Structural power imbalances in global organisations: analysing IT governance from a postcolonial perspective

ABSTRACT

IT governance has the potential to perpetuate power imbalances in international organisations. Important IT decisions are frequently made in the headquarters and then propagated through the rest of the organisation. This pattern can indicate a well-organised centralised IT governance focusing on security and efficiency. The same pattern, however, cannot only impede innovation and flexibility but also has the potential to continue colonial decision structures, especially as headquarters are often in the global north. Through 27 interviews with NGO representatives – a sector that should be highly motivated to counter postcolonial tendencies – we examine IT governance structures. We observe indications of a mismatch between organisational goals and IT governance structures that can only be explained when considering postcolonialism. We need to combine explanation approaches from IT governance and postcolonial theory to explain the observed phenomenon fully. Moreover, we critically comment on IT-related decision-making structures and call for more effort in research and practice to decolonise IT and IT governance.

KEYWORDS:

• IT governance
• postcolonial theory
• non-governmental organisation
• NGOs
• social sector

1. Introduction

Getting IT governance (ITG) right is challenging in large organisations, with difficulty increasing with an international footprint. Different and often conflicting IT goals, e.g., security, efficiency, or innovation, contribute to the complexity of optimisation, as it is necessary to align central goals with local IT requirements (Tiwana & Konsynski, 2010; Wareham et al., 2014). Often, this conundrum leads to structures where someone from headquarters makes IT decisions for the rest of the organisation. Here, headquarters tend to be in the global north,¹ e.g., the US, Japan, or Europe. Thus, power distribution in ITG appears to mirror still past (colonial) power distribution worldwide. We can explain these structures with ITG theory, where a drive for standardisation, IT security, and economies of scale encourages centralised structures (Agarwal & Sambamurthy, 2008; Brown & Grant, 2005; Johnston & Warkentin, 2010; Tiwana & Kim, 2015).

However, some real-world phenomena are not explainable by pure ITG theory. Governance in international organisations seems to be more centralised than theory would prescribe. Here, postcolonial theory (PCT) provides the vocabulary and theoretical lens to analyse the power structures critically. Central governance would imply that a (western) mainstream is acting paternalistically and systematically devoicing IT users (often in the global south) – also called the subaltern (Ashcroft et al., 1998). Hereby, the subaltern include local staff and other local stakeholders, e.g., customers or consumers. We study this phenomenon in the context of large international NGOs, as these typically work in former colonies and have their headquarters in the global north. As such, their internal power structures might be influenced by post-colonialism. Moreover, they should be interested in countering any postcolonial structures due to their societal mission, e.g., to support civil society development (Frewer, 2013). Hence, international NGOs should be especially incentivised to implement the UN’s social sustainability goals, such as inclusiveness. Therefore, we assume NGOs to be more open to critically reflecting on their ITG.

Examining the different perspectives on ITG is important as they form the basis for critically reflecting on current power structures within organisations and IT. As IT tools and decisions increasingly shape more organisations in developing countries (Avgerou, 2008), it is time to reflect on how far today’s IT governance structures follow rational arguments and how far they enforce old imperial and colonial power structures. New user-driven technologies like advanced mobile solutions can democratise decision processes and ensure every user has a voice (Bygstad, 2017). However, old power structures might curb this potential and even counter-act NGOs’ core interests: Prior studies have already established that NGOs do not always reach their ambitious standards (Frewer, 2013). Even worse, there is evidence that NGOs enforce postcolonial structures with their programmatic work (McVeigh, 2020), which would counter the social sustainability of NGOs (Biermann et al., 2017). Hence, it is time to revisit the role of ITG in these contexts.

Prior research has critically examined how post-colonialism affects NGOs in their work (Frewer, 2013; Makuwira, 2018), but IT and ITG have not yet been studied in these contexts. Researchers in information systems (IS) have used PCT to reflect critically on IS projects (Lin et al., 2015; Ravishankar et al., 2013; Sayed & Agndal, 2022). To date, no research has contrasted these perspectives with more established IS-specific theories on how power regarding IT decisions should be distributed – typically summarised under ITG. This research aims to start a discussion of what ITG taking up responsibility could look like. To create a valid starting point, we dive into the world of NGOs, conduct 27 interviews with NGO representatives, and study how IT decision structures can be explained with standard ITG theories and whether or not postcolonial structures shape them. We operationalise this objective with the following research question: What is the nature of the relationship between postcolonial contexts and the design of IT decision structures in NGOs? Based on the results of this research question, we follow Myers and Klein (2011) and take a critical stance advocating values of de-colonisation and calling for further research into understanding the values that drive ITG and ways to decolonise ITG.

2. Theoretical background

2.1. IT governance

Organisational governance incorporates all instruments and mechanisms to manage and structure an organisation’s direction, aligning the direction with the organisational strategy (L. Chen et al., 2022). Thus, IT governance refers to instruments and mechanisms for aligning and maintaining all aspects of organisational IT with the IT strategy (Peterson, 2004; Weill & Ross, 2010). To this end, decision-making structures, accountabilities and responsibilities concerning IT are defined and operationalised in an ITG system (Peterson et al., 2000). Thereby, the ITG supports and controls the achievement of the IT goals pursued in the IT strategy (Y. Xue et al., 2008). ITG is manifested in relational mechanisms, formal processes and structural arrangements regarding IT-related decision-making (Peterson, 2004). Table 1 gives a short overview of these three elements.

Table 1. Manifestations of ITG structure.

ITG manifestations	Definition
Relational Mechanisms	The exchange between IT and business, dialogues, shared knowledge, and communication (Haes & van Grembergen, 2005; Weill & Ross, 2010)
Formal Processes	IT’s alignment with organisational policies through the definition of formal processes such as IT investments and evaluation (van Grembergen, 2004; Weill & Ross, 2010)
Structural Arrangements	Defined roles and responsibilities concerning ITG, manifested in the degree of decision-making centralisation van (Grembergen, 2004; Weill & Ross, 2010)

The most urgent challenge in designing ITG cumulates in the decision between a centralised or a decentralised design of the ITG (Haes & van Grembergen, 2005; Prasad et al., 2010; Williams & Karahanna, 2013). A commonplace compromise between both extremes is a form of hybrid or federal ITG. Here, structures and decision-making channels are divided between central and several decentralised units (Weill & Ross, 2010). Decentralising governance structures usually leads to stronger participation of all stakeholders and democratisation of decision-making processes (Cheema & Rondinelli, 2007).

The choice of how to solve this challenge and to decide between more centralised or decentralised approaches to ITG is subject to the influence of several factors, which can be divided into three categories.

• First, organisations can have different IT-related goals supported by different governance mechanisms (see Appendix A for a detailed discussion of ITG goals). Organisational IT goals such as flexibility and innovation through and in IT are better achieved in decentralised governance (Tiwana & Kim, 2015; Sambamurthy et al., 2003; L. Xue et al., 2011; Mikalef et al., 2021). In contrast, goals such as IT security, efficiency, standardisation, or synergy call for more centralised structures (Weill & Ross, 2010).

• Second, the internal context influences the ITG structure, too. This includes both the overall degree of organisational centralisation, reflected by decision-making rights, authorities and patterns of communication and collaboration, as well as the general organisational structure and the actual power of the IT function within the organisation (Y. Xue et al., 2008).

• Third, the external environment, reflected in external pressure to increase efficiency or innovation, mimetic pressure to implement “best practice” governance approaches, and the availability of external resources, e.g., access to consultants, may influence organisations in applying certain types of ITG (Y. Xue et al., 2008). Over recent years, the availability of external resources has become more relevant, especially lightweight and consumer IT, which are increasingly finding their way into organisations (Gregory et al., 2018).

Recently, the literature on IT governance (ITG) has increasingly focused on the alignment of IT responsibilities between the IT department and the business, as demonstrated by the research of Schlosser et al. (2015) and Urbach et al. (2019). This discourse has spawned new concepts and frameworks for IT management, including the concept of bimodal IT (e.g., Haffke et al., 2017; Horlach et al., 2017; Magnusson, Koutsikouri, et al., 2020). Bimodal IT refers to the division of ITG into two modes, where in one mode, a centrally managed core IT remains, while in the other mode, the focus is on flexibility and innovation in a more decentralised and agile manner (Toutaoui et al., 2022). Bimodal IT exemplifies the growing attention to federated and increasingly decentralised concepts and perspectives in ITG (Grover, 2022; Leclercq-Vandelannoitte & Bertin, 2018; Vaia et al., 2022). These concepts have in common that responsibility moves away from the IT department and is transferred more to the business. This transfer can foster the business department’s autonomy in IT-related decision-making and thus contribute to a democratisation of the decision-making structures within an organisation (Magnusson, Koutsikouri, et al., 2020). The literature argues that the application of such models becomes more likely, and the transformation towards such models is eased with the use of lightweight technologies, enabling business units to acquire the capabilities to adopt external resources, as technologies stemming from this domain feature a high level of accessibility and low complexity (Gregory et al., 2018). This evolution requires a paradigm shift on the part of the enterprise in terms of the conscious use of IT, the development of IT capabilities, and an awareness of the responsibilities involved, which can be triggered by the use of lightweight technologies and user-driven IT (Koch et al., 2019; Leclercq-Vandelannoitte & Bertin, 2018). Although federated and bimodal approaches have been recognised by Hylving and Bygstad (2019) for their potential to foster innovation, particularly through the use of lightweight IT, these approaches also create tensions that are difficult to resolve (Toutaoui et al., 2022). Increasing shifts in power relations, for example, in the form of user-driven IT through the use of lightweight IT, can create these tensions. These tensions are reflected, for example, in a loss of focus in IT strategy, the possible misfit between agile and traditional IT goals and the scarcity of IT resources (Koch et al., 2019; Toutaoui et al., 2022).

