Challenges to business ecosystem alignment when implementing solar photovoltaic systems in the Swedish built environment

ABSTRACT Implementing solar photovoltaic systems in the built environment (BEPV) is critical for the construction sector’s contribution to mitigating climate change. While previous studies give various insights into innovation implementation, the challenges to value co-creation by construction actors remain underexplored. By studying the alignment of business ecosystems implementing BEPV in the Swedish built environment, we address this need. Drawing on the cumulative experience of construction actors, this study demonstrates how knowledge mislocation, knowledge deficits, cultural discrepancies, insufficient building codes, frequently changing regulations, and implementing a highly embedded innovation can disturb ecosystem alignment. The study contributes to the ecosystem literature, scholarly discussions of innovation implementation in complex construction projects, and the PV diffusion literature by offering insights into the realignment of ecosystems involving value co-creation by actors from previously unconnected industries. The study links PV diffusion research to the construction management literature by exploring the cumulative experience of implementation at the micro level. We highlight the significant industry reconfigurations required to accommodate a deeply embedded technological innovation, reconfigurations going beyond the challenges of implementing systemic innovations encumbered by fewer material dependencies. We also emphasize the critical importance of industrial interaction, coordination, and learning to accelerate the sustainability transition.


Introduction
Meeting the urgent sustainable energy challenge requires increased use of renewable energy sources, which involves the rapid diffusion of innovations in the construction sector (Hendrickson, 2012).This includes solar photovoltaic (PV) systems in the built environment (BEPV) (Magrini et al., 2020).Studies of BEPV diffusion have explored diffusion challenges related to the policy context, cost aspects, consumer behaviours, business models, electricity consumption patterns, and socioeconomic factors (Bankel & Mignon, 2022;Bennett et al., 2020;Curtius, 2018;Shih & Chou, 2011;Strupeit & Palm, 2016).However, while experience of BEPV installations has revealed implementation challenges (Andersson et al., 2019;Ottosson, 2018), less is known about impediments to BEPV implementation related to the particularities of construction projects.
Construction projects involve client organizations, builders, architects, consultants, and contractors, actors that are significantly interdependent in their ability to perform activities (Harty, 2005;Loosemore & Tan, 2000).This calls for specific conceptual understandings of innovation implementation in construction (Gann & Salter, 2000) that involves collective action, understandings that account for the interdependencies essential to the materialization of the final constructed product, that is, the value co-creation (Havenvid et al., 2016).Since previous studies have shown that the particularities of an innovation significantly condition its implementation (Lindgren, 2018), we identify a need for deeper insights into barriers obstructing construction projects that specifically implement BEPV.
Existing research on inter-organizational factors and collective action in the construction industry has frequently utilized network approaches (i.e.Havenvid et al., 2016;Håkansson & Ingemansson, 2013).However, the present focus entails paying careful attention to the interdependence between the two distinct industries of construction and PV that realize the installation of PV systems on buildings through value co-creation.While network approaches have proved useful for understanding collaborative work, the ecosystem approach has been identified as better for studying value co-creation across industry boundaries without complete hierarchical control, bringing multilateral complementarities to the fore (Shipilov & Gawer, 2020;Thomas & Autio, 2014, 2020).
Indeed, the business ecosystem literature has emerged as a promising framework in which to study the collective actions of deeply interdependent actors (Adner, 2017;Jacobides et al., 2018) and has proven useful for understanding actor dynamics and collective value creation in the construction industry (Aksenova et al., 2019;Pulkka et al., 2016).Existing studies of business ecosystems in construction (Aksenova et al., 2019;Pulkka et al., 2016;Viholainen et al., 2021) and in other areas, such as distributed energy systems (Hellström et al., 2015), district heating and cooling systems, (Peltokorpi et al., 2019), and the sharing mobility sector (Ma et al., 2018), reveal challenges of alignment for actors and the importance of paying attention to changes in their collective contribution to value creation through interdependent activities.Thus, the business ecosystem approach appears promising for shedding light on factors obstructing BEPV implementation as it emphasizes challenges hindering ecosystem realignment as an innovation is introduced into an established ecosystem.Against the above, this study aims to identify challenges to ecosystem alignment in BEPV projects.Two research questions guide the study: (1) How does introducing BEPV into construction projects influence ecosystem actors and their activities, positions, and links?(2) What challenges affect the realignment of these actors and their activities, positions, and links in BEPV projects?
Our empirical context is BEPV projects in the Swedish construction sector.Sweden is a suitable empirical context given that the strong market growth of PV projects there (Lindahl et al., 2022) provides good opportunities for gaining empirical insights through access to actors with rich, new experiences.PV diffusion in the Swedish built environment has historically been slow (Lindahl, 2015;Lindahl et al., 2022) versus in countries with similar conditions (IEA-PVPS, 2022).The proportion of solar electricity in the Swedish electricity mix increased from 0.1% in 2015 to 0.8% in 2021 (Lindahl, 2015;Lindahl et al., 2022), which is low compared with countries having similar conditions.For example, the proportion of solar electricity in the electricity mix is 4.6% in Denmark and 10% in Germany (IEA-PVPS, 2022).However, due to government incentives through capital subsidies and a feed-in premium scheme, the installation of grid-connected solar PV systems took off some 20 years ago.Since then, the installation rate has accelerated.Consequently, Sweden's total installed PV power has more than doubled annually since 2016 (Lindahl et al., 2022).The largest share of this increase has occurred in ground-based PV parks and the potential for solar PV installation on buildings remains great.Given recent increases in electricity prices, increased PV installation can be anticipated in the built environment.This PV diffusion is conducted through implementations in construction projects, where still only few projects include PV systems, whose installation is still considered relatively novel by Swedish construction actors.

