Business model innovation in circular start-ups: Overcoming barriers in the circular plastics economy

Circular business models

A business model is a holistic description of how a firm does business (Bocken et al., 2014; Zott and Amit, 2010), acting as a blueprint for the underlying core logic and strategic choices (Shafer et al., 2005; Teece, 2010). A business model has three basic components: (1) the value proposition - what the firm offers its target customers and why they are willing to pay for it; (2) the value creation and delivery system - how the firm creates and delivers its value proposition, such as its resources, capabilities, activity system and value chain and (3) the value capture system - how the firm generates revenue and makes a profit (Richardson, 2008). Traditional business model scholarship has explored the creation and capture of economic value for a firm and its value chain (Teece, 2010; Zott et al., 2011). Several scholars have expanded this to a broader view of value and stakeholders (social and environmental value beyond a firm’s boundaries) under the sustainable business model (SBM) concept (Bocken et al., 2014; Boons and Lüdeke-Freund, 2013). However, designing SBMs to capture economic value while creating environmental and social benefits is challenging (Schaltegger et al., 2012), and the literature does not provide clear directions (Henry et al., 2020; Stubbs and Cocklin, 2008).

CBMs are considered a more concrete sub-category of sustainable business models (Bocken et al., 2014; Guldmann and Huulgaard, 2020) that integrate CE elements (Henry et al., 2020; van Keulen and Kirchherr, 2021). This means designing the basic components of a business model so that it slows down, closes, narrows, intensifies and/or dematerialises resource loops (Bocken et al., 2016; Geissdoerfer et al., 2018). Materials thus, remain in use as long as possible rendering the ‘end-of-life’ concept obsolete and minimising excessive production, consumption, resource extraction and waste generation (Henry et al., 2020; Kirchherr et al., 2017). In theory, the business logic shifts from customary one-off sales to logics that continually create, deliver and capture (smaller increments of) value over time (Guldmann and Huulgaard, 2020). Consequently, firms benefit in several ways: reduced resource dependence, less volatile revenue, improved reputation, higher profitability and lower churn (Bocken et al., 2016; Vermunt et al., 2019; Werning and Spinler, 2020).

Role of start-ups in the CE transition

Businesses and business model innovation has a pivotal role in advancing the transition to a CE (Lewandowski, 2016). The premise is that if CBMs are scaled up and become mainstream, we can reap the full economic, environmental and social benefits of the CE (Vermunt et al., 2019). Accordingly, research on CBMs has significantly expanded in recent years (Centobelli et al., 2020; Ferasso et al., 2020; Hina et al., 2022). However, despite the support, interest and widely purported benefits of circularity for firms, the adoption and implementation of CBMs remain low (Vermunt et al., 2019; Werning and Spinler, 2020). Powerful incumbents understandably adhere to the existing linear regime (Milios, 2021), favouring circular solutions that are arguably more incremental, easier to implement and well-aligned with their existing assets (Henry et al., 2020; Hockerts and Wüstenhagen, 2010). Unfortunately, without more drastic, radical action from influential incumbents, it will be challenging to steer markets towards circularity and advance a circular transition.

The sustainable entrepreneurship literature suggests that one way of influencing incumbents to favour an industry’s sustainable transformation is to focus on start-ups. Unlike existing firms, CSUs and other born-sustainable pioneering firms (Todeschini et al., 2017: 765) are conceptualised with circular elements at their core from the outset. According to Henry et al. (2020: 2), CSUs are ‘new, independent and active companies pursuing a CBM’. Defined by an alignment of their business model with CE principles (Ostermann et al., 2021b), this term closely relates to ‘circular-born’ firms (Colucci and Vecchi, 2021: 870), in turn connected to the broader umbrella of born-green or born-sustainable firms. Such firms build on sustainable, collaborative and innovative products and processes from the outset (Ostermann et al., 2021a; Todeschini et al., 2017). They typically have a pioneering spirit, operate in an uncertain and risky environment, and find it difficult to balance profitability with environmental goals (Ostermann et al., 2021a). Unencumbered by lock-ins, they are more likely to experiment with and embrace higher levels of circularity and radical innovations than incumbents (Henry et al., 2020; Hockerts and Wüstenhagen, 2010).

Academia acknowledges and yet is confused by the extent of overlap between sustainability and circularity concepts (Asgari and Asgari, 2021; Blum et al., 2020; Geissdoerfer et al., 2017), reflected in subtle differences between sustainable and circular start-ups. In our study, we view CSUs as a subset of sustainable start-ups (Bocken et al., 2014; Geissdoerfer et al., 2017) as they adopt a more narrowly defined set of approaches to sustainable value creation (Cullen and De Angelis, 2021; Henry et al., 2023). These include R-strategies such as rejuvenating nature, refusing or reducing material use, reusing, refurbishing or remanufacturing products and recycling materials into resources (Henry et al., 2020).