The literature indicates that the internal and external contexts are subject to postcolonial influence. Boussebaa et al. (2012) show that despite representing themselves as transnational and diverse, multinational companies are deeply rooted in their country of origin, which is the global north for most international NGOs. This structure solidifies inequalities regarding staffing and power distribution (Ailon & Kunda, 2009). Moreover, Schwabenland et al. (2008) describe voluntary organisations being forced to adopt centralised management structures due to environmental pressure, omitting intended diversity. The case exemplifies how a postcolonial environment shapes organisational structures by forcing conformity to the mainstream. Organisational research, in general, has shown that the organisational structure is dependent on internal and external conditions and contexts that also shape the degree of centralisation through persistent and iterative alignment processes with external contexts (Hargrave & van de Ven, 2006; Thornton et al., 2012; Walsh & Ungson, 1991). If these are subject to postcolonial thinking patterns and assumptions, these may also be incorporated into internal structures and the corresponding ITG structures.

2.2. Postcolonial theory

PCT is a means of critical reflection calling attention to power asymmetries in social contexts stemming from colonialism, also used in IS research (Myers & Klein, 2011). This critical discourse started in the literary sciences with the book Orientalism by Edward Said. He introduced the idea of a carefully shaped narrative that enforced the hegemony of the west over the east (Said, 1978). For this research effort, we focus on the three authors most prominently discussed in PCT research (Tsibolane & Brown, 2016): Said, Spivak and Bhabha.

Edward Said (1978) was the first to examine the nature of the discourse in which the members of the hegemonic culture (the mainstream) impose their views on individuals at the social margins (subaltern) and thereby shape social reality. He drew on the discourse concepts by Foucault, but his focus was on how the discourse is used to shape an image of the “other” and to establish the actions of those in power as morally and intellectually correct. On the example of depicting the Orient in the west, he highlighted how the image was shaped to legitimise colonial ambitions. Here, Said builds on the paternalism concept Frantz Fanon derived based on his colonial experiences, e.g., restricting the subaltern’s decision freedom for their own good (Fanon, 1963). Hegemonic discourse and paternalism could affect NGOs’ IT decision processes if these are shaped so that mainstream decision-makers holding the resources and power in headquarters in the global north are positioned to make IT decisions for the subaltern in the global south – as they would, for example, do in a centralised ITG structure.

Gayatri Chakravorty Spivak (1988) concentrated on the subject of the subaltern in her work. A Marxist by training, she started with Gramsci’s (1971) conceptualisation and asked prominently, “Can the subaltern speak?” She posits that the subaltern do not have an effective voice to speak for themselves. They are not listened to and can thus not “complete the act of speech”. But it is also important to note that there is no uniform subject of the subaltern. They are influenced by their specific postcolonial context (Ashcroft et al., 1998; Spivak, 1988). Marginalising or devoicing the subaltern could be related to very centralised ITG structures with insufficient participation or consultation processes for organisation members in the global south.

Homi Bhabha challenged the simplifying duality of the mainstream and the subaltern. Following Bhaba, the context and the actual circumstances of the colonial relationship need to be considered. For example, the subaltern can exhibit a range of reactions to the culture, values and beliefs of the mainstream, which range from imitating or mimicking them (mimicry), over having positive and negative reactions simultaneously (ambivalence) to the emergence of a mixed culture or third space (hybridity) (Bhabha, 1984). Mimicry, ambivalence and the emergence of a third space are all potential behaviours of organisation members in the global south to NGOs’ IT decision processes. A true third space could emerge if the lines between former colonists and colonial subjects blur by having decision power distributed more equally, e.g., by moving headquarters to the global south as some NGOs such as Oxfam have done (Radojev, 2017).

IS researchers have adopted the ideas of PCT – most prominently in the literature stream around ICT for development or ICT4D (Chipidza & Leidner, 2019; Tsibolane & Brown, 2016) and IT offshoring (King & Torkzadeh, 2008; Ravishankar et al., 2013; Sayed & Agndal, 2022). At the same time, a dialogue within the IS discipline is emerging regarding research methods for examining marginalised groups (Myers et al., 2019, 2020; Ortiz et al., 2019) and how to improve the IT research discipline and teaching methods accordingly (Chughtai et al., 2020; Romano et al., 2023). The ICT4D research stream focuses on examining the effects of single ICT projects aimed at developing the communities in which they are conducted (Chipidza & Leidner, 2019; Lin et al., 2015; Walsham, 2017) and has thereby clearly established that individual IT decisions can be postcolonial. However, an examination of the IT decision structures and processes inside the organisations involved in such projects as a whole with a PCT perspective is still missing (see Table 2). What we have are first valuable explorations into the effect of the international context and the relationships between individuals in organisations from different countries in the offshoring context (King & Torkzadeh, 2008; Ravishankar et al., 2013; Sayed & Agndal, 2022). However, here we are yet missing a postcolonial perspective on IT decisions, IT-related decision-structures, and IT-related decision-processes in international organisations in general. With our research effort, we strive to close this gap. We employ PCT here as a means to critically examine if the existing IT decision structures perpetuate colonial patterns, thereby upholding postcolonial power-imbalances, which could adversely affect decision outcomes.

Table 2. Studies using PCT in core is journals.

Source	Titel	Focus and gap regarding our research question
Lin et al. (2015)	Extending ICT4D Studies: The Value of Critical Research	Ethnographic study and critical reflection using a PCT perspective on a single ICT4D project intended to improve the education and social welfare of the aboriginal people in Taiwan. Does not cover organisational structures in their postcolonial perspective.
Ravishankar et al. (2013)	Information technology offshoring in India: A postcolonial perspective	Explores the experiences and responses of an Indian vendor organisation to asymmetries of power in an international business context using postcolonial theory concepts as a theory lens. No specific focus on ITG.
King & Torkzadeh (2008)	Information technology offshoring: Research status and issues	Assessment of 43 papers dealing with IT offshoring – theoretical lenses identified include postcolonial theory. No detailed study of mechanisms and effects due to the meta-level of the publication.
Chughtai et al. (2020)	Demarginalizing interdisciplinarity in IS research: inter-disciplinary research in marginalization	Reports from a workshop that explored how interdisciplinary collaboration can help study the use and shaping of digital technologies by marginalised Indigenous communities on a general level. No focus on IT decision mechanisms but on research methods.
Myers et al. (2020)	Studying the other or becoming the other: Engaging with indigenous peoples in IS research	Reports from a panel discussion that explored the ethics and politics of the engagement with Indigenous peoples in information systems (IS) research. No focus on IT decision mechanisms but on research ethics.
Kane et al. (2021)	Avoiding an oppressive future of machine learn-ing: A design theory for emancipatory assistants	Uses emancipatory pedagogy developed to overcome postcolonial structures in Brazil to develop four design principles for emancipatory assistants. No focus on ITG but on machine learning and related oppressive futures.
Myers & Klein (2011)	A set of principles for conducting critical research in information systems	Proposes a set of six principles for conducting critical research, including using postcolonial theory in information systems based on the three main critical research streams by Bourdieu, Foucault and Habermas. No focus on IT decision mechanisms but on critical research in general.
Ortiz et al. (2019)	Giving voice to the voiceless: The use of digital technologies by marginalised groups	Reports on a workshop that explored cyberactivism and research methods for examining how marginalised groups use digital technologies. Marginalisation is hereby theoretically rooted in postcolonialism. No focus on IT decision mechanisms but on research methods.
Romano et al. (2023)	Indigenising the IT Curriculum by Design	Case study on how to indigenise the syllabus of an Information Technology (IT) unit as a part of a university-wide project at an Australian university in answer to postcolonial structures. No focus on ITG but on IT/IS curricula.

