Green public procurement implementation: a case-study analysis to overcome the intention-action gap Open Access

With a procurement volume of about 13% of gross domestic product (GDP) in countries of the organization for economic co-operation and development (OECD) (OECD, 2023), public sector institutions can wield their procurement spending strategically to accelerate the transition towards a more sustainable economy (Dimand, 2022). As Patrucco et al. (2024, 3) define, strategic public procurement refers to “the deliberate use of procurement resources and processes to achieve both public management and public policy-driven goals”, including the pursuit of sustainable, safe and socially responsible outcomes. Public authorities serve as role models in the socioecological transformation of the economy (Agenda 21, chapter 4 par. 23; Testa et al., 2012). Nearly all OECD countries have adopted strategies or policies to integrate environmental considerations into their procurement systems and the majority of these countries systematically monitor the outcomes of these policy initiatives (OECD, 2021). At the EU level, green public procurement (GPP) is recognized as a strategic policy instrument aimed at promoting sustainable consumption and production patterns as outlined in the European Green Deal and the Circular Economy Action Plan (European Commission, 2019, 2020).

Previous studies have examined the use of green criteria in supplier selection (e.g. Igarashi et al., 2015; Testa et al., 2016; Yu et al., 2020). While differing in scope, country context and product groups, these studies consistently conclude that sustainability criteria are not yet systematically embedded in public procurement decisions. This persistent misalignment between high-level strategic goals of GPP policies and the actual use of green criteria in awarding contracts highlights a gap between GPP objectives and procurement practice. The discrepancy between the normative ideal of green procurement and its limited realization is not a new phenomenon. Bowen et al. (2006, 151) referred to it as an “apparent paradox between the theoretical desirability of green procurement and the slow implementation in practice”. A similar gap has been observed in private-sector purchasing, where sustainable intentions are often overridden by other criteria such as cost, quality and brand awareness (Boulstridge and Carrigan, 2000).

While the literature has extensively explored the “why” and “what” of GPP, examining motivations, policy frameworks and the adoption of green criteria, there remains limited empirical research on the “how” of GPP, that is, the internal organizational processes and structures through which GPP is enacted, embedded and routinized in public institutions. As Hsueh et al. (2020, 709) stated on the implementation of sustainable public procurement (SPP), “future research should also consider how the implementation of a new policy – SPP or otherwise – occurs over time and how it reconciles pre-existing rules, procedures and decision criteria with new requirements.” This study addresses that gap by shifting the analytical focus to the internal dynamics of GPP implementation. Here, implementation is not understood as the execution of isolated GPP actions or the one-off application of green criteria. Rather, it is conceptualized as a broader process of institutional change through which GPP principles become embedded in the structures, routines and practices of public procurement organizations. Against this background, implementation is conceptualized as the operational link between GPP policy intentions and actual procurement behaviour (Igarashi et al., 2015; Lundberg et al., 2015). This study addresses the following research question:

Despite its importance, the field of public procurement “is still relatively under-theorized” (Eikelboom et al., 2018, 192). Moreover, “extant organizational research has offered a limited theoretical understanding of the factors related to SPP adoption” (Behravesh et al., 2022, 10). Although a few studies adopt theoretical perspectives (e.g. Roman, 2017; Dimand and Neshkova, 2023), the processes through which GPP becomes embedded in the day-to-day operations of public organizations remain insufficiently explored. Related studies by Brammer and Walker (2011), Grandia and Voncken (2019) and Nicolas and Schotanus (2025) have advanced the understanding of the factors influencing GPP adoption. However, their focus lies primarily on external pressures, organizational readiness or individual-level decision determinants rather than on the internal, collective processes through which implementation is enacted within organizations. Compounding this issue is the conceptual ambiguity surrounding both sustainability, which Roman (2017, 1049) describes as “extremely broad and ambiguous”, noting that “there is no ‘theory of sustainability’” and implementation, which, as Hjern (1982, 301) observed, “has come to mean many things”. As a result, prior research often draws selectively on disparate theoretical fragments without applying a comprehensive framework that could systematically guide data collection and interpretation. To address this gap, the present study applies May (2013) general theory of implementation (GTI), originally developed in the context of health care. Based on normalization process theory (Murray et al., 2010), May (2013) conceptualizes implementation as a social process of collective action. GTI integrates constructs from sociological and social cognitive theories to provide a coherent model for understanding how new practices are normalized and sustained within organizational contexts.

Applying May’s (2013)GTI to GPP enables this study to explore the conditions under which GPP becomes institutionalized as part of routine procurement practice. Selecting 12 qualitative case studies of high-volume public contracting entities in Germany, it investigates how organizations adapt and restructure to internalize sustainability principles in procurement. Case studies are well-suited for capturing context-dependent phenomena, including differences in purchasing behaviour across institutional settings and complex dynamics such as evolving supply chain interdependencies and benefit from methodological diversity (Dubois and Salmi, 2016). The study seeks to advance theoretical and empirical understanding of the persistent intention-action gap in GPP by identifying the internal structures, routines and social mechanisms through which implementation unfolds and how these might be leveraged to support more consistent and impactful sustainable procurement practices.