Barriers to CBMs

As CBMs operate under a different value creation logic (Fehrer and Wieland, 2021), circular businesses face different barriers than linear businesses (Tura et al., 2019). There have been substantial efforts to identify barriers to CBMs in general. In a recent review, Hina et al. (2022) found and analysed 126 studies on the drivers of and barriers to CE business models between 2016 and 2021 and new papers are continually emerging (Galvão et al., 2022; Geissdoerfer et al., 2022). These, and earlier works, have investigated CBM barriers in varying industries, regions and methodologies (literature reviews, case studies, surveys). However, despite this breadth of scholarship, studies that focus specifically on circular start-up barriers remain scarce, with two notable recent exceptions: van Opstal and Borm’s (2023) survey on CSUs in Belgium found that they faced significantly different barriers depending on their circular strategies, whereas Von Kolpinski et al.’s (2023) case study of young and small-scale firms with sustainable CBMs in Germany found that they tend to overcome barriers through their internal strengths. Nevertheless, barriers to concepts closely related to CSUs, such as born-circular firms or born-sustainable start-ups, have been investigated by Briguglio et al. (2021), Ostermann et al. (2021a) and Todeschini et al. (2017). They noted several prominent barriers for start-ups: the need for greater consumer awareness, lack of consumer demand, obstructive policy, insufficient financial support, inadequate alignment of values along the supply chain and the difficulty operating a circular logic in a linear system.

In our view, the plastics context is also under-researched regarding CBMs, with the exception of Bening et al. (2021) on packaging; Dijkstra et al. (2020) on plastics management; and Paletta et al. (2019) on manufacturing plastics; note that none of these focused on CSUs. Scholars, including Cantú et al. (2021), Kirchherr et al. (2018) and Werning and Spinler (2020), highlight the importance of context as an influence on the type of barriers that firms face thus, reinforcing calls from other authors for sector-specific explorations of CBM barriers (van Keulen and Kirchherr, 2021; Vermunt et al., 2019). Tura et al. (2019) concluded that circular business barriers and drivers are inevitably context-specific and cannot simply be transferred from one context to another. Echoing this view are Bassi and Dias (2019), Briguglio et al. (2021) and Demirel and Danisman (2019), which motivated our investigation into CSUs in plastics – a particular context that deserves research attention but remains underexplored.

Furthermore, studies on CBM barriers, except for Galvão et al. (2022), have often (perhaps unintentionally) portrayed them as almost-insurmountable blockades. One potential reason is that knowledge on how to overcome these barriers is scarce (Bressanelli et al., 2019), understandably so given the agreement among scholars that there is no panacea for addressing and overcoming all these challenges (Galvão et al., 2022: 1505; Werning and Spinler, 2020: 3). Nonetheless, we believe there is potential to explore how these barriers are addressed in practice (cf. von Kolpinski et al., 2023). By focusing on CSUs as engines of innovation, we expect them to be naturally agile and equipped to at least attempt to overcome their circular business modelling barriers. To this end, we employ the concept of SFs, the minimum requirements to ‘prepare a company for competition’ in a particular industry, as used by Ketelhöhn (1998: 335). For a new context such as the CE with its specific set of barriers, we see SFs as ‘entry requirements’ to overcome, or at least address these barriers in order to compete in the world of circular plastics, with the implicit notion that competitiveness is linked to navigating CE barriers successfully.

Case description: Plastics in the Netherlands

Materials that transform society – such as stone, bronze and iron – inspire the names of historical epochs, and the one we now live in is the plastic age (Schirmeister and Mülhaupt, 2022). Plastic has left an indelible mark on the planet: it is one of the geological markers of the Anthropocene (Zalasiewicz et al., 2016). As a versatile material that can be produced at an attractive price/performance ratio, plastic has become essential for almost any kind of consumer product (Bening et al., 2021; Bucknall, 2020; Schirmeister and Mülhaupt, 2022). Its design flexibility in terms of properties and applications outclasses other materials, as it can be tailored to solve a particular problem in a way that outperforms all others. As a resource-efficient, lightweight and low carbon footprint material, plastic can be designed to enable sustainability in areas such as mobility (lightweight transport), consumer goods (protecting contents and extending lifetimes), agriculture (enhancing output while reducing the need for pesticides) and the built environment (improving energy efficiency in buildings) (Schirmeister and Mülhaupt, 2022). Indeed, plastics have been continuously (re)invented for decades to help advance other fields (e.g. electrification and communication technologies), enable our modern lifestyles (widespread and secure access to energy, water, food and medication) and reshape the world, and they will continue to do so (Payne and Jones, 2021; Schirmeister and Mülhaupt, 2022).

Plastic production involves several processes in a long and complex value chain (see the blue arrows in Figure 1): from fossil fuel feedstocks, monomer building blocks are produced, polymerised into flakes or pellets, and further processed and converted into plastic intermediates. These become the plastic products we interact with and which we can potentially reuse, repair, refurbish or repurpose during their lifetime. Eventually, these products are disposed of, collected and managed as waste. CSUs offer solutions to divert these materials from landfill or incineration and return them to the value chain. One pathway is to valorise waste (the red arrows in Figure 1) through various degrees of recycling (Hahladakis and Iacovidou, 2019), resulting in alternative feedstock, monomer, polymer and intermediate material streams of varying quality and potential applications. While recycling improves the planetary footprint of plastics (Bening et al., 2021), alone it is insufficient to achieve sustainability and would eventually overshoot planetary boundaries (Bachmann et al., 2023). In addition, Bachmann et al. (2023) recommend switching feedstocks not only to plastic waste but also to renewable resources. This pathway occurs in parallel (the green arrows in Figure 1), where start-ups offer a bio-based, renewable substitute to fossil fuels, reducing emissions and even allowing for biodegradability in some instances.

OPEN IN VIEWER

Figure 1.

Material, knowledge, and service flows of circular start-ups creating new paths in the linear plastics value chain.

