirations of its members. This held as long as it was possible to share enough collective time, in a friendly atmosphere, so as to achieve informally the ‘free together’^[16] or ‘collective independence’^[17]. The strategy of spinning off into smaller local collectives united by a common brand platform should have made it possible to reconcile organic organisation and scaling up, but it generated a centrifugal dynamic and too much distance at the time when the institution of this cooperative meta-collective level was being deliberated.

There were also situational factors that played a negative role: the health measures linked to COVID-19, which reinforced this distance, but also the distancing of the two co-founders of the collective. Sébastien and Aurélien gradually shifted their energy towards the development of interoperable solutions based on the Solid standards (for ‘social linked data’) promoted by Tim Berners-Lee, the ‘father’ of the World Wide Web, whom they met in Boston in Spring 2019. The main field of application of these technologies is not accidental: a tool to equip and interconnect digital freelancers and their collectives, like a decentralised platform. A way of responding through tools, in the manner of engineers, to the problem of the institution of the meta-collective. But that is another story.

Bibliography

Blein A., 2016, “Le coworking, un espace pour les transactions hors marché ? La valorisation des réseaux sociaux pour travailleurs indépendants”, Réseaux, n° 196, 147-176.

Bureau M.-C. et Corsani A., 2014, “Du désir d’autonomie à l’indépendance. Une perspective socio-historique”, La nouvelle revue du travail [En ligne], n° 5.

Demoustier D., 2006, “L’économie sociale et solidaire et le développement local”, in J.-N. Chopart, G. Neyret et D. Rault (dir.), Les dynamiques de l’économie sociale et solidaire, Paris, La Découverte.

Fondeur Y. et Sauviat C., 2003, “Les services informatiques aux entreprises : un “marché de compétences”“, Formation Emploi, n° 82, 107-123.

Fondeur Y., 2013, “Services de conseil en informatique : recruter pour placer”, La Revue de l’IRES, n° 76, 99-125.

Fondeur Y., 2020, “À la recherche de nouveaux modèles de travail collectif dans les services numériques”, in F. Rey et C. Vivès (dir.), Le monde des collectifs, Buenos Aires / Paris, Teseo, 177-196.

Fondeur, Y., 2022. Les freelances du numérique au prisme de « l’économie collaborative ». Rapport d’études Dares, n°24, https://dares.travail-emploi.gouv.fr/publication/les-freelances-du-numerique-au-prisme-de-leconomie-collaborative

Granovetter M., 1974, Getting a job. A study of contacts and careers, Chicago, The University of Chicago Press, 2nd edition 1995.

Mauss M., 1925, “Essai sur le don. Forme et raison de l’échange dans les sociétés archaïques”, L’Année sociologique, repris in : M. Mauss, 1950, Sociologie et anthropologie, Paris, PUF, réédition 2012.

---