Business ecosystems as an analytical lens
A broad and eclectic stream of frameworks for understanding actor dynamics has been proposed under the term 'ecosystem'.Recent reviews and theoretical contributions have usefully systematised and clarified some early conceptual flaws in this emerging field (Aksenova et al., 2019;Cobben et al., 2022;Valkokari, 2015;Vigren, 2022).One clarification concerns the distinction between the innovation, knowledge, and business ecosystem frameworks.While innovation ecosystem perspectives focus on the development of specific innovations, and knowledge ecosystem perspectives concern knowledge creation through collaborative research and knowledge building, the business ecosystem approach focuses on end-user value creation (Aksenova et al., 2019;Pulkka et al., 2016;Ritala et al., 2013;Valkokari, 2015).Our focus on innovation implementation rather than innovation development requires us to select a perspective that captures the dynamics enabling value creation from a fully developed innovation.Furthermore, the business ecosystem approach focuses on value co-creation across industry boundaries related to a focal offering rather than on an innovation process as such (Shipilov & Gawer, 2020;Thomas & Autio, 2020).The business ecosystem approach emerges as the most suitable for our objective.
The business ecosystem approach draws on business strategy theories assuming that firm success depends on well-functioning inter-organizational networks of collaborative and competitive relationships (Moore, 1993).Recent contributions have identified characteristics that distinguish ecosystems from other types of conceptualisations of inter-organizational relationships across industry borders and serve as a basis for defining business ecosystems (Adner, 2017).First, ecosystems are delimited by a common focal user value proposition that only materializes through the combined contribution of efforts from various ecosystem participants.
This value proposition may include offering a product or a service defined as broadly as an electric car or as narrowly as Tesla's Model 3 (Kapoor, 2018).Second, since a common value proposition only materializes through collective action, thriving ecosystems require coordination between concerned actors (Adner, 2017;Iansiti & Levien, 2004;Jacobides et al., 2018).This coordination may emerge organically or intentionally from one or several central firms (Moore, 1993), requiring collaboration beyond conventional market-based or linear supply chain arrangements (Jacobides et al., 2018).Third, materializing the common value proposition creates simultaneous complementarities and interdependencies among actors (Kapoor, 2018).Complementarities unfold since actors' various efforts combine to create value, and interdependence emerges since a change in one contribution may affect others.Finally, ecosystems comprise multilateral relationships that cannot be understood simply as aggregations of bilateral interactions (Adner, 2017).Instead, ecosystems are nested systems of actors where the nature of one relationship depends on other actors' relationships (Jacobides et al., 2018).
These characteristics echo the interdependency experienced by actors in construction projects (Aarikka-Stenroos & Ritala, 2017;Pulkka et al., 2016;Viholainen et al., 2021), confirming business ecosystems as a useful lens for this study.When an innovation is introduced, work routines change.Modifying the focal value proposition alters ecosystem elements and vice versa (Moore, 1993).Thus, successfully introducing an innovation that leads to a new value proposition requires paying attention to necessary shifts in the ecosystem structure.Failure to do so will lead to misalignment and hinder the implementation of the innovation.Such shifts may include a need to develop the capacity of one or more actors to undertake new or modified activities, a change in actors' willingness to participate, and conflicting expectations regarding positions, such as who hands off to whom and who faces the end customer (Adner, 2017).
Ecosystem (re)alignment emerges when all actors strive to co-create value and are aware of and satisfied with their positions.Seeking (re)alignment involves striving for network stability through mutual agreement and compatible incentives among actors related to the new focal value proposition (Adner, 2017).Such network stability and alignment can be facilitated by conventions such as rules of engagement, technological standards, and codified interfaces (Jacobides et al., 2018;Kapoor, 2018).Scholars have chiefly used the ecosystem framework to describe ecosystem establishment and growth (Bremner et al., 2017).This study, however, examines the challenges to established ecosystems that arise when the introduction of an innovation alters their structure.Hence, we strive to identify challenges to the alignment of the positions and links of ecosystem actors when performing activities to implement BEPV.In doing so, Adner's (2017) ecosystem elements are used as an analytical lens that has been acknowledged as useful for studying the alignment of unpredictable, innovative inputs among involved actors (Thomas & Autio, 2020).Adner (2017) described ecosystem alignment in terms of four basic elements (i.e.actors and their activities, positions, and links) that characterize the requirements for realizing a value proposition.In an established and well-functioning ecosystem, these elements are aligned with one another and function coherently.However, when an innovation is introduced, it shifts the structure of an established ecosystem.In other words, innovation implementation may require that ecosystem actors develop their capacity to undertake new or modified activities, intensify their willingness to participate, and alter their value chain positions (Moore, 1993).This shift is necessary to restore alignment among the elements (Adner, 2017).It is central to our study, as we are interested in understanding barriers to innovation implementation as challenges to ecosystem realignment.Table 1 summarizes these four elements.
We capture the shift in these elements stemming from the introduction of PV in order to understand the consequences for an ecosystem, yielding in-depth insights into the subsequent misalignment.From this, the challenges to realignment that hinder successful BEPV implementation are captured and explained.

Research design and method
This study examines BEPV implementation in Sweden as a case of business ecosystem realignment in the construction sector following the implementation of an innovation.While previous studies of construction ecosystems have applied multiple-case-study approaches focusing on value creation within particular projects (see, e.g.Pulkka et al., 2016;Viholainen et al., 2021), this study explores challenges that recurrently emerge in many construction projects by gathering cumulative  (Adner, 2017).
Activities Actions undertaken for the value proposition to be created Actors Entities undertaking activities Positions Specified locations in the flow of activities across the system Links Transfers across positions; may or may not include the focal actor actor experiences.All projects, although individually unique in scope, follow roughly the same pattern (Orstavik, 2019), being designed by specialists (e.g.architects, structural engineers, and electrical engineers) and produced by general and specialized contractors (Barlow, 2000;Koskela & Vrijhoef, 2001).Our level of analysis is thus the project level.Our approach yields insights into recurring challenges for BEPV ecosystem alignment, complementing previous approaches and enabling more generalizable results.The boundary of this business ecosystem study is a value proposition shared among all actors (Adner, 2017), i.e. the provision of solar electricity to a building through a distributed solar photovoltaic system.

Data production
We used workshops with key actors as our main data production method (Ørngreen & Levinsen, 2017), since it allows shared experiences of a collective phenomenon to be captured (Jingmond & Ågren, 2015).The workshops were designed to capture informants' experiences by prompting discussions of barriers to BEPV implementation, allowing expression of a combination of information and opinions (Mohammed & Ringseis, 2001).Furthermore, insights beyond projectspecific data were captured by focusing on recurring experiences more relevant to challenges of a more general nature.Purposive sampling was applied, since it allows for a thick description and holistic comprehension that suits an intrinsic case study design (Ridder, 2017).First, actor groups involved in materializing the value proposition were identified.Since BEPV is implemented through construction projects, actors included traditional construction actors, grid owners, PV experts, and PV suppliers (see Table 2).These diverse actor categories gave us the wide range of perceptions needed for a case study (Stake, 1995).To minimize the risk of political manoeuvring among actors, which may undermine the problem formulation (Baer et al., 2013), and to sort the challenges specific to different actor groups, separate workshops were held for each actor group (see Table 2).Second, the groups were approached through open invitations via industry organizations and professional networks related to BEPV.Informants were self-selected.Given our selfselection approach, actors who did not find BEPV implementation important or challenging may have been missed in the data production.Nevertheless, our insights and the informants' judgments suggest that we captured a representative sample of all actor groups.The participants were promised confidentiality, and all documentation was anonymised.In total, seven workshops were conducted.Due to high interest from clients, two workshops were conducted for this actor group.To ensure high participation, the workshops were held in Sweden's two largest cities (i.e.Gothenburg and Stockholm), which offer good transportation access that simplified participation.The workshops had a three-step structure.First, informants were asked to identify challenges individually, drawing on their professional experience from participating in BEPV implementation projects.Approximately 15 min were allocated for this step.The aim was to prompt participants' personal reflection and allow us to capture all participants' perspectives and secure individual engagement.Participants noted the challenges they experienced in writing, and all notes were presented and discussed in the plenary.In the second step, informants sorted the challenges into clusters and jointly identified relationships among the clusters.This process, which involved significant participant engagement and took approximately half of the workshop time, provoked rich discussions of participants' opinions and experiences.Third, informants individually ranked the clustered challenges.The resulting individual rankings were then collectively discussed to expand on and clarify the individual experiences and perspectives.While we gained preliminary insight into the challenges from the individual notes and discussions in the first step, the discussions of opinions and experiences in the second step deepened our understanding of the actors' challenges with BEPV and provided opportunities for triangulation.
Due to COVID-19, workshops with PV suppliers and electrical contractors were replaced with nine semistructured interviews.The data production concluded with a two-hour verification workshop to clarify and revise the preliminary analysis, to which all informants were invited.This occasion also allowed for triangulation.All workshops and interviews were conducted in Swedish, recorded, and transcribed.The workshops and interviews continued as long as the discussions were informative and engaging (see Table 2).

