Introduction

Several Open Science Partnerships (OSPs) have emerged around the world. These pre-competitive public–private research partnerships adhere to principles of open science, which include freely sharing their research outputs in the public domain and precluding participants from seeking intellectual property (IP) rights protection on outputs from the partnership.

Open Science extends far beyond OSPs. It builds on long-standing openness norms and practices in science (Merton 1968, 1973; Dasgupta & David 1994; David 2008; Hosseini et al. 2022) and applies them to a broad range of activities within and beyond scientific research, including open data, open access, open methodology, open source, open peer review and open education (Watson 2015). Vicente-Saez & Martinez-Fuentes (2018, p. 428) defined Open Science as “transparent and accessible knowledge that is shared and developed through collaborative networks”. Earlier definitions emphasise its focus on free and broad dissemination of research knowledge, data and tools beyond personal networks (enabled by the Internet and digital technologies, see, e.g. Dai et al. 2018; Vicente-Saez et al. 2020; Stieglitz et al. 2020; Ramachandran et al. 2021), on early or immediate sharing of research outputs, and on communication of both positive and negative research findings (De Roure et al. 2010; Bartling & Friesike 2014; Schlagwein et al. 2017). Openness in science has been associated not only with benefits for the progress of science itself but also with economic benefits through increased efficiency and productivity in research and development as well as enablement of innovation (Fell 2019). Ultimately, it thus supports broad diffusion of research through commercial mechanisms (Fernández Pinto 2020).

Research on Open Science is part of a broader body of work on openness in research and innovation processes where openness is seen as a higher-order concept that refers to “accessibility of knowledge, technology and other resources; the transparency of action; the permeability of organisational structures; and the inclusiveness of participation” (Schlagwein et al. 2017, p. 297). For instance, Open Science has been linked to the related concept of Open Innovation in companies (Chesbrough 2003; Schlagwein et al. 2017; Bertello et al. 2024), and some studies actively seek to bridge the literature on openness in scientific research with studies of openness in innovation processes in the private sector (Friesike et al. 2015; Beck et al. 2020, 2023).

One stream of research on Open Science focuses on the role of open collaborative research in making research more efficient (Fecher & Friesike 2014), and this is the body of work to which the current study seeks to add. Collaborative research is an integral element in Open Science (Masum et al. 2013; Ramachandran et al. 2021), and open collaboration has been described as an effective and perhaps necessary means to tackle the complex scientific challenges facing society and industry today (Nielsen 2012; Gold 2021).

Some studies examine open collaboration among scientists, for example, big team science (BTS) collaborations where diverse researchers jointly address complex questions that are beyond the reach of individual partner groups (Alessandroni et al. 2023; Baumgartner et al. 2023). Other studies concern open collaboration that involves industry partners (e.g. Jong & Slavova 2014; Perkmann & Schildt 2015; Rotolo et al. 2022). Yet open collaboration at the “front end” of the innovation process—which includes the interface between science and industry—has been far less studied in the Open Science and Open Innovation literatures (Friesike et al. 2015) despite evidence that it can benefit companies (West & Bogers 2014) and bolster academics’ contribution to Open Science (Bikard et al. 2019).

In this study, we focus on open public–private research collaborations, specifically OSPs (Gold 2021; Ali-Khan et al. 2018a, 2018b; Gold et al. 2019).

OSPs have been defined as “private–public collaborations that have certain common elements: open access publications, open sharing of data, tools and materials and the absence of intellectual property rights that restrict improvement or use of jointly created inventions” (Gold 2021, p. 2). The open principles that define OSPs differ markedly from standard practices in pre-competitive research partnerships, which often restrict sharing of outputs and allow participants to secure the rights to any IP that may be developed in the collaboration (Stevens et al. 2016). Indeed, OSPs aim to remove “roadblocks not only to the sharing of information, but to its use” (Gold 2021, p.7) by mitigating barriers to university–industry collaboration (Perkmann & Schildt 2015) and to access and use of science (Morgan Jones et al. 2014; Morgan Jones & Chataway 2021).