While Figure 1 is simplified to show a single, linear flow of homogeneous ‘plastic’ material, myriad types of plastic can be tailored to equally diverse applications. This results in waste streams of diverse properties, recyclability and ultimately more complex (often inadequate) end-of-life management (Schirmeister and Mülhaupt, 2022). In 2022, only 5% of mixed post-consumer plastic waste was recycled, whereas 38% was landfilled and 57% was sent for energy recovery (PlasticsEurope, 2022). Coupled with a heavy dependence on fossil inputs (Bening et al., 2021), unsurprisingly, plastics have been linked to high emissions, pollution and environmental harm (Johansen et al., 2022; Nielsen et al., 2020; Singh and Ordoñez, 2016). However, because this value chain has evolved to become highly efficient under a linear system, it is difficult to reconfigure; in fact, volumes are projected to grow further (Bauer and Fontenit, 2021; Geyer et al., 2017; Jambeck et al., 2015). Clearly, the plastics industry faces immense pressure to take action and innovate its business models.

We investigated the plastics value chain in a circular innovation hotspot: the Netherlands. The Dutch have long aimed to be a pioneer in recycling, a role model for circular business and a nation that will be fully circular by 2050 (Calisto Friant et al., 2022; McCarville, 2019). Specifically, the Dutch CE Action Plan identifies plastics as a strategic sector (Calisto Friant et al., 2022; van Leeuwen et al., 2018). However, despite considerable progress, the Netherlands is still far from achieving its goal: current CE practices and policies are doing little to transform the country’s linear production and consumption of plastics, according to Calisto Friant et al. (2022). As the EU’s largest natural gas producer and with a booming chemical industry (McCarville, 2019), the Netherlands is a key global player whose economic and geopolitical interests are intricately locked into the plastic industry (Calisto Friant et al., 2022). We set our case study within the paradoxical context of a country leading circularity initiatives yet, failing to surpass linearity because it gives us the opportunity to uncover complex and interrelated barriers to implementing CBMs, potentially radical innovations by CSUs to navigate this complexity and a better understanding of how start-ups contribute to the circular plastics transition. This also responds to calls for in-depth field studies on circularity in complex value chains (van Keulen and Kirchherr, 2021) and circular plastics aspects beyond the technical realm (Calisto Friant et al., 2022).

Data collection

We collected data using diverse selection criteria (Seawright and Gerring, 2008), involving a wide range of start-ups to explore their potentially different CBM configurations across the plastics value chain (Figure 1). For our study, we divided the value chain into three groups (Figure 2): alternative raw material providers, alternative product providers and knowledge and service providers. Alternative raw material providers are CSUs that provide more circular substitutes for fossil-based plastic raw materials and feedstocks. These are either bio-based or waste-based in our sample. Alternative product providers offer more circular plastic in lieu of traditional products. Knowledge and service providers are CSUs offering knowledge and services for plastics circularity.

OPEN IN VIEWER

Figure 2.

Basic information on respondents.

Through collaboration with an industry-affiliated CE researcher, we recruited 12 start-ups hosted at a circular hub in the Netherlands. In addition, we contacted start-ups outside the hub through their websites and LinkedIn, and eight agreed to participate. These firms were selected based on size, fit in the plastics value chain, use of circular strategies in their business model, regional affiliation and relative success. In other words, they had to self-identify or be identified by third parties as start-ups; provide either raw, intermediate or waste materials for the plastics value chain, plastic-dominant products or products/services that complement or boost circularity; employ strategies that slow, close, narrow, regenerate or dematerialise loops (Geissdoerfer et al., 2018; Konietzko et al., 2020a); and be currently operating in the Netherlands (i.e. surviving or on the way to greater success). These criteria allowed us to investigate the barriers for start-ups at various parts of the Dutch plastics value chain and how start-ups overcome them.

We conducted 20 semi-structured interviews with start-up (co-)founders, executives and top managers between May and September 2021. Semi-structured interviews allow respondents more freedom in answering questions and offer unexpected insights while retaining sufficient structure to elicit data that answers the research question (Saunders et al., 2009). Each interview lasted between 41 and 74 min and covered the history and motivation for creating the start-up; the value proposition, value creation and delivery, and value capture elements of the business model; the challenges in developing and implementing it; actions they (plan to) take in response and their future vision for the firm. One researcher took notes, whereas another led the inquiry. Aside from the core 19.5 h of interviews, we approached experts to verify our initial findings and obtain further insight into parts of the value chain where we had little data. We conducted 2 h of expert interviews with (1) a Dutch circularity hub business manager familiar with the advanced materials trade, (2) a recycling executive who supplies start-ups with waste for their products and (3) two senior scientists at a large polymer producer who are familiar with the plastics value chain. All interviews were undertaken online due to the COVID-19 pandemic; each was recorded and transcribed using Microsoft Teams and Word with the interviewee’s informed consent. The interview recordings, transcripts and notes were stored in accordance with ethical review board guidelines. Furthermore, to establish construct validity and improve credibility (Yin, 2003), we examined archival data before and after interviews, for example websites, press releases and other material shared by the interviewees.