Data analysis
The data were analysed qualitatively in five steps, in line with systematic combining approaches that involve iteratively moving between a theoretical framework, in our case Adner's (2017) application of the business ecosystem, and the empirical world (Dubois & Gadde, 2014).All transcripts were imported and organized in NVivo coding software.Our coding strategy involved comparing and contrasting two coding structures: (a) a theoretically derived and (b) a data-centric open coding structure.In the initial step, we familiarized ourselves with the data to better understand the material.
Second, we coded the data deductively according to Adner's (2017) theory of ecosystem elements, using the overarching constructs of actors, activities, positions, and links (i.e.coding structure 'a').This coding involved gaining general insight, through an ecosystem lens, into how a BEPV construction project unfolds.Third, we explored these four coding constructs in greater depth by seeking to capture the unfolding of BEPV projects in terms of (mis)alignments from an ecosystem perspective.Fourth, we set aside coding structure 'a' and applied a data-centric open coding strategy (i.e.coding structure 'b') focusing on the actors' experiences of challenges.Guided by the qualitative thematic analysis method of Nowell et al. (2017), this coding procedure involved an iterative process of continuously revisiting the data and revising the emerging coding structure.This gave a profound sense of the material and its emergent patterns.The open coding procedure (i.e.coding structure 'b') initially consisted of 844 identified challenges.The thematic analysis procedure left an aggregation of 16 groups of challenges.Fifth, guided by systematic combining procedures, we iteratively explored linkages between coding structure 'a' based on Adner's ecosystem elements and the aggregated groups of experienced challenges, i.e. coding structure 'b'.As the process unfolded, we continuously reviewed the coding and the emerging patterns via peer debriefing among the authors to strengthen the rigour of the analytical procedure (Nowell et al., 2017).This exploration included discussions and collective sense-making.Furthermore, the main author's rich contextual knowledge as a professional in the industry strengthened the analysis.Any aspects that could have biased the analytical work were handled by close and active engagement with the peer researchers.Through this iterative process, we identified how the experienced challenges related to the (mis)alignment of the ecosystem resulted in six distinct challenges for ecosystem alignment.

Findings
The characteristics of BEPV projects in Sweden are presented in terms of the state of alignment of ecosystem elements, answering the first research question.

Ecosystem elements of BEPV implementation projects
BEPV is typically implemented in new building projects, major building renovations, or solitary PV implementation projects in existing buildings.Thus, the ecosystem actors in these contexts are organizations that work together to contribute to BEPV implementation.Figure 1 offers an overview of a standard-approach project that makes up a BEPV ecosystem and shows how the actor disciplines (horizontal arrows) are engaged in different activities (bottom table ).The sequence of the specific activities makes up the positions (boxes at the top) within the ecosystem.PV experts may be engaged at different positions, or not at all, in the process and their activities serve very different purposes (dotted arrows).The activities and actors are connected through links for channelling information, material, and funds (vertical links), such as communication about design and specifications, tenders, contracts, and goods.
During the initiation phase, the client must ensure that the actors in the design team have the necessary BEPV knowledge.Clients may ascertain this themselves or seek assistance, for instance, from specialist consultants such as PV experts.In this phase, the project boundaries are set, constraining the implementation since the initial actors determine the design and organization of the project through initial contracts that affect contracts established later on.These contracts serve as links that influence the activities of the various project actors.Apart from the client and the occasional PV expert, initial actors include architects and technical consultants (e.g.electrical engineers, structural designers, and fire consultants).In the design phase, the architects and technical consultants produce the technical specification documents, ensuring information transfer with a link to the procurement phase.If needed, a PV expert can also be assigned in this phase, establishing links of influence with the other involved actors.During the procurement phase, the client selects PV suppliers, contractors, and electrical contractors based on their ability to deliver on specifications and price.To further link the actors and activities to the next position, contracts for construction work are established, if necessary, with the support of a PV expert.However, this expertise can also be added by involving a PV supplier.The construction phase comprises the position of the actual construction work according to the contracted specifications and the connection of the system to the building and the electrical grid.The handover phase involves certifying the BEPV function through testing and warranty inspection.All technical documentation is then transferred to the client to ensure value realization.This phase ends the implementation with the handover and start-up of the BEPV system.

Indications of ecosystem misalignment
Several indications of misalignment emerged from our observations, and the following explores these misalignments as related to each ecosystem element in more detail.

Actors
Informants reported general uncertainty about what actors should be involved in a BEPV project.Many highlighted the client's responsibility, as the project commissioner, for successful implementation.They stated that PV-related knowledge is particularly critical in the case of clients, since they are the ones specifying and securing the required competence and actors for the task.Yet, several informants stressed that clients lacked sufficient knowledge to assume this role.One architect expressed this as follows: 'Clients … know too little, which risks putting a dead hand on the issue' (WS A).
Many informants noted the role of PV experts in meeting clients' and other actors' need for new knowledge.However, at the time of data production, PV experts were rarely engaged in projects since other actors often lacked insight into the value of their competence.One architect illustrated this using a fictional dialogue between an architect and a client: Client: No, we don't need such a super specialist.Don't you have an environmental specialist who knows a little?Architect: But they specialise in the environment.
Client: But it's probably finethat's enough for us.(WS A) On the other hand, traditional construction actors reported that newcomers to the construction context, such as PV suppliers and grid owners, lacked an understanding of construction work, leading to frustration among involved actors.One client representative claimed that PV suppliers lacked a holistic perspective on and insight into the need to collaborate with the project team to find the right prerequisites for the PV system's design.
Informants also stated that the lucrative nature of the PV market has encouraged opportunists to get involved, and some have failed to meet quality requirements or deliver the correct material: They have started with PV, but they have no real knowledge of it.They sell it and deliver the wrong stuff at the wrong time because they don't knowit is still too new.(I electr) This opportunistic behaviour of some actors also includes inappropriate appointments of subcontractors.One electrical contractor cited an example of PV suppliers hiring unauthorized electrical installers, jeopardizing installation safety: Everyone should use certified installers [to connect the system to the grid], but it can be done by anyone who has mounted PV modules.All that is required to connect to the electricity grid is that a certified installer should do it, but there are obviously people who [misuse their certification by authorising someone else's work].We see this since we have found electrical faults that we have had to fix.(I electr)