Public procurement operates within a defined political and legal framework, which forms the foundation for the strategic objectives of procuring organizations (Rendon and Snider, 2010; Vörösmarty and Tátrai, 2019). In this context, Krause (2011) emphasized the need for multi-level modelling, as implementation units are typically embedded within larger institutional structures. Rosell (2021) further differentiated between macro-level factors (e.g. national regulations and political decision-makers as key actors) and micro-level conditions (e.g. organizational capacity, training and awareness of public sector employees, who are the responsible actors for implementation). For instance, municipalities function within the broader political and economic frameworks of their federal and state governments and are tasked with enacting national and regional procurement policies (Krause, 2011). At the level of public policy, GPP constitutes a policy instrument intended to incentivize suppliers to develop environmentally friendly solutions, thereby stimulating competition and ultimately contributing to the transformation of markets for sustainable products and services (Rainville, 2021). Within public organizations, GPP is interpreted as a means of integrating environmental criteria into procurement decisions alongside traditional considerations such as price and quality by assessing characteristics of products, production processes, technologies and inputs (e.g. Testa et al., 2016).

However, political ambition alone is not sufficient to ensure successful GPP implementation. Strategic planning is required to translate these ambitions into concrete and targeted actions that guide organizations from their current state towards desired future outcomes (Storsjö and Kachali, 2017). In the GPP context, strategic objectives aim to integrate green criteria into actual procurement practices (Sönnichsen and Clement, 2020). Yet, the limited integration of such criteria, despite the availability of legal instruments and clearly defined policy goals, suggests an implementation deficit. This indicates a persistent gap between policy goals and procurement behaviour, underscoring the need for effective strategy execution (Grandia, 2016). The strategic goals, therefore, need to be translated and implemented into actual action. Strategy implementation encompasses the communication, interpretation and adoption of strategic goals and requires adjustments to perceptions, structures, processes, resource allocation and social dynamics within the organization (Friesl et al., 2021; Amoo et al., 2019). May et al. (2007) characterize implementation as an intentionally initiated process, in which actors intend to introduce and institutionalize new or changed practices through their own and others’ actions. It includes all related activities from initial planning to operational execution (Linton, 2002) and can lead to the routinization of new practices in everyday work (May, 2006).

The term “implementation” never refers to a single “thing” to be implemented. Rather, the introduction of a new way of thinking, acting or organizing produces a complex ensemble of material and cognitive practices (May, 2013). These depend on the ability of individuals and groups to effect change through their interaction with other actors, processes and contextual constraints (Campbell et al., 2007). Implementation involves mobilizing both material and symbolic resources, as well as securing cooperation, legitimacy and expertise within the relevant field of practice (May and Finch, 2009). Hsueh et al. (2020) operationalize SPP implementation as the extent to which local governments adopt and use standards and rules, establish routines and develop decision-making criteria for incorporating sustainability into their procurement activities. Similarly, Koplin et al. (2007) emphasize the need to align environmental considerations with overarching corporate strategies and policies. From this perspective, GPP Implementation can be understood as the conversion of strategic intentions into organizational action (intention-action link; Figure 1, Schültken et al., 2023).

Borrowing or adapting theories from other research areas can serve as a framework for theoretical development (Fisher and Aguinis, 2017). When transferring theories from other disciplines or to different levels of analysis, the applicability of the theory to the new context and its validity in that context may be questioned (Whetten et al., 2009). Frameworks from articles dealing with implementation in the context of procurement (Neessen et al., 2021; Behravesh et al., 2022) have similarities with May’s GTI. It provides a structured approach to examining the multifaceted aspects of implementation, such as organizational culture, stakeholder engagement and institutional constraints, which are similarly prominent in the GPP landscape (Behravesh et al., 2022). It emphasizes the importance of identifying “normalization processes”, which refer to the way practices become routine (Murray et al., 2010). This concept is beneficial in GPP, where policy objectives appear to fail to become institutionalized operational practices (Dimand and Neshkova, 2023; Grandia and Kruyen, 2020). It is linked to constructs from sociological theories of social systems, as well as social cognitive theories of psychology. The approach is to integrate existing theoretical constructs into new ways to achieve a more comprehensive explanation of the components of implementation processes. This theory uses four constructs: capacity, potential, capability and contribution (Figure 2).

The construct of “capacity” describes the ability of the organization to absorb and implement new ideas, technologies or working methods. Ideas and innovations flow through social networks, which consist of relationships between actors who exchange information, work together and influence each other. It is about whether the environment is able to act together and integrate GPP into practice. For enough capacity, it needs various resources and structures, as social norms, social roles, material resources and cognitive resources. Social norms are rules and expectations that define what is accepted or permitted, helping to control the behaviour of those involved (May, 2013). They are understood as directives or preambles, for example, for the GPP of an organization. Social roles determine who should do what and how actors should behave to create clear tasks and expectations for individuals or groups (May, 2013). In the GPP context, this social role can be covered, for example, by setting up a competence centre for sustainable procurement (Chiappinelli et al., 2019). Material resources are symbolic and actual currencies, things such as money, equipment, rooms or technical systems. To implement GPP, both is relevant, financial and technical resources (Behravesh et al., 2022). Cognitive resources include knowledge, information and evidence, real and virtual objects (May, 2013) and help those involved to understand and apply GPP. It covers, for example, the availability of time capacities and expertise (e.g. on procurement law considerations; Behravesh et al., 2022). To implement a complex measure successfully, the actors must be able to work together and coordinate. This means supporting and coordinating with each other, pursuing common goals and having access to the necessary resources and information. Capacity is therefore not an individual ability, but a collective possibility within the organization.

The construct of “potential” describes the willingness and motivation of the actors (individually and collectively) to participate in the implementation. In GPP implementation, it is about attitudes, beliefs and willingness to change one′s procurement behaviour, even if this is done voluntarily or with freedom (Behravesh et al., 2022). It is about whether they are inwardly ready and can imagine actively helping to shape GPP, both out of personal conviction and as part of a team. Even if an organization has structures, networks, roles and resources, GPP will only be implemented if the actors support it. This inner drive is a prerequisite for GPP to become a reality. It depends on individual decision-making (GPP decisions as a personal value and need, following Webster and Wind (1996); May (2013) refers to this as “individual intentions”, a term from which we deviate to avoid conceptual ambiguities) and shared commitments (if the actors are jointly motivated and convinced of the change). Shared commitment is about a collective team spirit based on common GPP goals, values and convictions (May, 2013). It is expressed through active communication between individual persons for exchange (Behravesh et al., 2022).