As we highlighted above, ICT4D research focuses on individual projects rather than the internal (IT) structures of NGOs. Similarly, NGO research has not yet combined the PCT perspective with a focus on IT. Nonetheless, PCT offers an effective lens to study NGOs’ actions due to their work contexts. Several authors use PCT to question how far NGOs fulfil their ambitions. Critiquing NGOs’ abilities to foster civil society and democratisation is not new (Mercer, 2002), but remains an important issue. Frewer (2013), for example, finds that NGO employees claiming to create good governance and establish a civil society in rural Cambodia fall short of their aim. As Jackson and Adarlo (2016) show, this can have severe consequences. They found that the NGO image is, for example, difficult for volunteers struggling with cross-cultural views. In addition, PCT is also used to examine power structures in international NGOs. Makuwira (2018), for example, criticises that the agenda for development work is defined in the West and that donor funding plays a dominating role in structuring NGO work. But NGOs are already working on decolonising aid. Key issues are the representation of the subaltern on governance boards (Worden & Saez, 2021), aligning incentive structures, and alleviating power imbalances (Konyndyk & Worden, 2019). Moreover, NGOs like Doctors Without Borders have, for example, started to critically reflect on their treatment of local staff and patients in the global south (McVeigh, 2020). Similarly, NGO research showed that IT decisions might follow the interests of donors from the global north (Godefroid et al. 2023). It is important to note that making different decisions is not inherently better, and decolonisation efforts can have unintended adverse societal effects, e.g., when micro-activism negatively impacts people or organisations (Young et al., 2019).

Based on our reading of the PCT literature and its application in IS research, we present our core concepts for this research in Table 3.

Table 3. Core concepts of PCT relevant to this research effort.

Aspect	Definition	Example from PCT	Example from IS
Main-stream	Individuals at the centre of culture, knowledge, and civilisation have the most power and resources Ashcroft et al., (1998); McLeod, (2000); Said, (1978).	Said (1978) analysed how painters and writers shaped a specific narrative of the Orient.	Lin et al. (2009) examined how a non-aboriginal school principal and the media controlled the narrative of an ICT educational project deployed in an elementary school in an aboriginal village in Taiwan.
Subaltern	Individuals who are removed from all lines of social mobility or, in other words, outside of the mainstream at the margin of culture, power, and civilisation Ashcroft et al., (1998); Spivak, (2005).	In her famous essay “can the subaltern speak”, Spivak (1988) concentrated on the perspective of Indian women and how their perspectives were marginalised within historical narratives.	In a study on Indian call centre workers, Pal and Buzzanell (2013) examine the reactions of the subaltern to systematic power imbalances.
Pater-nalism	The mainstream assumes they have the right and obligation to assimilate (help) those at the margin, often without consultation with the subaltern Fanon, (1963).	In the Algerian political struggles, Fanon highlighted paternalism when oppressed communities needed to acknowledge the west’s sense of moral before the legitimacy of their struggle was recognised (Fanon, 2008; Slisli, 2012).	In the ICT project in Taiwan, Lin et al. (2015) found a strong paternalistic view that the aboriginal community needed connectivity to further their development.
Mimicry, Ambiva-lence and Hybridity	The subaltern can exhibit a range of reactions to the culture, values and beliefs of the mainstream, which range from imitating or mimicking them (mimicry), over having positive and negative reactions simultaneously (ambivalence), to the emergence of a mixed culture or third space (hybridity) Bhabha, (1984).	Bhabha (1984) outlined how the Indian people reacted with a range of behaviours to the oppression during the English colonial rule. With this perspective, he challenged the historical narrative of the Indian people as the passive subject of their colonial rulers.	Looking at technology offshoring in India, Ravishankar et al. (2013) examined the hybridity in work practices that emerged in a knowledge management project. They highlight the impressive resourcefulness and strategic capabilities of the subaltern.
De-voiced	The subaltern is devoiced by denying them an opportunity to be listened to (1988).	Spivak (1988) highlighted how, on the subject of the Indian widow burnings, only the Brahmans and the English colonialists spoke – nobody ever asked the widows.	In the abovementioned project, Lin et al. (2015) also found that the notion of “helping” the marginalised ignored the actual local needs, and voices raising reasoned criticism were deliberately not heard.

3. Method

3.1. Data collection

To address our research question, we conducted 27 semi-structured expert interviews representing different NGOs operating locally and internationally.

To select our interview partners, we used purposeful sampling with a hybrid approach combining theory-based and maximum variation sampling (Suri, 2011). We composed our sample along a set of criteria detailed in the following. Table 4 consolidates all criteria based on the individual selected interview partners. We included representatives from two groups of NGOs: locally operating NGOs focused on one country only (group 1) and internationally operating NGOs (group 2). This is grounded in postcolonial theory that predicts that postcolonial structures manifest in the relationship between former colonist states and new nation-states mostly located in the global south (Bobby Banerjee & Prasad, 2008). Therefore, we assume that if postcolonial structures manifest within NGOs’ ITG, they would only do so in international NGOs, as these unite members from both the global north and the global south. For comparison, we also interviewed representatives from locally operating NGOs, i.e., NGOs operating only within a single country.

Table 4. Overview of interviewees, including organisational details.

#	Group	Country	Organisation cause	Job Title	Employees
1	Group 1: Representatives from local NGOs	Germany	Dance lessons for young people	Volunteer	7
2			Community work	Volunteer	10
3			Climate activism	Project manager	46
4			Community work and school	Senior manager	50
5			Charity	Administrative staff	51
6			Homeless care	Head of a local branch	70
7			Social care	IT representative	690
8			Peace movement	Head of IT	250
9	Group 2a: IT leads of international NGOs	US	Refugee/ Emergency relief	Head of IT	2,014
10		Ireland		Head of IT	3,900
11		US		Head of IT	5,000
12		Austria	Humanitarian/Development aid	Head of IT	39,000
13		Bolivia		Regional head of IT	10,000
14		Estonia Netherlands		Regional head of IT	10,000
15		US		Head of IT	37,000
16		Spain		ICT4D team lead	1,200
17		Ethiopia		Regional head of IT	100
18	Group 2b: Other managers involved in IT decisions of international NGOs	Austria		Head of Strategy	39,000
19		Ethiopia		Regional director	100
20		India		Country director	1,700
21		India		Regional strategy advisor	100
22		Sierra Leone		Country director	800
23		Nigeria		Country director	1,000
24		Germany	Wildlife preservation	Project manager	338,000
25		Malaysia US		Project Manager	7,000
26		Netherlands	Refugee relief	Specialist Programme Development and Support	1,000
27			N/A (interims manager)	Interims manager	N/A^2

We defined criteria for selecting organisations within both groups. We relied on the German market for group 1 to avoid cultural differences distorting the findings. With this, we deliberately chose a rich country from the global north to ensure that we could focus on sector specifics, not socio-economic effects like a general scarcity of resources. Moreover, one of the authors has worked in this sector for the past five years and was, therefore, able to ensure a representative sample. For group 2, we aimed to maximise variation concerning the country of origin. Within both groups, we opted for a maximum variation of the considered organisations regarding the organisations’ cause to avoid organisation-specific biased findings (Patton, 1990). Therefore, we aimed to include NGOs with humanitarian, ecological, developmental and other causes. For the same reason, we also decided to set organisation size as a selection criterion and aimed for variation here. Naturally, international NGOs are larger than local NGOs (i.e., NGOs only operating in Germany).

After choosing respective sample organisations, we identified knowledgeable informants within each organisation. Alvesson and Kärreman (2007) describe that knowledge cannot be separated from the person developing or working with it. Therefore, researchers have to carefully evaluate the choice of interview partners according to their field of knowledge (Palinkas et al., 2015). In our case, we were especially interested in the design of the respective ITG and the underlying reasons for IT decisions in the international NGOs (group 2). As such, we chose to select interviewees from different groups: Firstly, IT-leads who were either in charge of IT or specific IT-related topics, e.g., as the head of IT (group 2a), and secondly, other managers involved in IT decisions, e.g., as country director (group 2b). All interviewees were closely involved in developing the ITG structures or were directly concerned with steering the strategic IT decisions of the respective organisations. Our interview series ended in saturation, with only known aspects emerging in our last interviews (Guest et al., 2006).

The interviews were conducted between February 2021 and October 2022 using MS Teams or Zoom and lasted for an average of 50 minutes. The interview language was German or English, depending on the interviewee’s preference. Our interviews followed a semi-structured approach (Schultze & Avital, 2011), focusing on the use and governance of IT within the respective organisation and the five aspects of PCT (see Appendix B). We did not directly ask about PCT concepts or the design of ITG structures to avoid participants giving socially desired answers. This approach is in line with the guidance by Alveson and Kärreman that “rather than assume that the subject is reporting authentic experiences, we can see the subject as a politically motivated producer of what are, for him or her, favourable ‘truths’” (Alvesson & Kärreman, 2007, p. 1269). Instead, we asked about related topics, e.g., participation structures for field staff. We chose this approach as we got very adverse reactions when we discussed the topic openly in prior discussions with NGO representatives. All interviews were recorded and transcribed in consent with the interviewees to allow for further analysis. All information that can be used to identify persons or organisations is anonymised. This step allowed interviewees to respond without the fear of leaking confidential information. In the following, we refer to the interviews by abbreviation, e.g., I1 refers to interview number 1. We accompanied the information gained from the interviews with other sources of evidence (e.g., archival documents and general industry insights) to allow for cross-triangulation (see Appendix C for our data validation measures).