Data analysis

Given the exploratory nature of this work, we employed provisional coding as a jump-off point for our data analysis: here, a ‘start-list’ (Saldaña, 2009: 118) of provisional codes derived from the literature facilitated our coding of CSU business model elements and barriers. However, for SFs – a less explored topic for which provisional codes could not be made – we used structural coding (Saldaña, 2009): segments of data on CSU actions to address barriers were tagged with a conceptual phrase and clustered around five emerging topics. These first-cycle coding strategies laid the foundation for subsequent second-cycle coding (Saldaña, 2009) (see Appendix 3 for a coding summary). Beginning with the first-cycle codes, we deconstructed them into more nuanced sub-codes and visualised their interrelationships on a conceptual map. Going back and forth between the visualisations, transcripts and archival data enabled us to both (re)familiarise ourselves with the data as well as identify the most relevant interrelationships. In parallel, this process allowed us to combine codes into themes, review themes and create initial clusters of themes (Braun and Clarke, 2006): five sets of interrelated barriers, four SFs and several strategies-in-use for each success factor. As some of their elements were more closely related to others, we consequently reshuffled themes for our final configuration. Coding was done in NVivo 12 to facilitate collaboration within the research team. All interviews were double-coded to stimulate discussion, share alternative perspectives and address potential bias. Moreover, several interviews were triple-coded to ensure the research team commonly understood the code definitions (O’Connor and Joffe, 2020). The coding process is presented in Appendix 4.

High capital requirement but unconvincing business case deters investors

We begin our discussion of CSU barriers with a recurring theme during the interviews: the prominence of (new) technologies and novel materials. Start-ups said they often responded to CE opportunities through technocentric solutions. This included processes for turning waste into high-quality raw materials and feedstocks (thermal solvolysis, depolymerisation, pyrolysis), novel materials (bio-based materials and high-performance polymers) and 3D printing. Typically, these solutions are expensive and not yet proven on a large scale (R1–5, R7–13, R15, R17, R20–21). However, because these solutions are positioned as substitutes to virgin plastic (raw materials and products), potential customers could only be convinced to switch if they are guaranteed an ample and stable supply: ‘You need a full-scale plant, the first one [so] that you can show that you established yourself on the market as a supplier. You have to build trust in the market; you cannot build it with a proof of concept, right? You can only do it with a running operation’ (R1).

This set CSUs on a path of first having to prove their solution, technologically and at scale, which often takes a lot of time and effort, for example going from laboratory to pilot and then to demo plant scales, with several rounds of product testing, customer validation and funding acquisition, while still not producing substantial quantities of the product – a risk for investors: ‘. . .you need millions of funding without actually producing any product, and that is some disconnection between the finance guys and R&D guys. And it’s very difficult to overcome it’ (R7).

Beyond the amount of funding they require, circular plastic start-ups are also limited by a long payback period: ‘When you develop something from scratch like we do, we have an extremely long development cycle. It’s not long for us. . . but it’s extremely long for the established market, including a venture capital market’ (R7). This made it hard for CSUs to convince investors to fund their venture. Respondent 10 laments how ‘a lot of investors want to make quick money’ and ‘get cash the next year’. This forces CSUs to find ‘very patient investors’ (R10) with ‘deep pockets’ (R12) who were willing to invest for many years before making a profit or pivot their business to something else that could bring in cash faster (R10–11).

The combination of unproven technology, high capital requirement and long payback period deters investors from funding circular plastics ventures because the associated risk is ‘just too high. . . because so many things actually could, on paper, go wrong’ (R20). In most cases, this risk is framed in financial terms: ‘. . .if it doesn’t work out, do I feel OK about putting money into that project?’ (R15). Getting investors to trust CSUs is difficult because CBMs have neither been ‘proven’ nor are mainstream. Respondent 20 perceives this as a barrier from linear mindsets:

It’s very difficult to finance a business model which has not been proven. . . And circular business models are not proven or are not mainstream. . . So I think that what commercial institutions look at. . . [is] your business case, and it should be a viable business case or a positive business case in a linear economy. (R20)

Low credibility makes it hard to enter and influence complex markets and industries

For similar reasons, CSUs also found it hard to convince potential customers of their CE-oriented business case (R2, R6–7, R12, R15). Among alternative raw material providers, big customers did not find small-scale offers (like from CSUs in the early stages) attractive because of a mismatch with their large-scale operations (R2, R10, R20). On the technical side, a mismatch could come from customer perceptions of a shift to circular materials as a reason for changing ‘everything’ (R7). Fundamentally, it is a matter of risk:

They want to understand how they will eliminate risk, and there is no good solution for that. . . In many cases. . . companies decide in favour of already existing technologies or not doing anything because otherwise, they cannot validate the risk they have (R7).

This distrust of novel solutions also applies to alternative product providers: ‘Even if you end up having a great product, if the market is not yet ready for it, then you have to tackle that challenge as well’ (R12). Respondent 6 found the firefighting equipment market ‘very traditional and conservative’, ‘used to dealing with the same company for dozens of years’, with such ‘a long history of relationship and building up trust’ between customers and suppliers that it didn’t matter ‘whatever brilliant product’ you had. Echoing these words, Respondent 3 noted that the construction market is ‘highly conservative’ and hard to enter: especially due to a lack of history, customers have no prior experience with a novel product. In the ‘traditional’ and ‘hesitant’ carpet and furniture sector, Respondent 12 was forced to create proof points to influence the market.

Moreover, CSUs lamented the linear mindset of some potential customers: despite generating a lot of waste that CSUs could resolve, these customers still opted for cheaper but more polluting material (R8) or accepted greenwashing as the way forward (R17). Getting firms to switch to circular materials at a price close to virgin was one of their biggest challenges (R1): it was like ‘fighting against an established economy who ten years ago was not looking for recycled content’ (R10). As a recourse, many CSUs work to bring down costs and prices ‘to make it interesting’ (R21) for customers. However, bringing prices below virgin alternatives is not always feasible because production volume is insufficient to achieve ‘cost efficiency’ (R8) – which Respondent 8 believes applies to many circular products. To compensate, they sell with a premium: ‘It costs more than virgin plastic. . . But we are selling it with the premium mode, not because we would like to earn more money. . . [but] because we cannot. . . achieve a better cost structure [at current production volumes]’ (R8).