Activities
Our analysis reveals significant misalignment related to activities.Both established construction actors and actors new to the context are unsure about what activities to carry out and how to adapt traditional activities.Allegedly, BEPV requires adjustments to and completely new types of activities on the part of various disciplines within and outside the construction sector.Hence, construction actors must handle issues that are new to them, such as energy tax rules and energy market transactions.Construction actors expressed uncertainties about rules and regulations and difficulties keeping up to date with changes in them.One client explained: 'The regulations are so new that in the middle of the process, they can be changed, and then the conditions are different' (WS C).
Informants indicated that few resources were allocated to handling the legal and fiscal activities that come with BEPV projects.Also, BEPV-related regulations were perceived as far from the actors' core business, as another client explained: 'It is not our core businessthere is fear since one must stay up to date with what applies to legal requirements' (WS ver).
When clients initially establish the key actor constellation, they often include the incumbent construction actors but seldom PV-related actors.Thus, in the design phase, BEPV implementation activities are requested by actors such as technical consultants and architects who often lack essential BEPV competencies.One architect illustrated this: 'I don't know much [about PV], so it does not affect me until someone tells me that it affects me, and then, I guess, I do the best I can' (WS A).
At the client workshop, one informant shared this description of the problems arising when actors with knowledge shortcomings conduct critical design activities: PV comes in last … the tallest house is placed in the south, shading the roof with the PV modules.Hence, the design needs to be set from the beginning, otherwise, it will not turn out well.(WS C) When PV experts are asked to support other disciplines, they sometimes lack familiarity with these fields and cannot always translate their PV-related knowledge for specific disciplines.One PV expert stated: 'I lack knowledge of assessing the lifetime of existing roofs, which makes it difficult to advise customers' (WS PV exp).
These observations indicate no agreement on what activities BEPV implementation encompasses.In addition, architects instructed to include PV modules in the building design are seldom given a mandate to ask other actors to perform the required activities.One architect cited an example of this: The electrical engineers have been allowed to move freely and unhindered … and now when PV suddenly comes into the picture, they are unused to giving and taking [i.e.collaborating].(WS A) Actors reported that they were often assigned new activities in BEPV projects similar to their usual activities but that were nevertheless new activities for which they were untrained.For example, an electrical contractor stated that they were often asked to connect the PV system to the building's electrical system, which is a conventional task.However, they were sometimes also asked to mount PV modules, which is beyond their core competence.The electrical contractor stated: The biggest problem is that we electricians and most people who install PV have very poor knowledge of, above all, structural strength.When it comes to mounting PV modules on roofs, I see quite big problems from our side as well.There are design calculations, etc., but once you get to the site … it's not so obvious.(I electr) Another new activity for electrical contractors that requires new knowledge concerns connecting the PV system to the grid.A grid owner commented on electrical contractors' lack of knowledge related to this: 'The electricians connect PV systems to a grid they do not know [the technical details of].They connect something to something they do not know about' (WS GO).
Our data revealed many examples of uncertainty among actors regarding whether they were conducting activities correctly since they must rely on knowledge from unusual sources for construction.An electrical contractor even stated that, given this uncertainty, he conducted tasks and had to correct errors afterwards: I have Googled a lot and not really managed to find regulations and rules and stuff.So maybe not everything was done right, but gradually it had to be adjusted and made right.(I electr) These problems affect not only the function of the PV system but also other building functions.For example, mounting PV modules affects a roof's function, since the mounting affects the roof's load-bearing capacity and watertightness.A construction contractor mentioned that new types of activities due to BEPV implementation affect responsibility boundaries in contracts: 'It is probably a responsibility problem … you are interfering with each other's contracts, so to speak' (WS contr).
Issues related to structural strength, wind load, and fire and roof safety are also affected by implementing PV.Regarding the diversity of activities that BEPV installation actors must address, one PV expert underlined the complexity involved: 'PV is not only PV … but many think that PV is only PV … and that is an obstacle' (WS ver).

Positions
Regarding the positions of activities within the sequence of BEPV projects, there are uncertainties about when activities should be conducted and when to involve certain actors.This uncertainty is critical regarding the engagement of PV experts.PV experts said that, in cases when clients include them in a project to resolve uncertainties, they are often engaged too late.One PV expert illustrated this: 'PV comes in too late in the design process.In the worst case, it is included first in the construction phase' (WS PV exp).This late engagement of PV experts limits their influence on activities important to the installation, as these have already been undertaken.Consequently, tensions emerge among actors as the project unfolds.
Grid owners also expressed frustration, indicating misalignment of the positions of activities.By the time a PV system is to be connected to their grid, grid owners have no opportunity to affect the system's design, components, and quality, leading to connection problems.One grid owner explained: And those grids can be quite old … there were completely different quality requirements when building [the grid], so it doesn't have the standard required today to make connections, which calls for a lot of reinforcement work.But the customer wants to connect to the grid now.(WS GO)

Links
Links within BEPV projects consist of information transfer between actors through documents, such as building codes, regulations, and contracts.The links also cover materials used in construction projects.Construction actors report that building codes do not include information needed for BEPV-related activities.These codes set the framework for the technical specification documents created by the technical consultants in the design phase.Since not all actors are assigned to participate from the start of the process, sufficient information must be transferred through these documents.Because of insufficient building codes, the BEPV actors must gather information and knowledge from unusual sources for construction.Informants indicated that such documents often lack important information or may even include errors.An electrical contractor cited an example of documents not corresponding to on-site conditions: The roofs can be incorrectly measured so that the stuff does not fit.The [design of the PV] system does not take shading into account.So, one has to look at it once more on site.(I electr) Consequently, the technical fire consultants and PV suppliers who follow these specifications find that other actors question the specifications.One PV supplier explained: 'Fire safety regulations -I spend a lot of time talking to rescue service and clients' consultants who are worried' (I PV supp).
Beyond specific errors and deficiencies, information concerning PV system installation may be completely missing from documents.For example, a PV supplier said that the PV system was not covered in the specifications and that, consequently, the PV supplier had to resolve mounting and connection issues very late in the process and without the necessary equipment: Mostly, this kind of thing is not designed from the beginning … it is always I as a supplier who has to keep an eye on these projects.'Are you doing this, is this hole included?' Construction [actors] are great at knowing when plasterboard is to be included, etc., but to us, they say, 'now you can come and mount the PV modules', and by then, they have already disassembled the scaffolding.(I PV supp) It is apparent that many BEPV actors experience significant misalignment in their projects.These actors are seldom coordinated and informed but rather are uncertain about what to do and at what position in the process.They must deal with insufficient communication and flaws in information transfers, and they often experience unstructured handovers.