Though OSPs remain relatively rare, their number is growing, and they are gaining attention as a means of addressing long-standing challenges associated with university–industry collaboration with more restrictive policies on IPR and knowledge sharing. Recent work has shown that university–industry research collaborations are often governed by complex legal agreements to protect the parties’ interests. However, these agreements can limit the openness of science, further development of research results (Egelie et al. 2019) and thereby the wider benefits of the funded research (Lie et al. 2024). Calls have been made to rethink IPR practices given the growing focus on Open Science (Kumar 2024), and OSPs have been proposed as a mechanism to avoid restrictive IPR on research outputs to promote sharing of knowledge and data and, ultimately, greater efficiency in public and private research (Gold 2021).

Despite the growing interest in OSPs, scholarly attention has been limited. Existing studies focus on a single OSP (Perkmann & Schildt 2015; Morgan Jones et al. 2014; Morgan Jones & Chataway 2021) or address OSPs as a general concept (Gold 2021; Ali-Khan et al. 2018a, 2018b; Gold et al. 2019).

Although OSPs share the elements described above, it quickly becomes apparent that they differ in many other respects, including intended ends and means deployed to reach them. Research on pre-competitive public–private partnerships has long since acknowledged the importance of recognising and understanding how partnerships differ to better appreciate how to implement them and study their effects (Schaeffer & Loveridge 2002; Kurtmollaiev et al. 2023). Such differences have not yet been systematically investigated in OSPs, even though recognising them is key to understanding which types of OSPs are suitable for a given purpose, and which effects can be expected from different types of partnerships.

To understand the emerging phenomenon of OSPs and their potential impact on the progress and use of science, this paper takes a first step towards a better understanding of how OSPs differ. We identify features that are useful in understanding the variation across OSPs and explore how the differences may ultimately affect outcomes and impact.

We draw on the concept of organisational design (Burton 2006; Greenwood & Miller 2010) to identify design elements that can be used to characterise and distinguish between OSPs. Organisational design has previously been overlooked in the literature on management and organisation (Aubry & Lavoie-Tremblay 2018; Burton et al. 2019) but has gained traction in recent years (e.g. Burton & Obel 2018) in the study of established companies (e.g. Stea et al. 2015; Domínguez Escrig et al. 2020; Al-Atwi et al. 2021; Foss & Klein 2023) and entrepreneurial ventures (e.g. Colombo et al. 2016; Burton et al. 2019) and projects (e.g. Aubry & Lavoie-Tremblay 2018). It has also been deployed to the study of public–private partnerships by George et al. (2024), who emphasised the complexity involved in designing such partnerships and the role of design in aligning partners and enabling them to deliver intended outcomes. Understanding how university–industry collaborations are organised and managed is crucial to understanding their impact (Perkmann & Walsh 2007). Open collaborations can be seen as a distinct and viable organisational form (Levine & Prietula 2014), and organisational design offers a useful lens to study the processes by which open collaborations are formed and the resulting organisations (Aubry & Lavoie-Tremblay, 2018). This constitutes an appropriate starting point for our present study.

Based on Good et al. (2019) conceptual model for determining how organisations are designed in order to accomplish their goals, we focus on three elements of the design of OSPs: purpose (i.e. what goals they strive to attain), activities (i.e. what types of research activities they undertake, and how these activities are organised) and structure (including ownership, funding and how key decisions about how to pursue goals are taken). Specifically, the paper asks: How do the organisational design elements of OSPs in biomedical research vary, and what impact do these variations have on how OSPs are deployed and governed?

In this exploratory study, we use insights from case studies of five OSPs in biomedical research to generate a list of organisational elements and attributes that could be used to characterise OSPs. The five OSPs, which are briefly introduced in Table 1, are: Structural Genomics Consortium (SGC), Open Targets (OT), Enabling & Unlocking Biology in the OPEN (EUbOPEN), The Neuro’s Early Drug Discovery Unit (EDDU) and The Open Discovery Innovation Network (ODIN). Information on partners in the five partnerships, including industry partners, is presented in Table 2.