Furthermore, CSUs faced additional complications from the interconnectedness of the plastics value chain. For several alternative raw material providers (R1–2, R7, R9–10, R15), selling products involved convincing not only their direct customers but also those customers further downstream who might interact with products made by their direct customers, for example, waste suppliers, converters, brand owners and end-users: ‘. . .both the converter and the brand owner need to agree to sell your material. It’s very difficult for a small company to get in there’ (R15). This was not always straightforward because the specifications of various value chain actors were not always aligned with each other or with the CSU’s: ‘There’s a two-fold customer because the brand owners have different priorities than those that are converting, which means that things that we have to do for the customer [become] slightly. . . difficult. What is nice for one is too expensive for the other and vice versa, so that makes our lives difficult’ (R15). Even for alternative product providers, complexity in market segments is a challenge. Working with medical devices, Respondent 5 had to be able to distinguish and consider the user (patients), payer (insurance companies) and decision-maker (doctors). Despite the value of their product for patients, Respondent 5 encountered immense resistance from providers because they benefit less from the innovation. Respondent 12 compares this struggle with ‘building the mountain’, ‘even more so when you enter markets that [are] still developing’, referring to the associated work as ‘massaging’ and ‘preparing the market for your competitors’.

The value chain has to reorganise and collaborate more closely: That is no easy matter

CSUs also faced challenges from being part of a budding circular plastics industry – that is complex, expensive and requires closer collaboration in the value chain to close the material loop:

We’ve been looking at the whole circular economy model around waste and how waste can be processed into new products or new materials. . . you cannot do it on your own. It is so complex, so expensive, and probably that’s one of the reasons why it’s not in place (R16).

With the market for high-quality recycled plastics still in its infancy, a critical bottleneck for CSUs was a lack of stable, good quality and suitably priced waste feedstock (R7, R15), whose availability was ‘much smaller than the potential market’ (R7). Even for the same polymer (e.g. PET plastic), certain applications and technologies required uniform input streams: for instance, food application PET waste should not be mixed with that used for non-food applications (R15). This presented supply challenges because sorting and collection systems were not always compatible with CSU technologies. Respondent 7 found that securing feedstock was the most important activity, which required ‘extended network and negotiations and knowledge of the market: where to find the right material and the right quality’. Without proper feedstock, CSUs could not make products according to customer specifications or incurred additional costs to do so (R15). Ultimately, this placed plastic recycling CSUs at a disadvantage: the success of their solutions was at the mercy of upstream suppliers who did not always supply the requisite quantity or quality. At the same time, they shouldered additional costs to process complex inputs without being able to pass these costs downstream (R1, R6, R8–9, R15, R17).

With little power and status, start-ups in circular plastics relied on collaboration with other value chain actors and industries in order to improve their offering (R6–7, R9, R11, R15, R20): ‘It is really unique – the whole supply chain – because you’re so much in a development business. We all need each other to learn and develop together’ (R9). Respondent 10 sums this up as having to ‘work the whole value chain: the waste recyclers and the packaging industry and the plastics producers and FMCG brands’. Among alternative raw material providers, it is vital to develop in parallel with target customers: for instance, while Firm 9 develops their technology, their customers (petrochemical firms) also ‘have to learn and find out what can they change in their process, [to be] able to process more and more converted plastic’. Meanwhile, among alternative product providers, Respondent 12 highlights that their ability to redesign products for circularity will only work if the recycling industry also develops the capacity to handle them.

Ultimately, several CSUs find collaboration limited (R1, R11, R19): it is ‘difficult to work in a chain in an open way’ and ‘it’s also a matter of power’ (R19). For example, patents and non-disclosure agreements to protect the competitive advantage brought about by novel technology can hinder speeding up the development of a circular materials market and corresponding technologies (R7, R9).

CSUs have limited resources and no ‘knowledge bibles’ for improving their business model

The importance of resources cannot be understated for CSUs: ‘as long as you have funding, you can deal with all other bears on the road’ (R12). Otherwise, you might ‘feel like a dog in a circus. You have to jump through the hoops every time’ (R10). With limited funding, CSUs might find themselves lacking decent equipment (R4, R8), poorly equipped to enter established markets (R6), constrained to small-scale production (R5), unable to develop ideas in parallel (R11–12, R10, R20) or with insufficient personnel to accelerate growth (R10, R12, R14). Moreover, they might be forced to decide their business models early on (R11, R15–16), which could also limit their desired or potential socio-environmental impact (R5, R10–11, R20); for example, it is not feasible to enforce ‘a complete 360° business model from the start because that is very difficult’ (R20). Most importantly, without funding, CSUs have little leeway to experiment and make mistakes: ‘In the sense of funds and all kinds of things, you have very little space to make those mistakes nowadays, and that makes it pretty difficult’ (R1).

While start-ups obviously need to obtain more funding, searching for it actually competes with other important activities: ‘Even to go through the process of trying to find the financing is very difficult. . . You have to give a lot of information, and it takes a lot of time which you can also spend differently’ (R20). Respondent 12 also feels this time constraint: ‘The clock speed is so high. . . you don’t really sit down. . . and have the time to analyse’. With so many activities competing for attention, how should CSUs allocate their time and where do they make compromises? Here lies another challenge for CSUs: there is no ‘knowledge bible’ they can follow to know how and how much circularity to include in developing a viable CBM (R3, R5, R9, R11–13, R15, R17, R19–20).