The construct of “capability” means the ability of actors to apply a new, complex measure in practice and integrate it into their social and organizational context. Capability depends on how well the implementing measure works (workability) and how well it fits into the processual environment (integration; May, 2013). In the GPP context, workability is determined by the extent to which the actors can incorporate GPP into their operational practices, e.g. with a sustainability criteria catalogue as a concrete orientation. Integration is determined by the extent to which GPP fits into the existing processes. This can be achieved by formulating a GPP guideline (Sönnichsen and Clement, 2020).

The construct of “contribution” refers to the actions taken by individuals or groups within an organization, either independently or collaboratively, to enable the implementation of GPP in practice. Contribution should not be conflated with implementation itself. Rather than representing the outcome of implementation, contribution constitutes an explanatory factor within the implementation process, specifically one that strengthens the normalization and institutionalization of GPP. In this sense, contribution is a necessary but not sufficient condition for successful implementation. It involves not merely isolated or one-off activities, but continuous, conscious and collaborative engagement over time. GPP can only be successfully implemented if actors understand the purpose of GPP, actively participate, work together and continuously reflect on its impact and adapt their practices accordingly. Following May (2013), the construct of contribution focuses on coherence (sense-making), cognitive participation, collective action and reflexive monitoring. Coherence refers to an actor’s ability to make sense of a given GPP measure, understanding its purpose, contextualizing it within their professional domain and recognizing its relevance and applicability. Cognitive participation captures the extent to which actors consciously decide to engage in the implementation process, mobilizing the necessary resources and forming or joining a community of practice. Collective Actions frame how actors operationalize GPP in practice, contribute their knowledge, use existing resources and implement concrete measures, such as demand reduction, the application of sustainability criteria or the exclusion of non-compliant materials (Igarashi et al., 2015). Reflexive monitoring frames how actors evaluate the implementation process and its outcomes, gathering information, appraising effects and using this knowledge to inform future decisions and adjust their practices (May, 2013). For instance, this may involve assessing whether procured services meet defined sustainability criteria or evaluating the effectiveness of green award criteria (Govindan et al., 2015). Drake and Xu (2025) argue that the role of enforceability in influencing bidder behaviour and competition in public procurement has been largely neglected in previous literature despite its economic importance. Contributions are inherently social and do not emerge in isolation. They are shaped by institutional roles, formal rules, shared expectations and collaborative coordination. A comparative overview of constructs with exemplary operationalization of the items can be found in Table 1.

The GTI emphasizes that implementation is an emergent social process with four dynamically interacting constructs. Capacity creates structural conditions, potential activates actors to participate individually and collectively, capability specifies how implementation can take place at all and contribution is the visible realized action in the social space. If positive experiences arise through initial contributions, this can strengthen potential, legitimize capacity building or redefine capability requirements.

To find out why the intention-action gap in GPP persists and which factors contribute to the implementation of GPP policies in public procurement organizations, a comparative case study analysis was conducted with 12 contracting authorities from Germany. The GTI provides a structured analytical framework to systematically examine complex, context-dependent processes of implementation. Case studies deal with real context-bound processes (Warren and Bell, 2022), which GTI captures through its four interrelated core constructs: capability, capacity, potential and contribution. These can be operationalized into interview questions and coding categories, but also retain analytical sensitivity and openness, making them well-suited to the exploratory and interpretive logic of the case study method. Building on theoretical groundwork (e.g. normalization process theory, structuration theory and strategic action field theory), the GTI combines individual and organizational perspectives. A comparative case study approach involves a detailed within-case analysis as well as systematic cross-case comparison to identify patterns of difference and similarity (Gibbert and Ruigrok, 2010). In this study, each case represents an individual public contracting authority responsible for tendering in a specific product group. The selected product group “cleaning services and cleaning agents” serves as a suitable lens for examining GPP implementation, as it is widely procured across various public sector entities and has a significant environmental footprint due to the use of chemicals, water and energy [UNEP (United Nations Environment Programme), 2017].

All selected cases are public sector clients operating under the German Public Procurement Law. To account for the federal structure of public administration in Germany and to capture potential differences in governance, resources and institutional settings, cases were chosen from a national level (named “N1”, “N2” and “N3”), a regional level (named “R1”, “R2” and “R3”) and a local or municipal level (named “L1”, “L2”, “L3”, “L4” and “L5”). In addition, one case from a public contracting authority operating outside the conventional administrative hierarchy was selected (named “O”), offering a distinct organizational setting for comparison. The inclusion of authorities across these different levels ensures variation in institutional context while holding the overall national regulatory framework constant. This design strengthens the external validity of findings by illustrating how implementation dynamics unfold across typical types of procurement organizations. Case selection can best be categorized as a “most similar case” methodology, focusing on the contrasting mechanisms within similar contexts (Seawright and Gerring, 2008). This strategy aims to identify divergent implementation outcomes in otherwise comparable settings, thereby increasing the likelihood that observed differences can be attributed to internal organizational processes or actor-driven mechanisms, rather than to contextual heterogeneity. All selected organizations operate in the same regulatory environment, procure the same product group and serve public functions. This contextual similarity enhances the explanatory leverage of the study by enabling controlled comparison of variation in implementation processes. To further ensure the relevance of selected cases, authorities were chosen based on their procurement activity in the selected product group. Specifically, priority was given to authorities that had issued large-volume contracts in this domain over the period 2011–2021 (based on an independent evaluation of statistical data on awarded contracts). Additional cases were identified through existing professional contacts to ensure feasibility and access.