3.2. Data analysis

In data analysis, we followed the approach suggested by Alvesson and Kärreman (2007), building upon abduction (Peirce, 1978). We first analysed the IT governance structures in both national and international NGOs. To this end, we engaged in MAXQDA-based coding as a means of “more systematic work to develop our new understanding” of the phenomenon (Alvesson & Kärreman, 2007, p. 1272) based on existing theoretical understanding from ITG and PCT. We hereby follow the suggestion by Alvesson and Kärreman (2007) that ”reflexivity can be encouraged by using various theoretical and metaphors […] working with […] a different theoretical framework, and thus increasing the chance to be challenged when encountering empirical material”. Following Alvesson and Sandberg (2023), we triangulated our findings between theory, empirics, and our pre-understanding in the form of personal scientific and practical experience in the domain. The theoretical perspective was brought in using ITG and PCT lenses. By transferring theoretical assumptions to the practical context of NGOs, this approach succeeds in problematising theoretical correlations and assumptions that are regarded as given by introducing a perspective that is initially foreign to the domain of ITG in the form of postcolonial theory (Alvesson & Kärreman, 2007; Alvesson & Sandberg, 2020).

Two researchers read and coded all interviews separately to ensure rigour within the codes. With this approach, we ensure theoretical relevance from the beginning of the coding process but allow for data-driven research (Fereday & Muir-Cochrane, 2006). After the first coding cycle, all results were compared between the two researchers, allowing for a true discourse with the empirical data (Alvesson & Kärreman, 2007). Regarding inter-coder reliability, our coding results show a matching rate of 81%. In case of conflicting codes, the respective section was discussed within the group of researchers. To this end, each researcher re-read the corresponding text and reframed their position, intending to take incoherencies and contradictions seriously (Alvesson & Kärreman, 2007, p. 1269). Also, as one researcher has specific domain knowledge and experience, a potential confirmation bias influencing the interpretation of data needs to be considered. Therefore, all material was coded by both coding researchers, as the second researcher has no previous practical experience in the field. We identified and compared the observed ITG structures, formal ITG processes, and relational ITG mechanisms of both types of NGOs (national and international).

Through this coding and first analysis, several observations emerged where IT governance structures or mechanisms differ between local and international NGOs. Such occasions were taken as “signs of mystery that stand up to scrutiny” that were used to create and describe the phenomenon of overly centralised ITG (Alvesson & Kärreman, 2007, p. 1271; Alvesson & Sandberg, 2023). In order to shed more light on this phenomenon, all texts were reread and now specifically set in relation to given influence factors. In addition, research notes were prepared, discourse within the research team was stimulated, and interview statements were compared with secondary data. We specifically looked for incoherences or contradictions that could be explained through one or both theoretical perspectives of ITG and PCT. In doing so, we iterated between theoretical structures, empirical evidence, interpretation and critical analysis (Alvesson & Kärreman, 2007; Orlikowski & Baroudi, 1991). Appendix D shows an extract of our coding structure and describes the procedure of data analysis more practically.

4. Findings

4.1. ITG: observed ITG structures – differentiating international and regional NGOs

Within our data set, we found a majority of international NGOs applying centralised ITG regarding their structural arrangement. Most IT decision-making occurs at headquarters in the global north: “Some NGOs have a very strong head office where they organize everything” (I26). Only some international organisations allow local units a certain degree of autonomy in choice and decision-making but, nonetheless, set a strict frame: “Because almost by rights you are trying to democratize things. And you are trying to give people the chance to build these things and solve their own problems, build their own solutions, but at the same time, you have to develop like a cohesive or kind of coherent infrastructure for people to use” (I18). Formal processes, like the introduction of used software solutions, are, to a high degree, located in centralised departments and executed in a top-down approach without the involvement of local units: “The majority of systems, especially global systems, are prepared in Europe and then tested globally” (I14). This centralisation is maintained, although some organisations are well aware that it supports the manifestation of centralised structures disregarding innovations and experiences from local units: “I’ve heard of, and I’ve seen organisations that do not allow that discussion [about IT with local units]. No, this is the way it’s done. That’s the way IT is done well” (I10). On the other hand, relational mechanisms, like unofficial coordination, often occur in parallel: “We have some collaboration projects in Africa or Asia. I think it’s good that two [units] can collaborate and learn from each other and bring better effect to the beneficiaries” (I15). International NGOs operate to a high degree in centralised decision-making structures and apply centralised processes and mechanisms.

In contrast, for exclusively local operating NGOs, we observed different patterns. Here, we find much more democratic and decentralised structures, embracing the participation and active involvement on the part of employees and activists in the field and from all areas of the organisation: “As an organisation, it is important for us not to be perceived at headquarters as an ivory tower where knowledge is born. It is important for us to also look at projects that are born in the field and that may have a global impact. How can we develop them, what do our field workers say” (I7). Some local NGOs explicitly welcomed innovation and ideas from local staff and integrated them into formal decision processes: “Under the rolling wheel, we have made decisions and changed processes. All of us together, too. Because we have a very, very lively and interested workforce that is interested in the quality of their own jobs, and they said, I’ve just tried this out. It worked. And then we said okay, fine. We are happy. What exactly did you try out? And then we looked at the things and said okay, that actually works” (I8). Processes are not so strongly formalised but are very recognisable on a relational level and often lived as rather integrative relational networks: “There is actually almost something like a permanent network that meets regularly and exchanges ideas. So that’s our M&E Network, where we really say what works and what doesn’t” (I7).

In general, our interviews suggest that local NGOs emphasise stakeholder participation and involvement using decentral or hybrid structures and respective formal processes and relational mechanisms, while international organisations use a more top-down approach with centralised structures underpinned by restrictive formal processes and relational mechanisms (see Appendix D).

4.2. ITG: influencing factors

Regarding organisational IT goals, NGOs strongly focus on IT security, as these organisations often work with extremely sensitive (beneficiary) data. An example would be the health data of refugees obtained during initial care upon arrival (I9). An interviewee from a refuge relief organisation described: “We’re dealing with beneficiary information, we’re dealing with compromised people. And so the data exchange, the personally identifiable information, that type of information has to be secure” (I9). This need for data security is driving organisations to impose rather strict IT standards in the form of a more centralised ITG. Both the use of specific software solutions and process-related regulations are common: “We have to set guardrails, and those guardrails are, okay, you want a program or an application to do X, great, we will work with you, we will figure out if there’s something within the ecosystems we already have” (I26). In this line, the unmonitored use of technology can pose a threat to the integrity of the organisational IT: “So shadow IT comes up as a difficulty, because you want people to have the right tools, but not necessarily those they request and definitely not without your permission” (I11). These tools and solutions are typically defined centrally and applied similarly across the regional units. This tendency to centralise and regulate the use and development of IT undermines emerging approaches to user-driven IT. Another aspect mentioned in this context is the risk of a general reduction in (security) quality within the IT organisation: “And otherwise, of course, you can fudge something together with volunteers and computer science students on site. The only problem is that this de-professionalizes the work” (I9).

The goal of maximising IT security does not exclusively drive this quest for standardisation but is also influenced to a certain degree by economic considerations. As NGOs rely directly on the goodwill of donors to sustain their operations, they try to get out the best for the beneficiaries with the given financial resources and to avoid waste whenever possible: “But if you can spend your money more effectively, you can support more people. And this flatters the constituency, I think. So, it can improve your fundraising ratio by improving your customer experience” (I18). NGOs thus pursue efficiency goals by optimising IT-based processes and the related customer experience within the donor countries. Moreover, the interviewed organisations use technology to enhance productivity in back-office tasks: “And then technology, obviously, that helps us to be more effective at what we do. So that would be kind of back-office technology” (I24). We found evidence for a separation in the governance structure between systems applied in the global organisation and solutions only applied locally. In this case, the decision-making power is built upon funding streams: “If we are going to do federation projects, we’re going to use a federation system. If you are receiving international money, you have to use it just to be transparent to the organisation as a whole. If you are generating your funds fully locally, or not getting a penny, go on do your business” (I19). We also found that NGOs strive to achieve IT innovation and flexibility goals, demanding decentral governance structures. These goals and the efficiency goal mentioned above reflect the desire to provide the best possible assistance. As internationally operating NGOs face various circumstances and local conditions, they strive to adapt as best as possible. To this end, regional bodies must be granted freedom and decision-making powers, i.e., decentralised structures must be established that allow local staff and bodies individual scope for IT decision-making.