To address this, Firm 19 helped other actors with circular innovation. They found that many firms struggled with its interdisciplinary nature: ‘they were all struggling with circular or sustainable innovations because. . . you need knowledge from the social side, from the ecological side, from the economic side, from the technical side’ (R19). For instance, they struggled to develop revenue models for products in a cycle, where the linear model of making something and selling it once is not enough (R19; also with R11, R20). Indeed, Respondent 20 admitted that it was still difficult to finalise their business model and position in the value chain: ‘Am I someone who’s building factories there, who’s a technology provider? Or am I a project developer and an implementer who brings it all together? Or am I a consultant?’. Respondent 19 observed how firms struggle to measure the positive impact of using alternative models such as leasing and product as a service. This is echoed by Respondent 5, who found it difficult to quantify the impact of including take-back schemes: ‘We made this recyclable. Now we’re trying to make it circular. . . it’s unclear, the specific [. . .] cost impact it will have’ (R5).

Another challenge is the lack of role models (circular firms) to emulate because most CSUs are frontrunners (R12, R17). Respondent 12 notes that firms are moving in parallel without ‘articles’ about how to design and implement CBMs successfully. Not knowing how to best integrate circularity into the business, CSUs have to develop knowledge themselves by gathering a lot of information (R1, R4) or otherwise rely on assumptions on both the technology and business model (R1, R11, R20–21):

There are so many factors that you have to take into account: the technology, how long it will take to develop, how much it will cost and all kinds of things. A mistake is easily made. . . assumptions are the mother of all [screw]-ups. And it’s true: if you don’t have proven data on everything, you have to assume things, but the quality of your assumptions, whether it is technological or business-wise, determines how well you will perform. (R1)

Institutional elements are not yet geared towards shifting markets to circularity

Institutional elements that could help CSUs overcome barriers have not yet been fully developed for circularity. For instance, a key impediment to recycling plastics is a prevailing misconception: ‘in the discussion about circularity, plastic has a very. . . negative value’ (R13) associated with ‘a lot of low-quality connotations’ (R20). This poor perception applies to both plastic waste and recycled plastics. From a historical perspective, ‘an important argument for using recycled plastic was the lower price, and a lot of products were developed for recycled material for the [lower] quality that they could produce’ (R1). The novel technologies of circular plastic start-ups now allow recycled plastics to be of virgin-like quality but at a higher cost. Until recently, the market was ‘not ready to pay for the extra costs’ for circularity (R1), and in some cases, virgin plastic is still cheaper than recycled plastics (R19). The problem could be a broader aversion against paying for circularity – and in a wider sense, sustainability in general (R2–3, R5, R7, R21): ‘if they don’t feel it in their wallet, people are willing to become more sustainable’ (R2). While a societal shift is happening in favour of sustainability, Respondent 2 believes this applies to ‘outliers’ and CSUs cannot build a business around them.

CSUs recognise the vital role of supportive regulations for launching circular products and services. In Respondent 21’s words, ‘if they try to launch a special service or product. . . you have to have good regulations. . . [if] you have to deal with [unfavourable] regulations, then it can be quite a bit harder’. In some cases, laws, rules and regulations can be ‘in the way’ (R15; also R4), conflict across borders (R5), or be inadequate, such as for product labelling (R14), water waste (R16) and deposit-return schemes (R20). Similarly, standards and certifications may be additional requirements to fulfil before novel products for the CE can be launched in the market (R4 mentions material processing, R6 protective gear design, R9 waste declarations). Some standards could even deter greater circularity: ‘there is a European standard which regulates the properties, design, quality, etc. of firefighting gear, but it was made without thinking about sustainability and ecological problems’ (R6).

A few support mechanisms that CSUs would like to see in place are lacking, such as subsidies for biodegradable materials and government funding (R4, R8, R17). These factors sometimes have regional and cultural elements (R2–4, R8, R16). Respondent 4 found it hard to get funding for bio-based materials in Russia: with an economy largely based on oil, industrial and economic interests did not favour disruptive substitutes. In America, convincing investors (R2) and finding early adopters was easier (R16). In general, the geographical region could be a factor: for Firm 8, ‘it’s impossible to sell this product in developing countries. Only developed countries are ready for this product, I’m afraid. . . it’s too early to go to developing markets with CE start-ups’ (R8).

Circular value proposition design

CVPD is the ability of CSUs to design compelling value propositions that combine circularity with customer needs and wants. This success factor stems from CSUs needing to create offerings that embody a delicate balance between circularity (recyclability, take-back schemes, bio-based materials, reduced material footprints, etc.) and other customer demands (premium quality, convenience, affordability). Although any start-up is expected to have highly-compelling value propositions (R21), CSUs must achieve the same while burdened with often-costlier circular value for which not all customers are willing to pay a premium: ‘[if] you’re doing something circular, you’re actually adding additional costs to make it circular, right? So we kind of try to figure out how do we not do that and still be circular’ (R5). In the circular plastics context, we see this success factor being applied as CVPD strategies-in-use for (1) complementing circularity with other attractive values or (2) reducing costs and prices until circularity becomes a differentiator.