The 12 selected authorities from Germany accounted for around 10% of all awarded contracts (2,970 out of 30,788) and about 6% of the total reported contract value (315 mEUR out of 5,716 mEUR) for tenders above European threshold values (based on data from the “tenders electronic daily” platform). While these shares enhance the generalizability of the findings by covering a substantial portion of relevant procurement activity, they also provide a diverse and robust foundation for a comparative analysis. The cases revealed notable variation in how GPP is integrated as an award criterion. For instance, case “N3” emerges as a prominent example where GPP appears to be a considerable factor in the awarding process. In contrast, case N1, despite its focus on real estate and facility management – where cleaning services are regularly procured – shows limited application of GPP. Similarly, case R3 demonstrates a comparatively high GPP uptake, with green criteria applied in almost 40% of awarded contracts. This reflects differing approaches to GPP implementation and suggests that some organizations have progressed further than others in institutionalizing GPP practices. These observations, highlighted in Table 2, provide the starting point for a deeper analysis through the lens of the GTI.

Each case was examined in detail to ensure a thorough understanding of its context and internal dynamics. The GTI served as the guiding framework for the analysis and ensured consistency across cases while allowing sensitivity to their unique organizational complexities. The questionnaire was jointly developed by the authors and aligned with the GTI constructs.  Appendix 1 presents the interview questions. The interview partners were chosen based on their expertise and relevance to the study topic. The roles considered relevant were public procurement officers (responsible for procurement processes), managers (overseeing strategic planning and implementation), sustainability specialists (responsible for integrating environmental and social considerations in procurement) and specialized departments for cleaning (responsible for defining product and service requirements). While not all roles could be represented in every case due to time constraints, organizational structures or knowledge gaps, efforts were made to identify the person from the organization most involved in GPP implementation. This approach ensured access to the most informed respondents, even if role types varied across cases. Table 3 lists all interview partners and provides details on data collection and analysis. All interviews were recorded with participants’ consent, transcribed and reviewed by the interviewees. Transcripts were then coded by the first author using the GTI framework and discussed within the author team to ensure analytical consistency. The interviews were conducted between 2020 and 2021. Subsequent developments in the participating organizations are therefore not reflected in the findings. These and other limitations of the study are also listed in the final section of the study.

In all cases, initial efforts to build capacities for GPP implementation have been undertaken, with varying degrees of maturity. The social norms of GPP implementation, in particular, appear to be the most advanced element, as evidenced by the presence of preambles, mission statements and strategy documents, formulated with varying levels of detail. These encourage procurement staff to consider the environmental aspects of their procurement activities and substantiate the initial suspicion of high intentions for GPP implementation. The majority of interviewees reported the existence of social standards for sustainable procurement in the organization (L1, L2, L3, L4, L5, R2, R3, N2, N3 and O). These are regarded as binding and accepted. Regulations were the biggest triggers of sustainability activities for both purchasers and interviewees from specialist departments. In the absence of such a regulation or directive, procurement managers reported the lack of a basis for implementation. However, the existence of directives and preambles does not guarantee successful implementation. L2, for example, does not yet see any possibility of implementing the existing preamble, and L5 reported an outdated directive from 1995 that is hardly known within the organization: “As nice as it is that a decision has been made on one side, it’s not so nice when employees join at some point and don’t know about it.” (L5). The existence of social norms, therefore, appears to be a necessary basis but is not sufficient on its own for implementation. Without operational relevance and visibility, social norms remain ineffectual.

The prerequisites of material resources in the form of financial resources were described as similarly good by the interview partners. On this point, nine of the 12 contracting authorities surveyed were generally positive (L3, L4, R1, R2, R3, N1, N2, N3 and O). Some contracting authorities would also accept higher prices in favour of more sustainability: “So we’re on the road to saying that we want to have quality-assured cleaning all year round and are also willing to spend a little bit more on it.” (L3). Nevertheless, they also emphasized that the budget is an issue due to the principle of economic efficiency of the German Federal Budget Code.

This whole sustainability discussion is still being emphasized half-heartedly by the federal government in the public sector at the moment, and not in the right place. Why is that? Because we have enshrined the principle of efficiency and economy in the Federal Budget Code. Where is the principle of sustainability? When you get right down to it, you can only act sustainably as a public authority, which is obliged by the Federal Budget Code to act economically and frugally, if you bring it under the heading of “economic.” In other words, you have to calculate it with your cost structure, not with an economic cost structure and that’s where the problem begins. (O)

Others oppose this development in advance for cost reasons: “And of course, as a buyer, I always have to insist that the most important criterion is and remains the price.” (L1). In addition to the financial resources, the material resources required for GPP implementation also include technical resources. Technical aspects were not seen as a problem by any of the interviewees, and only one case had a technical basis in the sense of an electronic department store labelling sustainable products.

We have now introduced an e-procurement system, in which we also map all of the framework agreements that we have put out to tender sustainably. They are labelled, either ecologically or socially sustainable, as cradle-to-cradle, Blue Angel or local. Ultimately, five options can be categorized and displayed. The procurer can then also sort/filter according to sustainable products, which are automatically ranked better, i.e. displayed higher up, so that the hurdle is lower to simply procure sustainable products. (L4)

Overall, a lack of material resources was not reported as a major problem, but it does play a role and offers potential for implementation.