Table 1 Presentation of the five OSPs included in the comparative study.
Full size table
Table 2 Number and list of partners in the five OSPs examined.
Full size table

Based on interviews and a workshop with representatives of the five OSPs, the list of attributes was refined, relationships between organisational design components were examined, and archetypes of OSPs were developed. The OSP representatives served as interviewees, workshop participants and co-authors of the study and contributed through the writing and revision process to the analysis and findings reported here to ensure relevance for both scholars and practitioners.

We argue that understanding the purpose of an OSP is crucial to comprehending how OSPs differ. We distinguish between two key components of this purpose. The first refers to the predominant intent of the OSP, as indicated by the relative weight placed on the advancement of the progress of science vs. the advancement of the use of science, notably in the private sector. The second component refers to the nature of the research aims pursued by an OSP and focuses on whether these aims are directed or open-ended. Based on these two components, we propose four ideal types of OSPs that highlight the varied forms such partnerships can take. We also investigate how the purpose of an OSP shapes key attributes of its activities and structure, including how they are governed and how their openness principles are translated into practice.

Methods

An exploratory, inductive approach based on qualitative case studies

Given the limited work on OSPs and their emergence as a phenomenon, we chose an exploratory approach to this study as well as an inductive, grounded approach to ensure that the characterisation of the design of OSPs was informed by real-world cases. To this end, we undertook qualitative case studies of five OSPs within the biomedical field.

Focusing on OSPs within one field of research reduces potential distortion from field-specific differences in university–industry interactions (Eisenhardt 2021), which is important given the large variation across fields and industries in the degree and characteristics of engagement between academia and the private sector (Perkmann & Walsh 2007). The biomedical field was chosen as empirical context for the study, in large part because this is where the vast majority of OSPs have emerged (Gold 2021), although the field has traditionally relied, and continues to rely, heavily on IPR to appropriate returns from investments in research and development (Egelie et al. 2019). More generally, research-focused pre-competitive public–private partnerships have been common in biomedical and pharmaceutical R&D for decades (Schuhmacher et al. 2018; Olk & West 2020; Slavova 2022), pooling partners’ resources to pursue joint research and develop new technology platforms, tools, databases and/or predictive models (Stevens et al. 2013, 2016; Vertinsky 2015). Pharmaceutical firms face substantial R&D challenges, e.g. high complexity of unsolved problems, decreasing R&D productivity, rising costs of R&D and insufficient fundamental understanding of many diseases (Tralau-Stewart et al. 2009; Müller & Weigelt 2010). This has led firms to pursue new avenues, including open and collaborative innovation models (Müller & Weigelt 2010; Stevens et al. 2016).

The selected OSPs were identified via internet searches and assessed according to the following criteria: (i) formal goal-oriented agreements between at least one public academic partner and at least one private sector partner; (ii) explicit focus on open sharing of knowledge, data, tools, materials and other research outputs with no or minimal restrictions on sharing and further use of research outputs; (iii) ongoing at the time of study or terminated within the past 5 years.

The initial search revealed 17 partnerships that, based on a preliminary examination, had potential to meet the stated criteria. After more extensive document study, again based on publicly available online sources, most of the partnerships were excluded because they failed to meet one or more of the criteria. For example, some of them were not focused on goal-oriented research collaboration but served as platforms for data or knowledge sharing or aimed to eventually establish such collaborations. Ultimately, we identified five OSPs that met our criteria: the SGC, OT, EUbOPEN, the EDDU and ODIN.

Document studies and qualitative interviews

Data were collected initially through a document study of publicly available material and semi-structured interviews with representatives of each OSP, undertaken in late 2021 and early 2022 to identify common features and relevant singularities across OSPs.

Respondents were prominent academic and administrative leaders from the OSPs who were instrumental in managing and developing the partnerships as well as in their initial design and were chosen for their capacity to offer detailed insights about the objectives and functioning of the partnerships.