Challenges to ecosystem alignment
Our earlier account of misalignment identified and explained six distinct challenges that inhibit system realignment.

Knowledge mislocation
BEPV implementation requires that all actors engage in new and adjusted activities requiring knowledge of PV system technology, construction norms and methods, and fiscal regulations.Our analysis revealed that, while the required knowledge often exists, it is mislocated among the actors.For example, the client needs good knowledge of technical issues when allocating actors to the BEPV project and related planning activities and positions.PV suppliers and experts have this knowledge, but it rarely reaches clients and is not seen as standard client expertise.For example, the clients' lack of knowledge is shown by their uncertainty about what activities need to be added or adjusted or whether and when a PV expert should be engaged in the design phase.Also, uncertainties emerge regarding the quality of performed activities, since established agreements are lacking and there is great variation in PV experts' competence, educational background, and ability.Consequently, activities are often conducted in an ad hoc and unorganized way and critical actors are involved late or not at all.
The introduction of PV affects construction actors' traditional design work, requiring that they adapt their specifications to fit the requirements of the new technology.As in the case of clients, the data show that other construction actors lack knowledge of this task.Again, PV suppliers partly possess the required knowledge, but it has not been sufficiently transferred to construction actors.Relevant knowledge is not appropriately allocated among construction actors, and the lack of such knowledge creates a first challenge for activity alignment.Although misalignment was expected due to knowledge mislocation, our analysis revealed that this challenge was widespread.It was particularly surprising to realize that all actors expressed a lack of BEPV-related knowledge, not only related to other actors but also to their own disciplines.One electrical contractor described not knowing how to adapt electrical installations to PV systems: [We do not know] how to solve certain things, how to build certain distribution boards and so on … but we have no one [in the firm] who has knowledge of how it should be.(I electr) Also and relatedly, one electrical installer declared that they and others who mount PV systems have not accessed construction-related knowledge necessary for the task: 'We as electricians and most people who install solar cells have very poor knowledge of buildings' structural strength' (I electri).
One explanation for these knowledge mislocations is difficulty in meeting the increased demand for BEPVrelated competencies because the scale of BEPV implementation has increased and because PV-related technology is developing rapidly.This rapid development requires new links within the ecosystem through connections among actors' new activities.The emergence of these links is both a consequence of and a precondition for knowledge mislocation, since establishing the links necessitates system-wide knowledge of value co-creation, which in turn requires links for appropriate knowledge diffusion.In this way, the knowledge mislocation results in inadequate information and material exchange.

Knowledge deficits
In addition to knowledge mislocation challenges, actual knowledge deficits indicate a need to build new knowledge.The rapid development of PV technology has new implications for construction, with consequences for the sector's knowledge development and adaptation of activities.A PV expert provided an example by describing a PV-related issue that must be addressed by involved actors but for which no knowledge or experience exists: 'Electromagnetic radiation has become a question for some [to deal with] in built areas that are near military interests' (WS ver).
In addition, the study shows that knowledge of the construction sector's methods and norms needs to be combined and integrated into PV-related knowledge.Our data show that this integration not only includes a knowledge mislocation challenge but necessitates new knowledge development involving translation and adaptation.As such, deficiencies of integration constitute a knowledge deficit.
Informants reported that many PV suppliers underestimated the importance of integrating PV-related knowledge into the construction context.This knowledge deficit was exemplified at its extreme by the presence of what informants called 'opportunistic actors'.Another example of deficits related to knowledge integration was PV suppliers' use of unauthorized electrical installers.The PV suppliers' lack of relevant knowledge can lead to a deficiency in their capacity to assess actors' suitability for executing activities.In these cases, failure to meet knowledge demands and adjust activities to construction norms sometimes has severe consequences for the ability to conduct activities correctly and to align the position of these activities in the implementation process.This challenge leads to incorrect activity executions with significant implications for materializing the value proposition.

BEPV embeddedness
According to the data, PV is highly embedded in construction projects due to the multifaceted nature of PV combined with the high complexity of construction projects.This constitutes a challenge to value cocreation.
Regarding the multifaceted nature of PV, many informants highlighted the need to consider multiple aspects besides PV technology when implementing PV.This points to BEPV being highly embedded in several critical ecosystem activities.Besides the electrical system of the building, PV is also embedded in many other parts of which construction actors are unaware.For example, roof construction, architecture, building envelope, fiscal issues, grid connection, fire safety, building codes and regulations, building permits, and environmental impacts are also affected.
Regarding the high complexity of construction projects, construction actors repeatedly emphasized that actors new to the construction sector often fail to understand the challenging nature of the construction process, complicating BEPV implementation.These new actors lack awareness of how the many construction activities are interlinked and thus affect everyone involved.One architect described work in construction projects as follows: Buildings and building projects are such complicated things that every single step where there is a misunderstanding … [will jeopardise] the proper handling [of PV].(WS A) However, the workshops revealed that established construction actors also lacked awareness of the embeddedness of BEPV installation.The interdependence between construction project activities and the unanticipated embeddedness of the BEPV innovation challenged the actors significantly.These activities must be aligned to materialize a value proposition, requiring actor awareness of their interdependencies.Yet, actors often failed to anticipate how their activities must be adapted to enable BEPV implementation, and thus carried out activities insufficient.Instead, PV was initially seen by many actors as a component that was easy to add to a project.However, as the project progressed, the requirements of PV implementation emerged and the actors encountered unforeseen issues, perceived as hard to handle.Our data show how actors were confused about both receiving and mastering tasks.The failure to understand the consequences of the embeddedness also led to the unorganized execution of activities and the insufficient involvement of critical actors.
As shown, the embeddedness of BEPV innovation increases the need to align links and activities and obliges actors from previously unconnected sectors to collaborate around a shared value proposition.In our analysis, however, the lack of alignment is prominent, for instance, as links between actors are insufficient regarding material and information exchange.This insufficient linkage concerns the engagement of PVrelated actors (e.g.PV experts, PV suppliers, and grid owners) with construction actors (e.g.clients, architects, technical consultants, and electrical contractors) and new collaborations between construction actors.