The cognitive resources of GPP implementation are determined by the availability of time resources and necessary process- and product-related expertise. Six of the 12 cases were positive or neutral about the availability of time resources (L3, L4, N1, N2, N3, O). Although the preparation time for the procedures is between four and nine months, there is not enough time to integrate sustainability into these tenders. Another problem is the limited personnel capacity that is made available for procurement activities.

At the last procurement conference, one speaker said: “Chronically overburdened procurement offices.” I immediately felt comfortable with the term and said, “Yes, it is the same for everyone.” Even if you talk to other contracting authorities, contracts are always awarded in a hurry; there are too few people in procurement, even in the specialized departments. (L2)

A higher level of cognitive resource development becomes evident in organizations that have invested in designated roles and structures. Authorities that do not perceive time capacities as an obstacle also tend to report fewer issues with knowledge capacities. This alignment is particularly visible in cases with defined social roles, such as the competence centre for sustainable procurement of L4: “This is a full-time position, exactly, I can devote myself completely to the topic. We have around 200 decentralized procurers, and I am simply a kind of internal consultant for them.” (L4). Five authorities reported the availability of knowledge resources positively or neutrally (L3, L4, R2, N2 and N3). The majority wished for a better knowledge base: “Furthermore, the demand side and the tendering body lack the expertise to implement the aspects of ecology and sustainability in the tenders.” (L1). Building cognitive capacity requires not only the availability of knowledge and time but also a deliberate and structural commitment to prioritizing sustainability within procurement roles and responsibilities.

The social roles of GPP implementation are reflected in the definition and assignment of responsibilities and tasks to the social roles involved in the GPP implementation process. In approximately half of the cases, persons who are at least responsible for environmental and ecological sustainability have been appointed (L3, L4, L5, N2, N3); in the others, the allocation of roles or tasks remains undefined or delegated to procurement staff or individual demand centres. The institutional arrangements vary considerably across cases. Some reported the existence of cross-departmental working groups or designated competence centres for sustainable procurement:

I have the competence centre for SPP, and I am the contact person for the entire organization in this function. Of course, due to the proximity to the two specialist areas that are part of my department, i.e. the central procurement office of the financial authority, I am much more involved in the processes. (L3)

Buyers from organizations that lack clearly defined responsibilities or distribution of tasks reported frequent coordination problems. L2 reported that the integration of green aspects depends “on the subject of the tender and also on the willingness of the specialist departments to participate.” The definition of social roles also appears to be heterogeneously regulated in the analysed cases. All authorities reporting sufficient knowledge also assessed time availability as neutral or positive. This correlation points to an interdependence between capacity components.

The level of potential, understood as the degree of individual and collective readiness to act, varies significantly across the cases. The shared commitment to GPP implementation was determined by the willingness to drive the process forward together, to co-work transparently and to engage in an intensive exchange across departments. Eight of the 12 cases examined were in favour of this approach (L3, L4, L5, R3, N1, N2, N3 and O). Joint commitment to GPP implementation was determined, in particular, by the willingness to drive the process forward and work together transparently and in intensive dialogue across departments. However, the remaining four cases assessed the lack of cooperation and communication as a major problem. L2 reported that specialized departments communicate “unfortunately, often much too late, i.e. only when the award procedure is pending, i.e. actually too late for such strategic considerations.” (L2). Although the buyer wants to implement GPP, their influence remains limited if the specialist department or user dominates the specification process, operating in isolation. The ability of procurement to steer green outcomes thus hinges on early and structured collaboration.

The individual decision to implement GPP was determined by the proactive involvement of individuals in charge of procurement. As all interviewees actively agreed to participate in the interviews and shared their internal information, they had already shown a high level of commitment to GPP implementation. Although some cases have already integrated sustainability into their overarching strategy, buyers are rarely given concrete GPP incentives or targets. As one respondent remarked: “I also see precisely this incentive problem, because of course the buyer is supposed to bring purchasing success, but if he then implements sustainability particularly well, he doesn’t benefit personally at first.” (L1). Nevertheless, the implementation of sustainability aspects within the organization meets broad acceptance among all those involved. Although sustainability is not always declared an important development within an organization, all interviewees favoured greater commitment: “And [for] this other tender […] we paid a lot of attention to ecological aspects because, firstly, this is very important to me personally, and it is also available as an exemplary tender from the Federal Environment Agency.” (L3). Nevertheless, cases L3, L4 and L5 stood out as particularly proactive. In these organizations, dedicated individuals had initiated GPP-related activities, such as concept development or awareness-raising measures. However, this point should be interpreted with caution, as two of these three individuals also serve as institutional competence centres for sustainable development, making it difficult to fully disentangle personal commitment from role-related expectations.

All the public sector purchasers surveyed were aware of their role model functions. Even if the purchaser wants to procure sustainably, effective GPP implementation depends on interdepartmental cooperation. If it becomes involved in the process too late, the scope for manoeuvring is limited.

The workability of the GPP implementation was determined in the examined cases based on the availability and use of established criteria catalogues or concepts defining what is meant by environmental sustainability. In five of the 12 cases analysed (L3, L4, R2, N2 and N3), such instruments existed in varying degrees of detail. While some of them have developed targeted approaches, in other cases, they have not yet been proven to be practicable. L4 pronounced a clear orientation towards the cradle-to-cradle concept in all its procurement actions:

R2 is oriented towards 22 guiding principles based on the UN′s Sustainable Development Goals, from which the ongoing strategies originate. Conversely, others like R1 do not have something like a criteria catalogue. “At the moment, the fact that there are no specific requirements regarding sustainability from our higher-level authority for tenders for cleaning services prevents us from procuring ecologically.” (R1). Thus, sustainability is not a new topic for the public purchasers involved; however, something like a standard catalogue of requirements was not reported anywhere. The extent to which sustainability is translated into workable instruments varies significantly.