Based on data gathered through this document study and interviews, the organisational design of each OSP studied was characterised according to the three design elements purpose, activities and structure.

To this end, data were coded in two rounds. First, all components were identified to describe the OSPs’ purpose, activities and structure that could be detected in the data. Interview data were also used to generate possible attributes for each component, that is, to identify possible states, and a list of OSP design components and possible attributes was generated. One component is the scale of activities, and we identified two attributes: large scale and limited scale. With access to data on more OSPs, the list of components could be extended, and the range of attributes become more fine-grained. Thus, the design components and attributes identified here should be seen as a preliminary attempt to understand key features of how OSPs are designed.

Results were used to develop preliminary characterisations of the five OSPs studied. To gather missing data on OSP attributes identified after the first round of interviews and enrich the characterisations of the OSPs, a second round of interviews was undertaken in winter 2022–2023 with the same leading representatives who were interviewed in the first round. Some additional respondents were only interviewed in the first round. The list of respondents and interview guides are available in the Supplementary Materials.

Based on the data collected in the second interview round, the list of components and attributes for characterising OSPs was refined, and the characterisations of the OSPs were updated and sent to the interviewees to ensure factual correctness.

Cross-case analysis was used to identify common patterns as well as differences across the OSPs. Several design components that could be used to understand variations across OSPs were identified in this process step. To achieve a better understanding of these components and the relationships between them, an online workshop was held in March 2023 with representatives from the five OSPs to discuss key organisational design elements as well as possible dimensions to use in the development of OSP archetypes. Information about participants and content in the workshop is available in the Supplementary materials.

Based on inputs from the workshop, the stated purpose of an OSP was identified as the key factor in shaping the deployment and governance of such partnerships. Thus, the archetypes proposed in this paper draw on findings from the case studies as well as insights and experiences from practitioners developing and leading OSPs initiatives.

Collaboration with OSP representatives

A key aim of this study was to learn about the understudied phenomenon of OSPs based on collaboration between the team of social scientists who designed the study and practitioners who design and manage OSPs. In line with the “Open Innovation in Science” approach described by Beck et al. (2020, 2023), the study was designed to engage practitioners in the development, analysis and dissemination phases of the study. The approach was chosen to ensure that findings were informed by deep insight into the OSPs studied and relevant to other practitioners, as open intersectoral collaboration can increase the reliability as well as the societal relevance and impact of research (Beck et al. 2020).

The OSP representatives served not only as interviewees and workshop participants but as active co-authors of the study, engaged in four rounds of the writing and revision process and thus contributed to the analysis and findings reported here.

Results

Characterising the organisational design of OSPs

Our study revealed both similarities and differences across the OSPs examined. We noted three fundamental similarities. First, some form of legal framework, that is, a contractual basis ensuring that parties adhere to open science principles. Second, some form and degree of co-creation of ideas, activities and/or results among public and private partners. Finally, all require open sharing of outputs from the partnership but have different approaches to openness (see below).

Guided by the three organisational design elements—purpose, activities and structure—and based on the data collected in the study, a list of design components was developed to capture the differences that set OSPs apart from each other. These components are displayed in Table 3, which also lists the attributes or possible states of these components, illustrating the scope for variation observed across OSPs. However, the list of attributes is not exhaustive but only captures the variations identified in the five OSPs examined here.

Table 3 Identified components and attributes of the organisational design of OSPs.
Full size table

Two key design components reflect the purpose of an OSP

Based on interviews and particularly the online workshop, key design features of OSPs could consistently, and perhaps not surprisingly, be traced back to the purpose for which the OSPs were established. We identified two components, which are described in the following.

The predominant intent of the OSP

The first component, predominant intent of the OSP, focuses on the stated motivations of an OSP in line with prior conceptual frameworks used to classify research (e.g. Stokes 1997) or forms of university–industry collaboration (e.g. Perkmann & Salter 2012). Though stated aims are not binding, they send important signals internally and externally about the relative weight placed on, for example, industry aims and impacts. Moreover, stated aims can be expected to shape key decisions about how an OSP is designed and implemented.