Cultural discrepancies
During the workshops, we observed that actors within and outside construction corresponded to different sectoral memberships with distinct work routines, languages, and cultures due to differences in education, practices, and traditions.The actors themselves also acknowledged differences in work culture.This challenge shows that BEPV is a very recent application that alters how energy production and supply in the built environment are viewed.Also, BEPV requires activities that traditional construction actors cannot carry out.Therefore, actors from outside the sector with a significantly different sectoral culture must be added to the actor constellations.One example is the discrepancy in expectations between architects and PV suppliers about the need for extensive information and material exchange.One architect emphasized difficulties getting PV suppliers to respond to their requests when they asked for material samples, versus when they asked other suppliers.Also, the architect perceived that PV suppliers, who did not see value in collaborating outside of projects, needed to embrace the construction sector to understand its logic and ease PV implementation.There are signs that this is changing, as PV suppliers have begun to see the value of getting into the sector to appropriate its logic: '[PV suppliers] realise that they can't get into the 'jam jar' without acquiring companies dealing with roof construction' (WS A).
However, our analysis revealed that most PV suppliers still upheld a different sectoral culture.One way to overcome cultural discrepancies in the long term is through socialization between diverse actors.Yet, the PV suppliers did not mention cooperation or networking with construction actors, and the analysis shows that PV actors were unaware of the importance of interdependence and of bridging cultural discrepancies.This increases the importance of establishing well-functioning links to ensure the alignment of required activities and challenges for clients who need to coordinate culturally diverse actors to reach ecosystem alignment.The consequence of this challenge is evident in clients' uncertainties about actor constellations and work routines regarding what activities to assign to both new and traditional construction actors.This challenge also leads to uncertainty about division of responsibility and to lack of clarity about the boundaries of activities.Indeed, informants highlighted the need to routinize the positions of actors in the sequence of project activities.The lack of these informal routines also creates problems when establishing formal contracts.It is unclear how BEPV implementation affects the established division of responsibilities in conventional construction projects, as new types of activities affect responsibility boundaries in contracts.The combination of the aforementioned embeddedness of this innovation with the significant cultural discrepancies creates a particularly challenging situation since it requires the establishment of deeply interdependent ecosystem links between actors from diverse sectoral and disciplinary contexts.Our analysis reveals how these differences have hindered the creation of links between actors, and shows that actors often lack sufficient awareness of the required change and coordination.

Insufficient building codes
Building codes are standards that are established by national authorities and shape the conduct of activities in conventional construction projects.Our data confirmed that building codes are essential for handling project-based construction work.Actors perceived that these are not sufficiently developed for BEPV implementation, causing insufficient and unorganized activity execution.One example concerns the lack of PV-related fire safety regulations in the building codes, forcing fire consultants to create technical fire specifications without any justification in established standards.PV modules comprise various materials, which has implications for fire safety specifications, as explained by a fire consultant: When we look at the modules, we wonder what material they are … is it plastic or glass, we have to regard them as the building's surface layer, which depends on how they are mounted, on the façade or roof, [whether the PV modules are building integrated], they are part of the building's surface layer, and then these are not type approved, perhaps, or tested according to our regulations.(WS tech con) A structural engineer also reported that no information about PV module specifications was compiled in generally applicable standards or codes.Rather, in each project, they had to investigate products or materials that might be applied and consider a wide range of possible product specifics and mounting methods, affecting the structural engineer's design specifications.
As noted, insufficient building codes create major challenges for actors when planning and performing activities and establishing links for information transfer within the ecosystem.

Frequently changing regulations
Since the number of PV installations started to increase sharply some ten years ago, regulations established for other purposes have proven unfavourable for BEPV diffusion.These changes affect the ecosystem dynamics because they affect the conduct of activities, require new and changed information links, and create uncertainties among actors.BEPV differs from other construction innovations in that it concerns novel regulations outside the construction business core, such as fiscal aspects of energy production and supply.The informants expressed uncertainty about what regulations affect BEPV projects and the need to accommodate regulations that are not sufficiently adapted to BEPV implementation, leading to the insufficient execution of activities.Informants also claimed that, due to fast-changing regulations, it was not useful to learn regulatory details, as one PV expert explained: 'The regulations change rapidlythere is no point in putting any greater energy into learning everything' (WS PV exp).
The perceived frequent changes and amendments of regulations result in uncertainties about (1) what regulations apply, (2) whether changes will affect BEPV projects retroactively, and (3) how the regulations should be addressed.These uncertainties lead to misalignment in actors' activities and insecurity about the possibility of creating value over time.It appears that the described deficiencies in regulations affecting BEPV lead to uncertainties about what activities need to be carried out, how, and in what order (position).These deficient regulations and the lack of awareness of their importance create poor conditions for establishing the information and material links necessary to connect actors and activities.Our results highlight the importance of establishing wellfunctioning information links in order to realign positions and meet the quality requirements of activities.

Summary of challenges
Our analysis shows that, due to the lack of alignment within the ecosystem, mistakes are made when implementing BEPV, leading to long-term problems for the construction sector and the built environment.The mistakes occur because of the mislocation of knowledge of how to undertake tasks and actors' knowledge deficits regarding the boundaries of their activities and respective responsibilities.The strong interdependencies between actors as well as industry differences challenge the alignment of activities, links, and positions within the ecosystem.Also, since PV is new to the construction context, the actors have difficulties identifying trusted information sources and finding information about how to conduct activities.We identified six main challenges and explained their influence on the alignment of the positions of and links among actors when performing activities to implement BEPV.The distinctions among these challenges become evident through their different impacts on ecosystem elements.However, our analysis indicates that they are interlinked: the high degree of embeddedness of BEPV is part of the reason for the knowledge mislocation, which in turn explains why building codes are insufficient.An interesting finding was that all actors underestimated the impact of implementing the innovation.Apparently, actors first became aware of this impact during the installation process, and were therefore unprepared for the new required activities.Table 3 summarizes these challenges.