In the examined cases, the integration of GPP implementation was determined based on existing procedural guidelines for GPP. Again, some cases already work with defined GPP-guidelines (L3, L4, N2 and N3). In one case, initial steps have even been taken to digitally map the procurement process and integrate sustainability classification into the organization’s internal e-procurement system:

However, the majority of cases are still in early stages of integration: “Let me put it this way, at the moment there are no instructions from the top on how to do this.” (R1); even if the wish for this was explicitly expressed by interview partners: “[…] that I would like to see some kind of guideline, a framework for action, on how the purchasing department should set very specific requirements and then be able to check them. Just saying that you should keep out more harmful substances and that vaporizing is not an option is of no use to us in practice.” (O)

Similar to the workability, the processual integration of GPP appears to be inconsistent. Dependencies between capability items can be observed, as four of the five cases with some kind of criteria catalogues also have procurement guidelines. This appears to favour the development of GPP guidelines. There are similar cases with ambitions, as in the capacity section.

The coherence or sense-making of the GPP implementation was determined by the extent to which the objectives and paths addressed in preparing documents, mission statements and strategies were realized in procurement practices. Therefore, coherence reflects the alignment between engagement, prerequisites and practices, which is why it can only be considered when the engagement and prerequisites are present. Overall, interviewees reported a desire for systematic approaches to GPP implementation. For example, at L3, procurement guidelines were developed based on the procurement strategy: “[…]The environmental guidelines concretize the procurement law, including the obligation in the procurement law, but not in detail. Environmentally compatible procurement is then virtually mentioned.” (L3). In other places, this coherence is lacking; for example, when L1 has a sustainable corporate strategy but has been purchasing purely according to price, demonstrating a gap between rhetorical commitment and operational practice. The importance of a well-considered, structured implementation approach has been recognized, but its implementation still depends on various factors like organizational priorities, capacities and internal coordination.

The collective action of GPP implementation was determined by the implemented measures for a greener procurement result in the form of avoided procurement or environmental criteria in procurement documents. Eight of the 12 contracting authorities (L3, L4, R1, R3, N1, N2, N3 and O) reported active procurement avoidance for ecological reasons. As part of the environmental criteria, the respondents mentioned requirements for detergent ingredients, packaging, use of labels, dosing aid of cleaning agents, training of cleaning staff for dosing the cleaning agents and production process of cleaning chemicals. In addition, each client reported using at least one environmental criterion. The use of labels was the most frequently mentioned (10 out of 12 cases). Most criteria reported N3, including eco-labels and criteria on ingredients of cleaning agents, packaging, dosing and production processes. These were applied exclusively as mandatory requirements or qualification criteria. If these criteria are used as grounds for exclusion at the bidder eligibility stage, the contract is usually awarded to the lowest price or a combination of price and quality criteria. If sustainability criteria are used as award criteria (by L3, L4, N1 and N2), they are weighted at 5%–50% of the price.

In principle, we try to anchor sustainability in as many places as possible in the tender documents, be it in the bidder registration, in the specification, in the execution conditions or just in the bidder evaluation. But if it is not possible, because the market offer is still too small to set Cradle-to-Cradle criteria as minimum requirements, then we are always obliged to apply them as award criteria to at least 20%. (L4)

However, overall, they still play a minor role in the selection of bidders: “Sustainability criteria are rather less queried. […] the cleaning contracts contain appropriate criteria, […] but subsidized is the most economical offer without somewhere to pay attention to sustainability criteria, which would be evaluated somewhere in the tender.” (R1). To summarize, environmental aspects already find their way into the tender documents in different degrees of detail, as the preliminary ted-data-analysis of the green award criteria has already shown. However, the target direction behind their application is heterogeneous among the different cases and depends on internal prioritization and the capacity to act.

Cognitive participation in GPP implementation refers to measures to promote exchange with or the participation of other parties, such as network meetings, bidder conferences and advisory centres. Apart from the exchange with consumers, which all respondents reported, only three of the 12 contracting authorities were also actively involved with suppliers through bidding conferences or market dialogues aimed at identifying sustainable innovations (L3, L4 and R2). In the other four cases, market exploration with a focus on sustainability aspects was conducted in advance without communicating directly with market participants (L5, N2, N3, O). If a market survey is conducted in advance, tenders for alternative sustainable products can be more targeted. Cases L3, L4 and R2, which describe the most far-reaching exchange of knowledge and most intensive cooperation, have already implemented additional aspects of sustainable procurement, highlighting the role of cognitive participation as a facilitator of broader GPP transformation.

The reflexive monitoring of GPP implementation was defined particularly in measures for the evaluation of GPP activities, such as compliance with the criteria by the contractor and the control of environmental or economic performance. Almost all cases (11 of 12) reported initial compliance checks at contract inception (e.g. the use of certified cleaning agents by the cleaning service provider). However, only five of the 12 cases conducted ongoing checks during the contract period (when procuring cleaning services: L1, L4, N2, N3 and O). If compliance with contractual conditions is monitored, it is performed by a specialized department or user. The purchasing department receives information only in cases of non-compliance. Inclusion criteria that require specific expertise or excessive testing can make the evaluation difficult. In terms of performance measurements, the surveyed authorities were at different stages. Three of them (L4, N2 and O) already conduct a performance measurement or start engagements (“Through the internal product system, we also hope to be able to track prices more effectively in the future and observe how they develop over time.” L4), and three have in-house reports on their activities (L1, L3, L4 and N2). Nevertheless, interviewees frequently identified the absence of indicators, tools and measurement expertise as key barriers to meaningful evaluation, although L3 has already made initial efforts to calculate the cost difference between sustainable and conventional methods (“We’re simply not that stage yet. We don’t currently have a comprehensive overview in terms of figures.” L3). Public authorities review the effects of sustainability measures against the background of the publication of sustainability reports: “[We] need to do reporting. This will partly become our procurement policy, our procurement strategy, true to the motto ‘Do good and talk about it.’” (L5).