Predominant intent refers to the overarching motivation to establish an OSP, understood as the main expected impact communicated externally and by which the OSP would ultimately be assessed. While all five OSPs examined referred to advancing the progress and speed of science and to bolstering the utilisation of scientific results, the OSPs assigned different relative weight to the two purposes, meaning that a predominant intent could be discerned. Some stated motivations, for instance SGC’s and EDDU’s, prioritise advancement of science as a driver for the subsequent development of new and better treatment for patients, as stressed by a representative from EDDU:

It’s not just about doing projects with industry. It’s about learning and training [early career researchers] and putting knowledge out in the open. Our goal is really to see if we can work on diseases … where there has not been a huge amount of discovery or advancement in new therapies and ask, ‘how could we accelerate it and go a little faster?’

Other OSPs explicitly emphasise enabling and accelerating the use of scientific research outputs, particularly by bolstering or accelerating innovation in industry. Examples include OT, which aims to reduce difficulties and costs associated with drug development, and ODIN, which was established to increase and accelerate the use of science in industry.

In summary, the two main attributes identified for the predominant intent of OSPs are to advance the progress of science and to advance the use of science.

The nature of the research aims pursued

The second component is the nature of the research aims. The possible attributes identified for this component are whether the research aims are directed or open-ended. Some of the OSPs pursue very specific aims. For instance, SGC’s mission was, in its first 20 years, to understand all proteins encoded by the human genome. During this period, it determined the structures of thousands of human proteins and invented and disseminated hundreds of chemical probes, which have since been further applied in science research as well as in new drug discovery programmes in industry. In the next 10 years, the SGC will focus on Target 2035, a global initiative to develop pharmacological tools to modulate every protein in the human proteome by 2035. On a similar note, one of EUbOPEN’s primary aims is to create a “first generation” chemogenomics library comprising 4–5000 known compounds and covering 1000 targets. As explained by a representative from EUbOPEN,

Our research aims are very specific and highly numerical and were determined from the very start. Our focus has been to organise the consortium in a way that we work together to achieve these challenging goals.

Other OSPs are motivated by a desire to promote open research and collaboration within a specific topic. EDDU’s mission is to conduct fundamental research that can lead to the development of new and improved treatments for neurological disorders. All their activities build on induced pluripotent stem cells, but the scope and aim of individual activities can differ greatly. Similarly, ODIN was established to promote open university–industry collaboration within early-stage drug discovery research and provides funding for such collaborations, but the aims and contents of joint projects are co-developed by participating academic researchers and companies.

Four archetypes of OSPs distinguished by differences in the ends they pursue

Based on the two components described above, we propose four archetypes of OSPs, defined by their stated purpose (see Fig. 1).

Fig. 1: Four archetypes of OSPs.
figure 1

Pictograms sourced from Microsoft Word Icons.

Full size image

In the following, we introduce the four archetypes. The archetypes are not intended to characterise the five OSPs studied but should be seen as ideal types of OSPs that underline that different ends call for different means and affect how OSPs are designed. A real-life OSP may not fit neatly into any one archetype and may even have aspects of multiple archetypes.

The mission

Mission OSPs have specific research aims and attach relatively greater weight in their stated purpose on advancing the progress of science than on advancing the use of that science. SGC’s core activities to understand the functions of proteins encoded in the human genome are an example of a mission-oriented approach to OSPs. As described by a representative:

The genesis of the SGC was the appreciation that most of the proteins encoded by the human genome were understudied. The current system had no easy way to raise the finances needed to go after them …, so we needed to create a mechanism that allowed us to study them … How does one create an ecosystem where you can have multiple collaborators from academia and industry, from different countries, working towards one objective and a big problem that no one could solve? The only way you could do that was under an open science umbrella. … There’s just no other way to do what we wanted to do under any other collaborative framework.