Market sensitivity

MS is the ability of CSUs to sense the current market trajectory and react accordingly despite the uncertainty of the context in which they are embedded. The CSUs we interviewed operated in an industry in transition, where identifying customers could initially be problematic, customer sentiments and preferences could change over time and market evolution is not easy to predict. They therefore, need to gather data from a wide range of sources, make sense of it and act on often incomplete information to (re)design their business model and craft strategies for initial success and future growth. In the circular plastics context, we observed this success factor manifesting as MS strategies-in-use for sensing strategic information such as market trends and customer sentiments and for acting on sensed information to seize opportunities, for example, by timing your entry correctly and further developing growth strategies.

Networking prowess

NP is the ability of CSUs to evaluate their resource base and expand it through the external network, as well as influence and leverage that network to their advantage, for example, to access resources and capabilities outside the organisation or increase the number of opportunities available. CSUs need this success factor to overcome or manage resource constraints, offer their customers holistic circular solutions and navigate the complex markets, industries and value chains in which they operate. We observed this factor being applied in two NP strategies: assess the need to recruit partners and condition yourself to network, and expand your resource base and strengthen your business model by leveraging from your network.

Circular ambidexterity

CA is the aptitude of CSUs to switch between or combine linear and circular logics to their advantage. CSUs typically operate in an environment characterised by an ‘emotional rollercoaster of. . . ups and downs’, with ‘many frustrations on things not working or taking longer’ and ‘people not getting what you want’ (R12). Only purpose-driven individuals and teams can succeed on this path: ‘It is important that someone really has the inner drive. . . that you really want to do something circular and not only [do it] as a window dressing, because [on] this path, we are still in an early phase. . . and we see that many initiatives just [fail]. . .’ (R1). However, committing to something truly circular does not translate to naïve idealism in practice. Working in the circular plastics context requires CSUs to strike a careful balance between fully implementing circularity and playing the still largely-linear ‘rules of the game’. In essence, this means critically evaluating their desired circularity impact against their continued existence: ‘. . .you can do all those “saving of the planet” if you want, but if you don’t have customers, then you don’t have investors, then you’re not going to exist’ (R16). We observed this success factor manifesting as two CA strategies-in-use: (1) embracing new circularity mindsets and (2) balancing circularity with current linear realities.

Theoretical implications

We conducted this case study in response to the growing interest in CSUs and their role in advancing the circular transition. Previous research has posited how CSUs would play a key role in the circular transformation of various industries (Henry et al., 2020; Hockerts and Wüstenhagen, 2010). This is based on the assumption that CSUs, as newly created and often, born-circular entities, would have greater freedom to experiment with, develop and implement radical CBMs. We observed this does not seem to be the case in circular plastics – many start-ups in our study contribute to and focus on plastics recycling (cf. Dijkstra et al., 2020), a strategy further below the waste hierarchy (Bening et al., 2021) that requires less drastic shifts from prevailing business models (Henry et al., 2020). While recycling is necessary for the plastics industry to reduce its planetary footprint (Bachmann et al., 2023; Bening et al., 2021), it is insufficient to achieve a sustainable circular plastics economy (Calisto Friant et al., 2022). The CSUs in our study were developing businesses that would displace fossil fuels with arguably cleaner alternatives. Yet, few attempted to fundamentally change prevailing production and consumption patterns through strategies such as reuse, repair or product-service systems. Only a minority of start-ups could go beyond dealing with the waste problem and address such societal issues as accessibility and distributive justice, global health and well-being.

This highlights the limitations to the purposeful attitude of CSUs towards advancing a circular transition despite the agency implied by the SFs concept. A potential explanation is that circularity is a property of a system (Fehrer and Wieland, 2021), and the transition towards a CE requires an exchange of ideas and knowledge between various actors, such as incumbents, start-ups, suppliers and consumers (Adams et al., 2016; Ceschin and Gaziulusoy, 2016). On the one hand, we do witness an exchange of ideas and knowledge among the CSUs we studied. For example, NP is a critical success factor for CSUs to expand their resource base through partnerships that can provide complementary resources for them to deliver on their intended value proposition. This network includes entities outside the direct value chain, such as government, accelerators, trade associations, standardisation organisations and universities.

Nonetheless, for CSUs concentrating on alternative raw materials and products, apparently, the major collaborations are still with direct suppliers and customers, highlighting a firm-centric focus on value proposition, value creation and delivery and value capture. Thus, other actors inside and outside the value chain, as well as end customers, are seen as ‘barriers’ to overcome, not as entities for jointly creating value. Consequently, fewer radical innovations have emerged, as previous studies found: when firms decide to establish bilateral collaborations with suppliers or customers, they generate incremental innovations rather than the more comprehensive innovations required for radical CBMs (Brown et al., 2020; Velter et al., 2022).

Consequently, at the ecosystem level, the focus needs to be on the multilateral interrelationships between a firm and other actors that co-create and share value together (Blomsma, 2018; Bocken and Geradts, 2020; Boons and Lüdeke-Freund, 2013). Also, this implies that partner firm business models are just as critical to consider as the focal firm’s (Adner, 2017). Implementing a more radical CBM, therefore, requires not just one but multiple actors to change what they do (Konietzko et al., 2020b). Often, these ecosystem-level processes are beyond the control of individual start-up firms and create a certain degree of unpredictability and dependency on other actors (Kanda et al., 2021). This prevents them from enacting their radical innovator role in the transformation of industries, in turn slowing down the transition of businesses and society to a CE.