The organisations’ external environment also influences ITG. External pressures are forcing NGOs to redesign certain software solutions or processes at the local level. First and foremost, the regional legislature strongly influences ITG. Some governments regulate the use of certain tools: “Syria [and some other countries] are sanctioned by the US, so we could not use everything there. We had to ask for explicit allowance to bring the tools into that country and start using that” (I18). On the other hand, these countries may require a certain form of reporting, making alternative local solutions necessary: “But also the state requires reports. The local state. They need the reports as well in some particular format. The requirements of governments are different; it can even be on the level of the municipality” (I15). Our additional sources of insight indicated the existence of movements that put pressure on international NGOs to discuss their own social inclusiveness (e.g., Charity So White, 2022). These could be understood as another form of external pressure potentially influencing ITG structures. Our interviewees acknowledged this external pressure in passing: “[T]his whole decolonization, how [do] we organize ourselves, how was our staff management, etc., is really, really a hot topic in the sector” (I27). Moreover, the availability of resources influences the design of ITG to a certain degree. NGOs are dependent on multiple factors when operating in the field. First, the regional infrastructure influences the selection of technologies used and the decision-making for or against a certain solution: “In terms of access, I guess more or less, maybe we have a couple of points in which maybe we need to improve our access but depending on the location again, for example, Venezuela’s daily problems with electricity, that’s something outside of our control” (17). But we also found regional differences in education and the availability of skilled staff to be influential: “There are multiple factors that influence why we prescribe a lot, first of all, financial reasons, but also regional experience, skill, and knowledge, as well as political and geographical factors” (I15). In addition, there are mimetic pressures within the domain of NGOs, forcing organisations to adapt to widely used standards: “So I think I would also say that most of the NGOs are playing catch up, I guess, I think, and especially the big NGOs” (I9).

However, we also find that underlying organisational structures and NGOs’ overall internal context influence ITG. Organisations state that their ITG structures grew according to the overall organisational structure: “I am currently the link between the ICT and the country offices. Our HQ being in London, so I link up the rest of our offices outside of the UK, according to our organisational structure” (I15). This structure likely stems from the organisation’s history. As mentioned, financial considerations also play a strong role in NGOs’ daily business, as they mostly rely on external funding. One interviewee from an international NGO explained this in a nutshell by emphasising the lack of capabilities of regional or country-level divisions to launch projects due to scarce financial situations and missing fundraising efforts: “The vision is there, but all the ideas, all the initiatives, all the projects, all the innovations to move our division forward, we don’t have the finances that go with it. And our fund development and communication activities are extremely low” (I22). This funding predominantly originates from the global north, which means that decision-making processes and structures are closely linked to the needs and requirements of this region to satisfy donors and other opinionated stakeholders in this domain: “It is very difficult to convince people to invest in [IT] – it is not the cheapest sector, you have to convince to invest hundreds of thousands of Euros into that and give them the understanding that we need to care for our beneficiary data” (I18).

The emphasis on centralised structures reported above creates a mismatch between the objectives pursued and the existing ITG structures for global NGOs – a “mystery” we cannot explain with ITG only. While central governance structures fit well with security and standardisation goals, efforts for flexibility and innovation are undermined or inhibited. The internal context and the external environment would, from a sheer ITG perspective, call for a more federal or hybrid model, even for core systems, to prevent power imbalances. Instead, we found that most globally operating NGOs apply an ITG with a centralised structure, where the influence and needs of stakeholders from the global north are accentuated. While the difference between local and global NGOs is quite clearly explainable with a more homogenous environment and more stable circumstances, this can also be explained by the lack of influences of the postcolonial context on the organisation’s structures. Examining this ambivalence further requires a review of existing structures for postcolonial evidence in the respective international NGOs.

4.3. PCT: mainstream

NGO funding usually originates in the global north, giving donors and their representatives in these countries significant power. Subsequently, (perceived) donor interests shape ICT investment and use: “We know that more and more donors, especially institutional donors, corporate donors, they want to have [data transparency]” (I24), leading to subsequent investments in tools that serve these interests. This pressure also changes the overall management culture: “When you ask our management, […] they’re very process-oriented. But when you ask them about the data or the KPIs. They are a little bit, you know, because […] with data, you create transparency about yourself. […] But it’s getting better because of donor pressure” (I24). Surely, more data transparency is a justified request by donors and a valid use of IT; however, one could question whether this is the best use of scarce resources in line with Makuwira’s (2018) challenge of the NGOs’ understanding of local needs. However, more problematic is that the need for transparency leads to more centralised IT decision-making and standardisation, increasing the decision power of the mainstream even further (I24).

In addition, there also appear to be (western) financial interests at play both for IT companies and the NGOs themselves. In contrast to an ideal world where only those IT solutions are selected that best fit the purpose, other factors also play a role: “Of course, foreign policy and development policy and economic policies are entangled to some degree. It is an aspect because certain tech is adopted, and an obvious one. If you have a billion-dollar company, of course, they have ways to push their tools in the development sphere. They know people at embassies, foreign offices” (I14). A selection based on political interest and not actual task-technology fit is potentially problematic as one of our interviewees stated that in his organisation, for example, 95% of solutions are externally sourced from vendors (I1). But also regarding the IT use across the NGOs activities, we can observe a bias towards activities conducted in the global north – primarily investing in fundraising activities: “I would say that the main part of ICT investment is on the fundraising part already. Because there, you need the most expensive systems, […] So, most of the time, the field intervention, program design, and program development is financed via external partners” (I18). Such a focus is logical as the amount of funding decides the NGO’s subsequent action possibilities. However, it seems to slightly contradict the clear focus of the IT leaders we interviewed: “I’m not here to fix PCs. I’m not here to do anything but to make our services and our work for beneficiaries better” (I24).

Finally, as described in the prior section, IT centralisation gives key decision-makers in the global north (the mainstream) the power to decide what everyone else needs: “So executive level, they make contracts with an IT supplier. They mostly do the implementation, not really in an agile way. And then, they start training the users in the new system. […] So, people get a fixed system” (I18). As key decision makers control most resources available for IT, they make decisions for the subaltern based on their perception of local needs. But here, we also observed change, even if it might be slower than some interviewees would like: “Ground-up development of tech does not happen very often. Still, western NGOs and companies develop tools and try to be inclusive in a way. But it is certainly not ground-up. […] I see the change, but I think it is lower than it’s being proclaimed. We are too far from any satisfactory situation” (I14).

4.4. PCT: subaltern

Regarding perspectives on the subaltern, we found that they were simultaneously seen as unskilled or lacking resources and a source of innovation. With this, we understand the subaltern to encompass both field staff and participants in the global south. We found significant differences between the perception of individuals from international headquarters (I8, I24, I12) and those closer to the field (I10, I14, I15). While interviewees from headquarters, in line with their function, focused on developing and implementing tools for the whole organisation from a centralised view, local interviewees focused more on differences and how to help members with different development trajectories. Even though interviewees from headquarters were aware of specific innovative projects, the general view was that the subaltern needed help (and thus centralised and standardised decision-making) and was rather dominated by the perception of weaker organisation members. Such a perception could explain the abovementioned observation that international NGOs use a more top-down approach, allowing for less participation in decision-making.

Regarding a lack of IT skills, one of our interviewees pointed out: “In many countries, we really need to start with basic things, like Syria, for instance, or Kyrgyzstan. But we don’t need to do it anymore in the Baltics, Poland, or Czech Republic, where they are already getting a good education from the state. This is the difference” (I15). But this can still mean that “in some villages far to the east, you have to explain what an email is” (I15). Interviewees who were more closely involved with the global south often emphasised significant differences across countries (I18, I26, I10). While headquarter members shared perceptions like “the constituency is not aware of the importance of ICT and not willing to pay for it” (I18). An issue that also arose was the availability of IT resources: “There are people who have such proper equipment, and everything is wonderful, but not everybody has something like that” (I5). However, skills and understanding of IT security needs were also perceived as an issue. These are especially important as NGOs serve vulnerable communities and risk losing their funding in case of any data leaks. Two interviewees highlighted that the subaltern tend to use any “mom and pop” IT without proper security if left unsupervised (I9, I10). We are sure all our interviewees referred to personal experiences and made valid observations. However, the focus should be on upskilling co-workers and participants – not making decisions for them. A task that some international NGOs are already engaged in (I15).

Nonetheless, the subaltern is also perceived as innovative and motivated: “For example, in South Africa, regarding data management, the big majority of people are young, and many of the young people are pretty well qualified – with no real job opportunities. I have met a lot of people using those tools – they learn quickly and need some training” (I14). In line with this perception, we also encountered deep respect for the innovativeness of the subaltern: “The folks out in the field have worked with minimal, everything. And so, they’re all MacGyver out there” (I8). One interviewee even perceived co-workers in the global north as less technology-friendly: “My impression is that people in the civilized world are actually very critical. I think that in developing countries, the opportunities rather in the foreground, […] that outside Europe, technology is already perceived very, very positively” (I12).