The infrastructure

Infrastructure OSPs are also primarily science-driven but pursue activities that build on an existing scientific infrastructure, biobank, set of methods or the like. EDDU is an OSP that develops and pursues collaborations based on its existing capabilities and infrastructure. As explained by a representative:

When we were set up, we had funding to get a small core platform up and running. … At the core of everything we do are human induced pluripotent stem cells … and we use our knowledge and tools to both further understanding of diseases and to see if we can advance new therapies or at least contribute a little bit towards that.

The club

Among the OSP archetypes that attach greater relative weight to supporting the use of science to bolster innovation in industry, we find club OSP, which are industry-oriented partnerships that pursue specific research goals. OT fosters collaboration among pharmaceutical companies and research labs to accelerate target identification and validation in key strategic areas of interest for the industry participants. A representative explains:

A driver for the formation of Open Targets in 2014 was the translation of findings from genetics, which can improve the odds of success in drug development, and genomics from academia to industry. With a focus on target identification and prioritisation questions, before candidate drug molecules are developed, pre-competitive collaboration is possible between multiple commercially competitive industry partners and academic groups.

The hub

Finally, the hub OSP, like the infrastructure, has open-ended research aims. Unlike the infrastructure, it focuses on providing a platform for academic–industry collaboration rather than pursuing specific research aims. ODIN stimulates use-oriented basic research collaborations that are co-developed and co-executed by academic researchers and industry partners within the ODIN network, as explained by a representative:

Part of the uniqueness of ODIN comes from the capacity it gives researchers to develop their own ideas for collaboration with companies, and for companies to find researchers who can help them solve the challenges they face. … We start [a call] with company pitches, where companies share research challenges that are open to interpretations and can be addressed by many disciplines. We ask researchers to pitch possible solutions and discuss the problem together from many different perspectives. … We also turn it around and allow researchers to pitch their ideas and methods that could be useful to industry and form the basis for a collaborative project.

As mentioned, OSPs usually do not only belong to one archetype. For example, EUbOPEN can best be described as a combination of a mission OSP, pursuing very specific research aims to advance science on 1000 proteins, with aspects of a club pursuing open collaborative projects with a set of industry partners.

OSP archetypes and the means they employ to pursue desired ends

We have argued that appreciating the purpose of an OSP is crucial to understanding other key design features, including how it is governed and deployed in practice. Insights collected through interviews and the online workshop with OSP representatives confirm a link between the purpose of an OSP and other design features related to its activities and structure. These observed links are described in the following—grouped by whether they appear linked to the nature of an OSP’s research aims, its predominant intent or both—and summarised for each OSP archetype in Table 4.

Table 4 Design component attributes, by OSP archetype.
Full size table

Organisation of research activities

How research activities are organised in an OSP appears linked to the nature of its research aims. OSPs with directed, specified goals are likely to have top–down designed research programmes that ensure that activities set in motion will be able to realise the chosen goals. For instance, SGC, OT and EUbOPEN have extensive coordination mechanisms in place to organise and align distributed research activities. In contrast, OSPs with open-ended goals tend to have bottom–up developed projects that may have little in common besides a common infrastructure or emerging from the same collaborative hub.

Relatedness of research activities

Not surprisingly, the extent to which activities within an OSP are related or heterogeneous is also linked to the nature of the partnership’s research aims. Activities within a directed OSP are likely to be related and complementary to allow for coordinated efforts to contribute towards a shared goal. Meanwhile, activities in an OSP with open-ended goals do not require coordination, as research aims are determined for individual projects rather than for the partnership.

Scale of research activities

The nature of an OSP’s research aims is also associated with the scale of activities. OSPs with open-ended goals tend to include a broad and often heterogeneous set of activities, which can include small-scale projects. In contrast, directed OSPs with very specific and ambitious scientific research aims are likely to necessitate an efficient, larger-scale set-up suited for high-throughput, high-efficiency research.