In summary, we observe that start-ups in circular plastics find themselves focusing on the implementation of more incremental CBMs despite the theoretical argument that CSUs would have greater freedom to experiment with, develop and implement radical CBMs (Henry et al., 2020; Hockerts and Wüstenhagen, 2010). Our study thus deviates from and brings nuance to Henry et al.’s (2020) novel contribution where start-ups pursue higher-level CE strategies than corporates. We also expand Hockerts and Wüstenhagen’s (2010) view where the freedom of start-ups to be innovative and radical with their business models advances an industry’s sustainability transformation. What we witness in the plastics industry is that start-ups do hold agency over their business model elements to a certain extent yet are unable (or perhaps do not even want) to advance the circular transition more radically (cf. Hofmann, 2019 for the role of firm agency in implementing CE). This is a second important contribution suggesting that several exogenous influencing factors still put a ceiling on the role of agency to drive the circular transition more radically.

Indeed, Aarikka-Stenroos et al. (2021) noted that local and regional institutional settings and social institutions shape and constrain the ability of individual actors (e.g. customers, firms and governments) to act for the CE. Consequently, to advance the circular transition, CSUs need crucial external support and engagement at a meso-level with developing their business models (e.g. increase collaboration within and across industries and value chains) and at a macro-level (e.g. create policies that stimulate them to develop CBMs around higher value retention options beyond recycling) (Bidmon and Knab, 2018; Huijben and Verbong, 2013; Verleye et al., 2023). Accordingly, we call for future research to enhance our understanding of the embeddedness and role of CSUs in CE ecosystems ¹ and their potential to engage and support CSUs in achieving a more radical circular transformation of the plastics industry.

In particular, entrepreneurial ecosystems ² (EEs) – regional communities focused on facilitating the start-up and scale-up of new entrepreneurial ventures – have been found to largely determine the rate of emergence and rate of stability of start-ups (Bala Subrahmanya, 2017) and play a central role in the transition towards more CBMs (Pizzi et al., 2022). The strength of these ecosystems (in terms of financial support, mentoring and provision of material and intangible resources) improves the stability, scalability, competitiveness and ability of such ventures to have a greater impact on the economy (Bala Subrahmanya, 2017, 2022; Santisteban et al., 2021). In essence, entrepreneurial ecosystems empower start-ups by providing them access to several SFs we identified. As such, in the CE, EEs represent emerging structures with the potential ³ for transformative change (Berghuis et al., 2023).

Thus, although CSUs should take agency in adopting a broader view and move from a firm-centric focus towards an intensive interaction with CE ecosystem actors (Pieroni et al., 2019), a corresponding response from the entrepreneurial ecosystem side is also needed. Important research questions are: What is the extent of agency and role of CSUs in ecosystems? How do we facilitate CSU collaboration and support within the value chain and ecosystem for a more radical transition to a CE? How do we orchestrate BMI at firm and ecosystem levels to achieve transformation towards the CE? How do we engage future intra- and inter-industry stakeholders needed for operating these collaborative business models? Also, for entrepreneurial ecosystems in particular, what competences do they need to engage and support CSUs to play a key role in a more radical circular transformation of the (plastics) industry? What kinds of programmes, services, tools and knowledge can EE actors develop (e.g. universities, incubators, science parks, etc.) to help CSUs develop or gain access to key SFs?

Managerial implications

These theoretical implications point to several managerial implications. Firstly, beyond identifying and categorising barriers, we provide in-depth empirical insight, as strategies and SFs, for CSUs to influence elements of their individual business model and overcome barriers. Our research also shows that they cannot design or shape the entire ecosystem business model alone. A multi-stakeholder collaborative process is not easy, as stakeholders often have different business models that hinder alignment (Oskam et al., 2020). Value in a circular ecosystem is created, delivered and captured differently than in a linear value chain and requires diverse partners to align their activities (Linde et al., 2021). This means that business model elements must not just be internally consistent and aligned (Ritter, 2014). For an ecosystem business model to work, CSU managers must also continually iterate their business model components to (re)align with relevant ecosystem stakeholders. Equally, Hofmann and Jaeger-Erben (2020) confirmed that CBMI requires new value creation spaces for organisational realignment. Thus, in view of the need for continuous strategic alignment and realignment as ecosystems evolve, CSUs must develop dynamic capabilities. In particular, the ability of their management teams to design, refine, implement and transform business models is an outturn of their higher-order dynamic capabilities (Teece, 2018). For future research, it would be interesting to offer insights into the specific micro-foundations underlying the dynamic capabilities CSUs require for implementing more radical CBMs.

Limitations

Despite our efforts to design and conduct this research conscientiously, it has several limitations. Firstly, our results are valid for the context we explored: start-ups in the Netherlands connected to the circular plastics value chain. However, the term ‘plastics’ covers many polymers with different properties and circularity potential. We attempted to simplify this by grouping CSUs under alternative raw materials, alternative products and knowledge and services for circularity. Still, CSUs within each group will experience a different intensity of related barriers and SFs. Moreover, the majority of participating start-ups approached ‘circularity’ through recycling and product design. This could have been influenced by our selection criteria: we excluded businesses with only partially plastic products (e.g. appliances and gadgets); thus, business models considered more radical (product-service systems, reuse and repair schemes) were not strongly represented in our sample. While we attempted to recruit digitalisation-enabled, service-oriented (potentially more radical) start-ups, they did not respond. Considering these business models could differ substantially from product-focused models, we anticipate that additional (or different) SFs will apply to them. Finally, the SFs we identified are largely influenced by the fact that the transition to circular plastics is in its infancy. As this progresses and barriers evolve – current ones will be overcome, but new ones will appear – the corresponding strategies-in-use are also expected to change.