Overall, efforts are being made to improve the contributions of contracting authorities. However, many contracting authorities are still in the early stages of institutionalizing robust contribution mechanisms. Particularly in the area of monitoring, only initial structures are in place, despite widespread recognition of its strategic importance. However, there is growing awareness regarding the need for further development, willingness to act and incremental progress towards GPP contributions. The results are summarized in  Appendix 2.

The analysis of GPP implementation based on the GTI reveals that embedding green criteria into public procurement processes requires an interplay of multiple mechanisms. Notably, the data indicate that GPP is strongly anchored in normative frameworks. Thus, the need for GPP is symbolically and politically recognized. However, assigned responsibilities are often missing, as is the consistent provision of material and cognitive resources. While financial means are available in some cases, others (especially at the local level) lack time, expertise and technical capacity. This discrepancy between normative commitment and practical resources points to a structural implementation gap, where formal capacity may be present, but functional capacity remains limited. The data also reveal a pronounced degree of shared commitment across nearly all levels of governance, particularly with regard to communication and participation. This indicates a generally supportive collective climate for GPP implementation. Wang and Li (2014) also highlight the importance of having a strong decision-making team in implementing GPP. Surprisingly, individual initiatives are more frequently observed at lower hierarchical levels than in higher-level or operational contexts, especially in smaller urban organizations such as case L4.

In several cases, GPP is supported by a certain degree of operational workability, most notably through the use of standardized criteria catalogues. The presence of such tools across different levels indicates a general availability of instrumental knowledge and formal guidance. However, integration of GPP into existing organizational processes is inconsistent and support measures for institutional embedding are only sporadically present. This is an unexpected finding, given the wide array of available support instruments for GPP in the German context, but it was consistently linked in the data to missing organizational capacities. As a logical result of these varying conditions, the degree of contribution also varies. Label-based green criteria are the most commonly applied, while more complex criteria, such as those related to production processes or dosing, are rarely considered. Reflexive monitoring instruments, such as evaluation, reporting or tracking systems, are inconsistently established across the cases, which limits the ability for organizational learning and iterative improvement. Nevertheless, cognitive participation is evident in many cases, suggesting a positive tendency towards participatory implementation practices.

The GTI posits that the four constructs, capability, capacity, potential and contribution, do not follow a fixed linear order but rather interact dynamically and reciprocally. This is also evident in the GPP context. Strategic capacity through interorganizational policies or funding does not automatically lead to practical capability when integration into workflows and process support is absent. Collective potential, expressed through communication and shared values, does not necessarily result in individual initiative, pointing to the need for empowerment measures. Contribution is especially strong when practical tools such as criteria catalogues coincide with motivational commitment. Conversely, the lack of reflexive monitoring inhibits the feedback loops needed to advance both capacity and capability. In this light, GPP implementation appears as a circular and adaptive process, in which contribution is not merely an outcome, but also a driver of the other constructs. Sustainable institutionalization of GPP depends on the continuous alignment and mutual reinforcement of capacity, potential, capability and contribution. Thus, the GTI provides a compelling theoretical lens for explaining how the intention-action gap in GPP can be bridged over time.

The items, constructs and interrelations in May (2013) were reflected in the interviews on GPP implementation. The case study analysis identified distinct expressions of the four GTI constructs and revealed a spectrum of implementation maturity across the cases. Importantly, these maturity levels should not be interpreted in binary terms (as fully met or unmet), but rather as positions along a continuum. The GTI framework allows for the nuanced representation of cases at early stages of implementation like “beginners” (e.g. L1, L2, R1, L5, O), lacking strategies, structures, time or knowledge resources, as well as more advanced “pioneers” (e.g. L3, L4, N3), where multiple criteria are applied, market research is conducted and monitoring mechanisms are being introduced. These findings confirm the analytical transferability of GTI from healthcare to the domain of GPP. GTI provides conceptual clarity for differentiating types of implementation behaviour, assessing the internal dynamics of organizations and explaining variation in GPP practices. It also enables structured comparisons across cases and supports the identification of behavioural deviations, contributing to a better understanding of the persistent intention-action gap in sustainable procurement.

A cross-check with Table 1 on the use of sustainable award criteria reinforces the findings, as cases L3 and N3 align with the advanced stage of implementation. The lower positioning of L4 in this comparison may be explained by the competence centre having been introduced only in 2020, towards the end of the observation period, while R1’s unexpectedly low ranking may reflect self-reporting bias. Despite similar levels of implementation maturity, respondents may have over- or underreported progress due to differing expectations or social desirability. While GTI offers a valuable framework for analysing internal implementation processes, the findings also point to additional institutional and interorganizational influences currently underrepresented in GTI. Some respondents attributed implementation challenges to a lack of external mandates, insufficient regulatory pressure or supplier readiness and thus, factors located outside the organization′s direct sphere of control. Grandia (2016, 184) has referred to such externalities as a “black box” in GPP research and Behravesh et al. (2022) emphasized the importance of “institutional setting” for implementation.