Ownership (autonomy)

The five OSPs examined have very different organisational set-ups. The relationship between purpose and ownership is not entirely clear from the exploratory study but does appear linked to the nature of the research goals. More specifically, the study indicates that large-scale directed programmes are likely to involve multiple partners and thus to be independent organisations, for example, independent legal entities like SGC or research consortia like EUbOPEN and OT. OSP representatives pointed out that having an independent organisation rather than being embedded at a specific university offers greater degrees of freedom in identifying relevant researchers to connect to the partnership without being limited by available resources at the university, and shields the OSP from the risk of changes to university policy regarding open collaboration. An independent organisation is often a given in the case of large consortia with multiple partners to ensure a stable and efficient organisational structure, as illustrated by a representative from EUbOPEN:

A large consortium such as EUbOPEN has the advantage to be able to leverage the strength of the different partners. Given a well-functioning organisational structure, which can be challenging in itself, a large international consortium can achieve more and help spread best scientific practice in an unprecedented way.

OSPs with open-ended aims are likely to be less reliant on a specific organisational form and could in principle be independent or embedded at a university. A representative from ODIN reflected on the benefits and implications of a university-based set-up:

One of the advantages for us of being embedded within a university is that we can draw on the internal support that the university offers. But, of course, whether you’re part of a university or an independent organisation affects the goals you have to meet. So, we not only look at our own objectives, but also have to consider how we can add value to the university.

Open sharing

The exploratory study indicated that practices surrounding open sharing from an OSP are related to the predominant intent of the partnership. OSPs with primary focus on the advancement of science are likely to have extensive and strict protocols to ensure unmediated and immediate public sharing of outputs. As described by a representative from SGC:

The idea is that no matter what we make, whether you’re in industry or an academic or a foundation, we will allow you to have it or we make it available to you … And we don’t put any intellectual property restrictions on anyone who purchases it. We’ve done that for thousands of reagents … At least 25,000 samples of chemical probes that we have made have been distributed around the world, 10,000 papers reference them, and they are not encumbered by any intellectual property strings. Everyone can use them as they will, and discoveries made with these reagents are being tested in nearly 100 clinical trials now. That’s one of the impacts we measure: how many of the widgets that we made are used by people to make meaningful discoveries.

In contrast, OSPs with a greater relative focus on advancing the use as opposed to the progress of science are more likely to give participants options for mediating open sharing, including reviewing public disclosures for potential IP or possibilities to delay the timing of disclosure. These were described by one practitioner as “safety valves” which are rarely if ever used but can mitigate perceived risks for industry partners in open science collaborations. For instance, although the OT consortium aims to share all research outputs in the public domain, there is a process to assess and potentially raise IP before public disclosure.

Influence of industry participants

Another component of the design of OSPs which is related to the predominant intent of the partnership is the extent of influence private sector partners have on strategic decisions in an OSP. OSPs with a greater relative focus on enabling the uptake and application of science are likely to give industry participants greater opportunities to influence key decisions about for instance which activities are pursued compared to OSPs with greater relative focus on the advancement of science. A respondent from OT describes the value of engaging closely with industry:

For Open Targets research to impact drug discovery decision making, insights and input from industry are sought as part of new project generation. Industry and academic scientists with related interests are matched to develop new proposal ideas together, working in partnership is designed in from the start. Without this approach to engagement, interactions between industry and academia can be transactional, and scientifically strong proposals could be developed that do not address industry drug and target discovery questions sufficiently to be impactful.

Decision-making authority

The final component is linked to a combination of the nature of an OSP’s research aims and its predominant intent in the form of the archetypes. This component focuses on whether key decisions in an OSP are made by a small centralised unit or by a larger collective. The exploratory study indicated that concentration of decision processes in a small centralised unit is likely in both missions and clubs given the importance of securing coordination of distributed activities for the OSP’s ability to achieve its overarching research aims. At the same time, given their tendency to have larger-scale operations, both types of OSPs are likely to involve some delegation of scientific decision-making authority to smaller groups or work packages. Centralised decision-making authority is also likely in an infrastructure OSP to ensure that the total portfolio of projects adheres to principles for development and use of the central infrastructure.

Finally, in hub OSPs, key decisions about scientific activities are likely to be decentralised, delegated to participants in individual projects. Nonetheless, the hub is likely to retain a measure control of the OSP’s overall direction and guiding principles.