4.5. PCT: paternalism

Paternalistic structures still exist in NGOs. For example, in some organisations, the social hierarchy still appears to defer to western personnel: ”One is somehow, unfortunately, the white person who comes from Europe to now explain things. […] the mentality of many people here still is so that you are simply taken more seriously if you somehow come from afar and bring knowledge” (I1). But apart from such blatant post-colonialist tendencies, there are also more subtle views regarding the education of the subaltern. For example, in a discussion on user-driven technology, an IT leader stated, “We need to recognize […] digital opportunities. We need to develop a mindset in people so that they actually recognize how that problem that they face or how that process that they are doing [could be digitalized]” (I24). Similarly, another interviewee stressed that especially younger participants must be trained in IT security (I5). This attitude is also reflected in the perception of a need for centralised decisions to ensure IT security (see above). The mainstream also assumes they have the right and obligation to help those at the margin implementing the systems the mainstream deems necessary to achieve digitalisation: “If a country actually refuses to use certain technologies, we cannot enforce them. And that, of course, makes our digitization process very difficult” (I12). A perspective that could motivate less participation and involvement of the local level in global decision-making through formal processes, which would force a confrontation over paternalistic ideas. Again, we found that interviewees on a local level saw less need for “enforcing” digitalisation (I15).

Additionally, we found a more technical version of paternalism. Big IT providers and the NGOs’ headquarters that negotiate with them assume that their solutions can be deployed worldwide without issues. It appears, however, that they are optimised for their largely western clientele. In that regard, one of our interviewees observed that IT solutions did not work as well in his countries – potentially “because corporations or big organisations, maybe, they don’t have too many clients in certain countries like Jamaica, Bolivia, Nicaragua, Honduras, so that also creates a lot of dilemmas and problems for the digital agenda for us” (I10). Here it is again the mainstream – in this case, both the key decision-makers of large IT providers and the NGOs from the global north – who decide what the subaltern needs without consulting them.

4.6. PCT: mimicry, ambivalence and hybridity

In the following, we detail our findings on the spectrum of different behaviours the subaltern exhibited in reaction to mainstream actions. Looking for patterns of mimicry proved to be slightly difficult, but we found instances which indicate that this effect is also at play. For example, when we spoke to representatives from country offices, it was difficult to discern whether they truly appreciated the IT initiatives driven by their headquarters or mimicked the hegemonic discourse’s opinions and actions. In line with this observation, we found that country offices not receiving large-scale subsidies from headquarters were much more critical than those that did (I20, I22, I23). Another effect we observed was local decision-makers not pushing IT initiatives, which would serve country-specific interests of the subaltern, and justifying it with “mimicked” efficiency goals: “We don’t push for having micro individual initiatives at country level because it’s proven with a lot of studies and research that this is not scalable and at the end the investments done there disappear” (I16). While this might make much sense from the view of international headquarters, it does not answer Bhabha’s (1984) call to account for the unique contexts and differences within the subaltern in any way. Again, from an ITG point of view, such rationales make perfect sense, but interestingly, the subaltern adopts this narrative, which justifies not addressing local needs.

In the smaller NGOs, we found a general ambivalence regarding the mainstream push for more efficiency through digitalisation (I4, I6, I7). For example, one of our interviewees from a small local NGO argued, “The impetus for digitization. […] we’re less focused on it in the social sector, […]. Because economic interests play less of a role for us, and perhaps that’s why the question of efficiency is more of a secondary consideration” (I7). A feeling that the large international NGOs apparently do not share (I18, I10, I15). However, we did find a certain ambivalence regarding the available funding for ICT projects. When asked if local ICT projects receive less funding because, for example, large-scale ERP projects are favoured, the answer was quite ambivalent: “I’d say yes or no. […] There are good years, and there are bad years. There are good years where we got good funding for our digital projects. And there are years where it’s like OK, now we need to re-prioritize someone else” (I17). Apparently, the important aspect was that there was always some budget for local needs, making the adverse prioritisation less problematic (I19).

What we did find quite explicitly was the emergence of NGOs as a third space with some level of hybrid culture. There was a general perception that NGOs as a whole are marginalised regarding ICT. Our interviewees perceived their workplace as a “low-tech environment” (I2) and to “lag behind the private sector” in their IT use (I8). This perception is underlined by the charity efforts of big tech companies: “Of course, you have to always deal with budget limits. But that’s not a big issue, […] Salesforce and Microsoft […] there are great discounts for non-profit organisations” (I18). Thus, it seems that NGOs perceive themselves as a whole as needing help regarding their IT solutions (I8, I26). This perception makes sense if not so long ago decision makers at all levels of the organisation prioritised delivering physical resources to program participants over ICT expenses: “But you can speak to the wall, if the people on the ground, they are not listening to you. Because they make comments like which one is important, the milk for children or the licenses that you require? And I’ve heard this personally, like 12/13 years ago” (I17). But this perception is changing, and NGOs perceive themselves to have to do “a lot of catching up” (I10).

4.7. PCT: de-voiced

As outlined in the ITG section, most IT decision processes appear to be centralised to increase standardisation, and the users (in the majority part of the subaltern) are given little opportunity to raise their voices. They are neither involved in the selection of the tools nor their development. This systematic devoicing leads to standardised solutions that are not optimal for user needs and are thus often ill-received (I18, I14, I10). Most NGOs have official consultation processes, but these do not necessarily fulfil their intended purpose: “They come, and they do consultation, but the consultation process is almost like just to tick the box. If I wanna be very honest, you know most of the feedback again linking it with the analysis, I don’t know to which extent when Sierra Leone brings that input, […] all the different countries, right? They bring their inputs. Who does the analysis, how is the analysis done […]?” (I22). Structural devoicing can be observed in the case of the (very centrally oriented) ITG of core systems. Since development and decision-making power here lies almost exclusively in the headquarters or central IT realm, matters from the field are not necessarily considered. However, the NGOs themselves are conscious of this fact: “There is like a very low diversity within our headquarters. […] I’m talking about civil society organisations and private organisations and consultancy companies in general. I think that European and US-based corporations, NGOs, etc. lack hugely about diversity” (I16). In some organisations, the lack of diversity in decision boards is also actively countered, e.g., through a dedicated digital strategy written with and by the subaltern: “For example, in Latin America, we created a regional digital agenda with country offices” (I13). Along these lines, dedicated efforts create specific solutions that allow the subaltern to raise their voice, like a platform for youths to promote good manners regarding technology (I10). There also seem to be efforts underway to facilitate more user-driven development, like whole departments focusing on ICT4D projects (I16). Thus, changes in international NGOs’ power structures are underway to empower the subaltern. This shift appears to be driven by external global developments: “Because there are some markets, for example, in the case of Latin America, you have Mexico, you have Brazil, Chile, there, they are good economies […] these countries they were invited to raise more funds […]. And obviously, it was connected to the decreasing of international funds coming principally from Europe” (I10). Such shifts in financing offer an opportunity to democratise structures and mandate a re-definition of the subaltern. If funds are raised locally and thus the degree to which the subaltern can fund their own IT systems increases, they can make their opinions heard. However, as these effects remain marginal, other structures are needed to ensure that the subaltern gets a say in the selection and implementation of IT solutions. Otherwise, the risk remains high that IT solutions do not fit local needs.

5. Discussion

5.1. Summary of findings

The IT goals of the international NGOs we observed in our interviews cannot completely explain the central governance structures we found – a central mystery that we identified in our empirical data. This mystery manifests in top-down oriented relational mechanisms, more formal and less inclusive processes focusing on headquarters and central organisational units, and an overall higher degree of centralisation in ITG (see Table 5; more details on this analysis are given in Appendix D).

Table 5. Observed ITG structures.

Element	Observed structures
Relational Mechanisms	While locally operating NGOs foster the participation of all stakeholders and appreciate their involvement, international NGOs use a more top-down approach, focusing on implementation mechanisms like organisational change management for decisions taken in the headquarters.
Formal Processes	Local NGOs run more participative processes, including field and operational staff such as regional managers in decision-making and requirements management. These are organised in network structures. International NGOs operate a more centralised relationship management, located mostly within the headquarters, only taking up local needs from an outside perspective.
Structural Arrangements	Local NGOs operate more decentral or hybrid ITG structures. Decision-making is divided between different authorities. In international NGOs, core IT and strategy are very centralised. Some have hybrid structures but only focus on project and programme levels.

In addition to the security and economically driven standardisation goals that could explain centralised governance structures, we also found innovation and flexibility goals that would rather call for decentralised structures. Moreover, these goals, by and large, do not differ between local and international NGOs. As such, other factors must have a decisive influence on the observed ITG structures, i.e., the internal and environmental context of international NGOs.