These insights raise the question of whether and how GTI can and should be extended to account for these external influences. The study suggests that macro-institutional factors and market signals may shape internal processes indirectly; for example, supplier readiness and institutional setting may influence the constructs of potential and capacity, particularly through social norms and perceived legitimacy. In this regard, insights from legitimacy theory (Suchman, 1995) offer a useful extension, as legitimacy perceptions are shaped by regulatory expectations, professional norms and stakeholder pressures and can affect how organizations interpret their roles and responsibilities in GPP implementation (Nicolas and Schotanus, 2025). Legitimacy theory also supports a more nuanced interpretation of “intention” within the GTI framework.

While May (2013) does not exclude intention from his theory, but conceptualizes implementation as a dynamic social process shaped by agency and structure, in which intentions play a central role, our application of GTI focuses specifically on the intentions and decision-making within the procuring organizations. Broader, system-level or political intentions are considered part of the institutional environment rather than the implementation process itself. From this perspective, legitimacy theory offers a useful extension by highlighting how external mandates, societal expectations and perceptions of legitimacy influence internal commitment, shared meaning and organizational self-understanding, which are closely linked to GTI constructs such as potential and social norms. Thus, rather than drawing a rigid line between intention and implementation, a more integrated view should be taken in which intentions emerge through social processes within organizations, shaped by their institutional context. Incorporating legitimacy-sensitive dimensions into GTI could strengthen its explanatory power in sustainability-related domains like GPP. Future adaptations of GTI may benefit from a more permeable understanding of organizational boundaries and a clearer theorization of how legitimacy pressures and market dynamics condition internal implementation pathways.

The findings offer several practical insights for enhancing GPP implementation in public organizations. A strong internal foundation includes the existence of a criteria catalogue, a clear strategy, formal in-house regulations, designated roles or specialized units with access to technical, financial and human resources. Although Sönnichsen and Clement (2020) identified organizational size as a key driver, the findings indicate that smaller organizations (e.g. L4) can also achieve high implementation levels when these conditions are in place. Moreover, the “pioneers” L3 and L4 demonstrated proactive market engagement, including market analysis, supplier outreach and bidder conferences. This contrasts with other cases, where procurement officials expressed reluctance to engage with suppliers due to concerns over competition law compliance. Conversely, concerns regarding low bidder participation were raised (e.g. by N1), despite empirical evidence from mature markets such as cleaning services indicating that green criteria have little impact on supplier interest (Drake et al., 2024).

Previous literature (Vörösmarty and Tátrai, 2019; Bovaird, 2006) advocates for a more collaborative and less hierarchical understanding of procurement, viewing the supply relationship as a co-productive process rather than a transactional exchange. Patrucco et al. (2022) identified the role of market intelligence and poor supplier matching as critical levers for implementation. Supplier engagement is also emphasized in the UNEP (2022) recommendations and is consistently identified as a determinant of GPP effectiveness (Dimand and Neshkova, 2023). Thus, public organizations aiming to increase GPP implementation should not only invest in internal capacity building, but also cultivate market-oriented capabilities. This includes establishing robust mechanisms for supplier communication, conducting readiness assessments and co-developing award criteria. Doing so requires not only technical and legal clarity but also inter-organizational coordination and a shift from a compliance-based mindset towards a more interactive procurement culture. Such a transformation would enable public procurement to act more effectively as a strategic instrument for sustainability.

The case study analysis demonstrates that the GTI provides a robust and empirically sensitive framework for analysing GPP implementation processes. Its constructs, capability, capacity, potential and contribution enable a structured analysis of both facilitating and inhibiting factors in implementation. The comparative case study approach revealed substantial variation in implementation maturity, from cases with minimal structures or strategies to others that apply diverse criteria and actively monitor results. The most frequently cited obstacles to GPP implementation include insufficient staffing, time constraints and a lack of knowledge. One promising response to these challenges is the establishment of dedicated SPP or GPP positions. These roles can serve as focal points for expertise and coordination, but require strong support from leadership to become institutionally embedded.

In many cases, the procurement process is already too advanced by the time procurement officers become involved, leaving limited scope to influence environmental outcomes. Earlier involvement of technical departments acting as product experts could help improve the integration of green aspects into procurement. A key difference between the more advanced and less developed cases lies in the degree of cross-functional collaboration and engagement with the broader supply chain. In successful cases, technical departments interact with users, buyers and suppliers and conduct market analyses. The ability of contracting authorities to procure more sustainably also depends on market readiness and supplier knowledge. To facilitate greener tendering and market development, enhanced cooperation among contracting authorities, e.g. through harmonized tender documents and aligned award criteria, may prove beneficial. Such coordination not only lowers entry barriers for suppliers but also creates stronger signals and incentives for innovation in sustainable product and service markets.

However, certain limitations must be acknowledged. The study is limited to 12 contracting authorities and only two specialist departments, which restricts the generalizability of the findings. Further, the data reflect predominantly the perspective of procurement staff, sustainability managers and product specialists. The data provide a snapshot from the years 2020/2021 and focus exclusively on the procurement of cleaning services in Germany. The study does not assess the environmental effectiveness of procurement decisions, either in terms of the environmental degree or measurable environmental impact. One recurring challenge identified by procurement professionals was the limited availability of reliable criteria to evaluate the environmental performance of cleaning products, especially concerning their ingredients. As is often the case in qualitative research, the results may be influenced by subjective perceptions, both in the interview data and in the interpretation by the researchers. Future research could build on these insights by using quantitative methods and focusing on market readiness, supplier engagement and the alignment between public demand and the sustainable offerings at the market.

The authors acknowledge financial support by Universität der Bundeswehr München.