This section has explored the relationship between the purpose of an OSP and the design of its activities and structure. As summarised in Table 4, the four OSP archetypes demonstrate distinct design features, tailored to accommodate their intended purpose.

Discussion

The growing number of OSPs calls for a better understanding of how such partnerships may vary. Our study underlines that OSPs, despite their obvious common features, differ in important respects. In this exploratory study, we take a first step in describing their organisational variety, building on prior studies of university–industry collaboration in the biomedical field (e.g. Stevens et al. 2016) and offering deeper insight into the nature and possible sources of variation in open collaborative partnerships.

The study also adds to the broader literature on public–private partnerships, which has drawn links between the function and chosen form of partnerships (Kurtmollaiev et al. 2023) and explored the implications of whether the aim in a partnership is well defined or open-ended (Schaeffer & Loveridge 2002). In a similar vein, we apply an organisational design perspective to examine systematically how ends and means are connected in open public–private research partnerships.

We propose that understanding an OSP’s intended ends, or its purpose, is crucial for understanding how they are designed. We identify two dimensions of this purpose—the nature of the research aims and the stated predominant intent—and propose four archetypes of OSPs based on them. Finally, we reflect on how the purpose of an OSP affects how its activities and structure are designed, adding further depth to the description of the four archetypes proposed. We thus also extend prior work on OSPs (e.g. Gold 2021; Ali-Khan et al. 2018a, 2018b; Gold et al. 2019) by highlighting the different forms OSPs can take, depending on what they were established to accomplish, and identifying crucial elements in the design and, ultimately, impact of these partnerships.

The propositions put forth in this paper are exploratory, as they are based on insights from only five OSPs. As the number of OSPs increases, and more data can be considered, the list of design components and attributes, and our understanding of the relationships between them, can be expanded and become more fine-grained.

Moreover, as mentioned, the archetypes represent ideal types and not necessarily real-life OSPs. In practice, an OSP may include multiple programmes and projects that fit into different archetypes. The purpose of the archetypes we propose is not to capture the full complexity of individual OSPs but to highlight crucial differences in their objectives and how these differences affect the way OSPs are designed.

Future work could seek to validate and extend the archetypes we propose and explore whether some OSP archetypes are more suitable under certain conditions than others. Future work may also investigate key success factors across archetypes. For instance, it is probably crucial for mission OSPs to maintain a high-performing, high-efficiency and high-throughput operation and to ensure buy-in from critical partners and funders with limited influence on the overall aims of the mission. For infrastructure OSPs, the quality and scope of potential applications of the infrastructure as well as their ability to build an appropriate portfolio of activities over time are likely to matter for their success. For clubs, key challenges may involve making sure that the moving parts fit together as planned and that “safety valves” do not undermine the OSP’s open principles. Finally, for hub OSPs, ensuring intermediation that supports the development of collaborations and projects as well as ensuring that distributed activities in distinct projects adhere to key principles and the overall aims is likely to be crucial.

More research is also needed to explore how different archetypes and designs of OSPs may affect the open principles upon which they stand. Though research has long been attributed with Mertonian ideals of open science, this is not always the case in practice (Hosseini et al. 2022; Macfarlane 2023). Indeed, multiple barriers to open science have yet to be addressed (Ross-Hellauer et al. 2022; Pence 2023). Openness must therefore be actively supported throughout the research process (Levin & Leonelli 2017), and appropriate incentives for open sharing of methods, data and results must be put in place (Manco 2022). Similarly, openness of collaborations comes in many different shades (Masum et al. 2013; Stevens et al. 2016) also in the OSPs we have studied. There is therefore a need for further exploration of the mechanisms OSPs use to mediate openness.

Ultimately, we hope our exploratory analysis will offer an informed starting point for researchers who are interested in a better understanding of OSPs and their role in the advancement of science, industry and society, and an inspiration for practitioners on how to ensure that the intended ends of the OSPs they design and manage are supported by the means chosen.