Discussion
We add to the ecosystem literature through a case study of misalignment following the introduction of an innovation.Thus, we respond to the need for empirical studies applying the ecosystem concept from specific standpoints in specific contexts (Cobben et al., 2022).In particular, we uphold the growing body of literature on patterns of ecosystem dynamics (e.g.Ansari et al., 2016;Khanagha et al., 2022).While this literature chiefly concerns ecosystem emergence (e.g.Hannah & Eisenhardt, 2018;Thomas et al., 2022), we contribute through insight into the realignment of an existing ecosystem following the introduction of an innovation that involves value co-creation by actors from previously unconnected industries.Our findings reveal how this particular dynamic underlines the significance of cultural discrepancies among actors from culturally distinct sectors.This discrepancy expands on previous insights into the consequences of diverse perceptions of ecosystem rules and their impact on coordination (Jacobides et al., 2018).
In relation to this, our study adds to ecosystem research emphasizing the importance of conventions for coordination (Jacobides et al., 2018;Kapoor, 2018), particularly in the construction context (Pulkka et al., 2016), by exploring the relationship between conventions, in this case building codes, and the knowledge and cultural aspects in an ecosystem.When building codes are insufficient or absent, as in the present case, the actors involved in the construction process must agree on shared operating rules as early as possible (Peltokorpi et al., 2019).Our study reveals that reaching this agreement is more demanding when actors come from diverse cultural contexts and greatly depends on the client who initiates the construction process having sufficient knowledge.In this way, our study connects the ecosystem literature on conventions with the literature on focal firms for ecosystem alignment (e.g.Adner & Kapoor, 2010;Chen et al., 2020;Iansiti & Levien, 2004;Lingens et al., 2021), highlighting the role of the client in achieving alignment.
Moreover, we show how frequently changing regulations cause uncertainty within the ecosystem.Often, regulations are considered external to the ecosystem and excluded from analyses of ecosystem dynamics.While previous studies have provided some preliminary insights into the connection between regulations and ecosystem dynamics, showing that a system's ability to respond to regulations can create competitive advantages (Hannah & Eisenhardt, 2018), our study provides complementary fine-grained insights into how regulatory uncertainty causes frustration for actors and greatly hinders value co-creation.
We contribute to the innovation construction literature by supporting previous findings (e.g.Aksenova et al., 2019;Pulkka et al., 2016) with novel empirical evidence demonstrating that the ecosystem concept is a valuable lens for understanding value co-creation in the construction industry.The identified misalignment between elements of BEPV business ecosystems following the implementation of a systemic and highly embedded (i.e.complex) technological innovation yields insight into requirements for industry reconfiguration.This adds micro-level insights from a novel technological context that advance our understanding of the need for increased collaboration across industries and of key actors' roles in driving change (cf.Aksenova et al., 2019).The construction industry's challenges to renewal, such as individualistic behaviour, lack of actor collaboration, and deficient knowledge sharing (Håkansson & Ingemansson, 2013), are particularly prominent according to the ecosystem view, which focuses on inter-organizational rather than firm-level value creation.
In addition, our specific focus on the alignment of Adner's (2017) ecosystem elements allows us to contribute to the construction innovation literature beyond identifying characteristics of construction ecosystems, by exploring innovation-induced challenges to (re)alignment within the ecosystem with consequences for value co-creation.Our findings show that all the challenges identified here affect the alignment of the elements of activities within an ecosystem.This reveals how the misalignment materializes in the execution of both new and traditional construction activities.In line with previous studies of innovation implementation in other sectors (Adner & Kapoor, 2010), this study provides the construction literature with some empirical evidence of how aligning value co-creation following an innovation entails design alterations in both existing and completely new activities.In line with Pulkka et al. (2016), we show that the ecosystem framework is suitable for capturing the construction logic.In addition, the framework complements network approaches (i.e.Havenvid et al., 2016;Håkansson & Ingemansson, 2013) in construction research by offering new perspectives on the study of innovation implementation.These perspectives include understanding innovation-induced network change in which the ecosystem elements constitute a basis on which key actors can develop a strategy for innovation implementation, including preparing for changes that affect existing aligned elements (Adner, 2017;Aksenova et al., 2019).
Second, this study identifies new implementation barriers by focusing on actors from previously unconnected industries.Existing literature has identified attitudinal and industrial barriers (Vennström & Eriksson, 2010), lack of technical capabilities (Hartmann et al., 2008), insufficient regulations and building codes (Gambatese & Hallowell, 2011), innovation liability (Rose et al., 2019), and rigid contracts and procurement processes (Gosselin et al., 2018) as hindering implementation.While identifying the first two aspects, this study also demonstrates how knowledge mislocation, frequent changes in regulation, cultural discrepancies, and the highly embedded nature of BEPV constitute implementation barriers.
Third and relatedly, we offer insight into a highly embedded case.Earlier studies have characterized PV as a modular innovation (van Oorschot et al., 2021), implying that PV only affects the implementing organization (Slaughter, 1998).This expectation was widely shared by informants who were surprised by the extent of activities affected by PV implementation.In contrast, our results show that the implementation requires significant changes in many activities of several actors, indicating that BEPV is a systemic rather than a modular innovation (Taylor & Levitt, 2004).In addition, our study has several similarities to other studies of implementing systemic innovations in construction.For example, building information modelling (BIM) has been characterized as a systemic innovation (Lindgren & Emmitt, 2017) as its implementation has consequences for many actors and their activities (Papadonikolaki, 2018).Evidently, achieving ecosystem alignment following BIM implementation requires considerable time and effort (Aksenova et al., 2019), and this study demonstrates a similar pattern.However, our case has more substantial material properties than BIM since BEPV is a product innovation that involves physical installation, augmenting its embedded and systemic character.Thus, our case emphasizes how the innovation's highly embedded and systemic character and the construction process's interdependent nature significantly hinder alignment for value co-creation.
By revealing the importance of accounting for the material embeddedness of systemic product innovation, our findings suggest that the extent of BEPV embeddedness exceeds that of process innovations such as BIM (Lundberg et al., 2022;Papadonikolaki, 2018;Papadonikolaki et al., 2019) and logistics innovations (Hedborg Bengtsson, 2019).The present case yields insights into the specificities of implementing energy innovation, which necessitates taking into account factors beyond those highlighted in the literature as critical for systemic innovation.We demonstrate that, in addition to the adaptations necessary in the implementation phase, incorporating energy-related innovations has additional implications for buildings as end products, since their functions are transformed into active components of the energy system, requiring new industry cooperation and lifelong maintenance for the end-user (Viholainen et al., 2021).Consequently, we can anticipate more significant challenges when implementing systemic energy innovations than those shown in other studies of systemic process innovation.Fourth, our study casts new light on the importance of knowledge for value co-creation in construction.While earlier research has emphasized the importance of knowledge and information sharing (Bröchner, 2016;Eriksson, 2013), this study shows that mutual knowledge building within and across projects is essential for innovation implementation.We show that knowledge mislocation and BEPV embeddedness emerge as significant challenges affecting all ecosystem elements and that the requirement for mutual knowledge building is closely connected to the embedded nature of BEPV.Thus, for the industry to advance through innovation, innovation-specific knowledge must be allocated strategically, intra-organizational learning must occur between projects, and crossorganizational learning must occur within projects through collaborative efforts.This study's contribution supports emerging evidence of the importance of collaborative knowledge building for value co-creation in construction business ecosystems (Viholainen et al., 2021).
Fifth and as indicated above, we add to the literature highlighting the critical role of focal firms in ecosystem alignment and to related contributions from the construction sector (Viholainen et al., 2021).Our study shows that knowledge mislocation and deficits affect both PV-related and traditional construction actors, leading to disorganized and improvised activities.Although the mistakes and knowledge shortcomings of any actor negatively affect the whole ecosystem (Brusoni & Prencipe, 2013), our analysis finds that the lack of knowledge on the part of the client, as an ecosystem focal firm, is particularly critical as all actors rely on client-mandated activities.Previous studies show that the knowledge shortcomings of critical ecosystem actors, here, clients, can be addressed by engaging additional actors (Lingens et al., 2021).Our findings reflect this: clients sometimes involve PV experts during the early design phase.However, this is not always the case, and clients' lack of technical knowledge and experts' involvement at an unsuitable position in the process can obstruct the alignment.Our study identifies a need for deeper insights into the client's role in aligning BEPV ecosystems.It supports previous construction studies emphasizing the client's central role in shaping value co-creation through procurement, contracts, and requirements (Eriksson, 2015;Rose et al., 2019;Winch & Leiringer, 2016).
Finally, our study adds to PV diffusion research that has chiefly focused on a higher level of analysis (Andersson et al., 2021;Osseweijer et al., 2017;Palm, 2022;Vroon et al., 2021) by offering insights into how previously identified diffusion obstacles transform and merge with the particularities of value co-creation in construction projects.For instance, while previous contributions identify that inconsistent regulations (Andersson et al., 2021;Osseweijer et al., 2017) and weak links between the PV and construction industries affect PV diffusion (Vroon et al., 2021), this study supports these findings by offering novel fine-grained empirical insight into the consequences for value cocreation.In this way, we connect the PV diffusion literature with the construction management literature.
Despite the challenges identified here, the number of BEPV installations is rapidly increasing in Sweden, albeit from low levels (Lindahl et al., 2022).However, our data indicate that mistakes occur in the implementations, leading to malfunctions.While the misalignment has been somewhat addressed on site, the informants stressed that solutions are unsystematic and resource intensive.Thus, BEPV ecosystems in Sweden are still significantly misaligned, much to the frustration of the involved actors.Although this misalignment is sustained by the challenges identified here, this is expected since aligning an ecosystem takes time.For instance, establishing a BIM ecosystem in Finland took at least 15 years (Aksenova et al., 2019), so it could take several more years before BEPV ecosystems achieve alignment in Sweden.