As outlined in the previous chapter, this context appears to be shaped by postcolonial structures for large international NGOs. As these NGOs operate most frequently in contexts affected by their colonial past, e.g., Africa, Asia, and South America, the NGOs’ context is, to some extent, also shaped by postcolonial structures. This organisational context also affects the NGOs’ structures and then shapes the ITG structures.

Even though nearly all former colonies became independent again in the last century, “colonization appears as a foundational shaper of these [international] orders, to a degree and with effects still under-appraised” (Epstein, 2014, p. 294). We saw this international order in the flow of resources, e.g., donor funds, from the global north to the global south in the NGOs. The programmes they operate usually aim to improve the lives of people in the countries in the global south and are financed by donors in the global north. NGOs often have similar developmental goals to ICT4D projects; thus, their actions are similarly “initiated by donors and researchers in the global North (developed nations in the Northern hemisphere) on behalf of communities in the South (developing nations in the Southern hemisphere) [which] underscores the existence of these power imbalances” (Chipidza & Leidner, 2017, p. 1). Such insights resonate strongly with our findings on the influence of IT leaders in the global north and donors (from the global north) on IT selection and focus, e.g., on data and analytics, to create transparency. As such, power appears to reside in the global north.

Going further, we also find distinct aspects of PCT as captured in the core concepts highlighted above (see Table 3). We find that the mainstream indeed makes the most IT-related decisions for the subaltern. Following a paternalistic ideology, this often happens with the best interest of the subaltern at heart but with a limited understanding of their needs. Hereby, the headquarters focus on countries with more issues regarding IT skills and resources and generalises them to all the subaltern. The subaltern, in turn, react with mimicry and ambivalence, but we also see the emergence of a hybrid culture or third space. Even though there are positive examples, only a few organisations give the subaltern an active voice; instead, they appear to be largely devoiced. As the organisational context largely shapes the IT decision structures, they are not in themselves postcolonial.

The structural design of ITG in itself is context-sensitive and is, therefore, often based on prevailing organisational structures and grown processes. For example, IS research shows that “centralized IT governance is associated with centralized organisational structures” (Y. Xue et al., 2008). As laid out before, the prevailing structures for most NGOs centralise decision-making power in the global north. These structures lead to a corresponding ITG design, which manifests itself in centralised IT-related decision-making powers and processes that primarily occur in instances of the global north. This design can be advantageous regarding standardisation, security, and cost-efficiency goals but risks devoicing users and stifling innovations. While the IT goals pursued seemed reasonable, their motivation sometimes remained unclear. Our impression was that the individuals themselves were unaware that they pursued some of their goals based on assumptions resulting from postcolonial structures, like a paternalistic impulse of the mainstream to decide for the subaltern. For example, in pursuing a high security and standardisation goal to make the most out of the given finances in the field, self-determination is suffocated in the global south by lapsing to even more centralisation. Thus, we have to conclude that ITG in itself is not postcolonial but that a centralised ITG can reinforce detrimental power imbalances stemming from a postcolonial context.

5.2. Theoretical contributions

With these insights, our research offers five main contributions to theory. Firstly, it extends our understanding of ITG through PCT. Secondly, taking a critical perspective, we contribute updated discussions of factors that influence ITG structures. Thirdly, our insights specify the benefits and limitations of more modern and hybrid forms of ITG. Fourthly, we contribute to PCT theory by broadening the scope of its application in IS research while, fifthly, focusing on the application of PCT in general.

As a first contribution, this research extends the understanding of ITG with the help of PCT by identifying additional influences on ITG structures along the categories of organisational IT goals, internal context and external environment (see Table 6 for a brief summary of this contribution). While we could not identify how the IT goals were influenced by post-colonial structures, we observed indications that both the internal context and external environment of international NGOs were, in fact, partially shaped by prior colonial structures. Regarding the internal context of an organisation, our data highlights that the post-colonial reality in which it is situated might influence internal structures and processes. Organisations with headquarters in the global north and active in the global south are prone to be influenced by post-colonial thinking, which might affect their general organisational structure and ITG structures. Regarding the external environment, external pressures (in society and from shareholders) might force organisations to become more socially sustainable (Strand et al., 2015). One specific example is the Charity So White movement that addresses especially international NGOs (Charity So White, 2022). More generally, the discussions around corporate social responsibility and ethical investments encompass ecological and social sustainability, also reflected in the United Nations social development goal 16 (“Promote peaceful and inclusive societies”). From an ethical perspective, active engagement with, analysis of, and treatment of any post-colonial structures, even in the realm of ITG, is a concrete component of this social sustainability and is incorporated in corporate social responsibility programmes (Matten & Moon, 2004). As society demands more sustainability from global organisations, the ethical consideration of post-colonial structures must also be included in the realm of ITG. Thus, regarding the external environment, the external (competitive) pressure of sustainability requirements somehow affects the organisational ITG.

Table 6. Updated influences on ITG structures.

Cate-gory	Original Description	Empirical Observation	Critical perspective	Updated Description
Organisational IT goals	(1) flexibility and innovation, (2) security, (3) standardisation, (4) efficiency or (5) IT synergy	NGOs pursue these organisational IT goals. The goals, in fact, influence the design of ITG mechanisms, processes, and structures. As an example, organisations that focus more on IT security have a more centralized ITG (I26).	Organisations should also pursue the goal of empowerment to avoid perpetuating colonial structures and to achieve higher degrees of social sustainability.	(1) flexibility and innovation, (2) security, (3) standardisation, (4) efficiency, (5) IT synergy or (6) empowerment
Internal context	(1) The overall organisational centralisation, reflected by decision-making rights, authorities and patterns of communication and collaboration, as well as the general organisational structure (2) The power of the IT function may influence the actual design of ITG, as a great power can lead to the IT department enforcing its own goals	Aspects of the internal context are influenced by prior colonial structures. As an example, organisations with headquarters situated in the global north also have central IT departments situated in the global north, where “the white person [coming] from Europe […] now explain[s] things” (I1).	Organisations need to acknowledge the way prior colonial structures might influence their internal context.	(1) The overall organisational centralisation, reflected by decision-making rights, authorities and patterns of communication and collaboration, as well as the general organisational structure. This general organisational structure and the corresponding degree of organisational centralisation might be influenced by post-colonialism. (2) The power of the IT function may influence the actual design of ITG, as a great power can lead to the IT department enforcing its own goals
External environment	(1) External (competitive) pressure forcing organisations to engage in increasing efficiency or standardisation (2) Mimetic pressure forcing organisations to adopt specific technologies/structures (3) Availability of external resources for actual users within the organisation, which would strengthen their power	NGOs are subject to external pressures to overcome post-colonial thinking and become more socially sustainable (e.g., Charity So White, (2022)		(1) External (competitive) pressure forcing organisations to engage in increasing efficiency or standardisation or to become more socially sustainable (2) Mimetic pressure forcing organisations to adopt specific technologies (3) Availability of external resources for actual users within the organisation, which would strengthen their power

Secondly, a critical perspective on our findings regarding the effects of postcolonial structures suggests that an additional goal to counterbalance these effects is necessary: The formation of ITG structures cannot only be influenced by the choice of security and innovation goals – we need to evolve mechanisms and ways of thinking that incorporate ethical aspirations into our decision-making. One potential answer is to expand the goal horizon by incorporating goals that actively dismantle postcolonial structures and balance power distributions between organisational stakeholders. From an ethical standpoint, this expenditure should empower structurally disadvantaged organisational units and field workers in the global south. Thus, we need to extend organisational IT goals with an empowerment goal. This extension is in line with the call of Sarker et al. (2019), who argue that “many IS researchers have forgotten or ignored the premise that technologies need to benefit humankind overall […], not just their economic condition” (p. 705). Building on both our empirical observations and our critical perspectives, we thus contribute to and update the influences on ITG structures (see Table 6).

Third, our study (and especially the analysed prior literature on ITG) indicates that modern approaches to ITG could help decolonise IT organisations. Agile management approaches towards IT projects build upon values such as democratisation, bottom-up decision-making, and inclusion of interdisciplinary voices (Beck et al., 2001). As such, these approaches towards ITG in smaller project settings should inherently be less prone to post-colonial influences such as paternalistic behaviour. Acknowledging the security, efficiency, and standardisation pressures organisations face, bi-modal IT has been proposed as one hybrid approach of ITG that integrates traditional and agile IT management (projects). More specifically, bi-modal IT conceptualises two modes: A centrally managed core IT remains in one mode, focusing on stability and security. In the other mode, the focus is on flexibility and innovation in a more decentralised and agile manner (Toutaoui et al., 2022). This decentralised and agile mode of bi-modal ITG could alleviate post-colonial influences. As such, bi-modal IT might be one way to design a more emancipatory ITG. Moreover, recent developments in the area of citizen developers or user-driven IT also reduce the power of a mainstream who designs IT for the de-voiced subaltern. Instead, the prior subaltern are empowered to create their own artefacts. This change, again, could form a more emancipatory ITG with a lower degree of power imbalance between the former mainstream and subaltern.