Conclusions
This study has identified six challenges to ecosystem alignment in BEPV projects and shown how the characteristics of the specific innovation explain the misalignments occurring in a Swedish construction ecosystem when the innovation was introduced.This has yielded novel insights into the obstacles to BEPV implementation.Knowledge mislocation hindered the adequate conduct of activities and led to unreliable exchanges of information and materials due to faulty links among actors.Knowledge deficits regarding PV created ambiguity about what actors must be involved in the implementation and about how to conduct activities.The high degree of embeddedness of BEPV required novel constellations and links between previously unconnected actors.It also required changing activities in ways that actors were unaware of.Cultural discrepancies between actors within and outside construction reflected differences in their pre-understandings and expectations of their own and others' activities and areas of responsibility.Insufficient building codes also emerged as a challenge, as the codes proved unsuitable for PV installations on buildings and actors experienced tensions and confusion, occasionally leading to faulty installations.Finally, the frequent changes in governmental regulations affecting BEPV led to uncertainties among actors about how and what activities were influenced by the changes.
Our study contributes to the business ecosystem literature by focusing on the realignment of an existing ecosystem involving value co-creation by actors from unconnected industries.It adds to the scholarly literature on innovation diffusion in construction by supporting previous findings with novel empirical evidence demonstrating the usefulness of the ecosystem concept for comprehending value co-creation.In particular, we identified new implementation barriers in the form of frequent changes in regulation, cultural discrepancies, and the highly embedded nature of the innovation in focus.Furthermore, the study links PV diffusion research to the construction management literature by exploring the cumulative experience of implementation at the micro level.We have shown that BEPV has a highly systemic character as an innovation, which has implications for understanding and handling its implementation through construction projects.
Although this study is confined to a Swedish setting, it offers opportunities to translate insights to other geographical and institutional settings.Challenges related to knowledge mislocation, knowledge deficits, and cultural discrepancies would be expected to emerge in any setting involving the integration of two previously unconnected industries.Although the institutional setting of Sweden directly conditions the construction regulations and building codes considered here, it can be assumed that adaptations, such as the need for increased BEPV standardization in building norms and stable regulatory conditions, will be needed in any setting.Furthermore, the material characteristics of BEPV as a systemic innovation and the subsequent challenges related to its embeddedness will be general to any graphic and institutional setting.
Several topics for future research emerge from this study.We have offered insights into overall challenges, but ecosystem actors' particular obstacles and roles deserve further attention.In particular, the role of clients merits attention since they emerge as the primary beneficiaries and orchestrators of the ecosystem.
Furthermore, this study was conducted within the particular contextual setting of Sweden; additional studies from other national contexts would provide insight into the influence of the institutional setting and help build knowledge of generalizable patterns.Another future research avenue would be to revisit this case in ten years.Such a study would reveal how construction actors have overcome energy innovation challenges related to the sustainability transition in the built environment (e.g.installing energy storage), mastering ecosystem realignment, and identify what challenges still prevail.As an alternative, countries with a more mature trajectory of PV than Sweden and where PV is a standard application in construction could be subjects for a comparative study.
Our findings have implications for practitioners in both policy and business.Regarding policy, this study identifies a need to support the robust adaptation of formal conventions, such as regulations and building codes.This includes carefully considering balancing frequent changes to formal conventions against the resulting uncertainty, as both create obstacles to BEPV implementation.Furthermore, governmental regulations (e.g.PV specifics incorporated into building codes), fiscal rules regarding electricity production and distribution, and grid connection regulations must be adapted, stabilized, and predictable for actors to incorporate PV in their professions.Also, decision-makers committed to accelerating the energy transition need to support the development of the required information and knowledge exchange between construction and PV actors.Regarding business actors, the study highlights the need for a coherent understanding of the ecosystem and to pay attention to how actors' activities interact with those of others.In particular, key actors, such as clients, must recognize the diverse aspects and challenges of BEPV implementation and develop the capability to orchestrate such projects by acquiring competence at the right time.Also, business actors must secure organizational learning and promote the development of conventions among novel constellations of culturally distinct actors within and among projects.Furthermore, it is critical that the construction industry have realistic expectations of the magnitude of the challenge that implementing such complex innovations entails.Our research highlights the significant industry reconfigurations required to take on such a systemic and deeply embedded technological innovation, going beyond the challenges of implementing a systemic innovation with fewer material dependencies, such as BIM.It also emphasizes the critical importance of industrial interaction, coordination, and learning in order to accelerate the sustainability transition.

Figure 1 .
Figure 1.Implementation of a PV system in a construction project (see separate file).

Table 1 .
Elements of business ecosystem structure

Table 2 .
Details of workshops and interviews.

Table 3 .
Overview of the challenges and their influence on the ecosystem elements.