Fourth, this research broadens the scope of the application of PCT in IS research while simultaneously focusing the application of PCT in general. As outlined above, post-colonial structures and thinking can be understood as part of an organisation’s internal context and external environment. Context is a key component in several organisational-level studies in IS research (e.g., Baiyere et al., 2020; Ein-Dor & Segev, 1982; Tumbas et al., 2018). Our study exemplified that, especially if this context is multi-national, influences of post-colonial structures and thinking can have an impact. Thus, PCT research in our field should not only focus on single artefacts or projects (Lin et al., 2015; Romano et al., 2023) but also include IT decision-making structures. Building on this, we also contribute to critical theory studies using a PCT perspective in IS. With our study, we build on the great work of the IS scholars already focusing on critical theories (Myers & Klein, 2011) and especially those working with a postcolonial theory perspective – be it in offshoring (King & Torkzadeh, 2008) or ICT4D project contexts (Lin et al., 2015). We extend this perspective to focus on permanent IT structures within international organisations. Our findings that postcolonialism affects ITG through the organisational context highlight similar findings to those discovered in ICT4D projects. However, the internal organisational structures we examined are more permanent than project-based work with a specific time frame. At the same time, we see a third space emerging in NGOs, a development surely empowered by closer integration within organisations compared to the relationship of two separate entities in an offshoring context.

Fifth, we also contribute to the general (non-IS) stream of PCT research. While PCT research in general traditionally focused on societies and organisations at large (Kagee, 2019; Kerner, 2018; Loomba, 2015), there is ample room for further studies on specific areas within organisations, taking technological aspects into account. Apparently, post-colonial contexts do not only influence the thinking and behaviour of individuals and societies. They also become inscribed into technological artefacts and organisational processes and routines. While research on the social aspects is necessary and fruitful, PCT research can also be valuable in socio-technical settings. Our study indicates that even “mundane” aspects such as ITG can perpetuate colonial thinking and require further decolonisation and emancipation efforts.

It is important to note that while we took the social sector as an example because NGOs are already trying to decolonise their programmatic work, our findings also apply to the private sector. We believe organisations there will also arrive at the same challenges NGOs currently face. Prior research has already shown that postcolonial effects might be at play, too (Alcaraz et al., 2012). Especially when large multinationals have their headquarters in the global north, power structures might be similar to those observed in this study. Oftentimes, central IT departments reside in the headquarters and make certain decisions for local business units and subsidiaries. Again, de-colonisation and a corresponding emancipatory ITG would be needed.

5.3. Contributing a clear research need

Building upon these contributions, we still need to acknowledge that modern ITG approaches do not solve the fundamental problem: How can we balance the justified interests regarding stability, scale, security and the contrasting justified interests regarding emancipation and empowerment, especially regarding core IT decisions? As a discipline, we need to challenge the comfortable assumption that IT solutions are neutral tools and that our IT-related decisions do not warrant critical examination. The great work on gender bias in voice recognition (Bajorek, 2019) and race bias in image recognition (Cavazos et al., 2021) should have already alerted us, as IS researchers and practitioners, to the ethical dimension of our discipline. In an increasingly digital world (Rennstich, 2008), honest intentions are not sufficient anymore. We need to go further. Creating emancipatory IT governance should also be possible if we can design emancipatory assistants (Kane et al., 2021). Our research highlights how inherently correct IT decisions – centralising IT governance to ensure the security of marginalised communities NGOs work with – can have adverse effects if they systematically devoice organisation members in the global south. We need to develop prescriptive knowledge on designing ITG to balance the justified interests regarding stability, scale, security, and profit with the justified interests regarding emancipation and empowerment. This prescriptive guidance can build upon insights from adjacent fields. On an organisational level, the simultaneous pursuit of seemingly conflicting goals of profit and (social and environmental) sustainability has been discussed for over two decades (e.g., Elkington, 1998; Norman & MacDonald, 2004). On a societal level, research focusing on the United Nations Social Development Goal 8 (“Promote sustained, inclusive and sustainable economic growth”) has produced insights into balancing profit, people, and the planet, too (e.g., Biermann et al., 2017; Mio et al., 2020; Smith et al., 2022). However, developing such prescriptive knowledge on designing ITG should not only build upon results from adjacent fields but also refrain from being (solely) based on values and ethics from the global north. Instead, it should also rely on the philosophies of the global south to minimise potential biases that come from a mainstream perspective on post-colonialism (Masiero, 2021, 2023).

5.4. Contributing to social sector research and practice

On top of these main contributions and the call for value-conscious research on prescriptive guidance, we contribute to social sector research and practice. Firstly, we contribute to the social sector ITG research. With this study, we contribute to closing the research gap on optimising ITG in the social sector. The challenges international NGOs face coordinating across borders are not unique, but working with vulnerable communities and the non-profit orientation surely makes finding the right balance of IT goals more challenging. The social sector represents a unique organisational context, which is yet underexplored in ITG research, in contrast to, for example, the public sector (Magnusson, Khisro, et al., 2020; Tonelli et al., 2017). Since IT solutions will likely substantially shape the future (Rennstich, 2008), NGO research should also examine how IT decisions are made. So far, NGO studies appear to focus more on the effects of single solutions like social media (Sheombar et al., 2015) or postcolonial aspects in NGO work (Fowler & Mati, 2019) as separate concepts.

Our work also has important practical contributions. We hope that practitioners in NGOs and international organisations already critically reflect on their IT decisions and structures. Again, we can only highlight that centralised ITG can satisfy important IT goals. But taking into account that NGO leaders are already engaged in efforts to decolonise aid (Konyndyk & Worden, 2019; Worden & Saez, 2021) such as moving their headquarters to the global south (Radojev, 2017), we are sure that they will also include IT decision-making in their more general re-shaping efforts. However, our work hopefully also paves the way to enabling NGO practitioners to access more general insights from ITG research. Large international NGOs do not focus on profit maximisation. Instead, they are incentivised to act as responsible stewards of donor money and to pursue efficiency goals, e.g., to keep administrative costs low (Burkart et al., 2018). Thus, aspects of ITG literature can alienate NGO practitioners, and insights need to be translated.

6. Limitations

Although we designed our research effort with the utmost care, a few limitations remain. The most relevant issue is that we chose not to ask directly about PCT aspects. We felt that the guidance to complement the line of inquiry with several “friendly” and “non-threatening” questions (Yin, 2009) would not suffice here. We knew from prior discussions and our work in the sector that merely suggesting postcolonial tendencies could irritate our interview partners to the extent that they would not be willing to talk to us anymore. Instead, we discussed the more general topic of decision-making, consultation, and participation with our interviewees, within which the postcolonial influences of the context the operations operated in typically manifested. The second limitation is our selection of interview partners and the NGOs they reflect. The social sector is diverse, with many different organisational causes and organisation types. We covered a broad set of causes (e.g., humanitarian, ecological, or developmental). Especially those international NGOs working in humanitarian aid and refugee relief are likely to be aware of the postcolonial implications. As such, selecting exclusively organisations with an e.g., ecological cause might lead to different findings. Last, while we benefitted from the extensive industry experience of one of the authors, her long-term involvement in the sector might also have introduced potential biases. For most of her career, she was based in the Global North and thus might also have been accustomed to mainstream NGO thinking. We tried to reduce the effects of this potential bias in data collection through the involvement of other authors in conducting the interviews and in data analysis through the rigorous application of the PCT lens by multiple authors.

7. Future research

Next to a general call for more studies on PCT in IS and especially on more prescriptive studies on the de-colonisation of ITG, we would like to highlight three areas for future research: More critical reflection of IT decisions using PCT, more work on IT use in international organisations operating in a postcolonial context, and finally, more research on ITG in NGO contexts in general. Firstly, as IS researchers, we need more insight into how IT decisions might perpetuate postcolonial power imbalances. The level of IT decisions might vary from the selection of specific software to the formulation of general IT strategies. As our study of ITG highlights, this might imply perfectly justifiable decisions, which just become problematic due to the context in which they are made. Otherwise, we risk becoming unconscious agents similar to the painters and writers who helped to enforce colonial ideas about the Orient (Said, 1978). It would mean extending the focus of PCT studies in IS. We must question how and where IT is used and who makes these decisions – not just in ICT4D projects and offshoring contexts. Secondly, as we highlight above, what we found true for NGOs is likely to apply to other private sector organisations as well. Addressing problematic ITG structures in these organisations might be even more impactful if we aim for the empowerment of the global south. More evidence from an IS perspective is needed to understand the intricacies of ITG in these contexts. Finally, during the preparation of this research effort, we could only identify a few publications on the ITG of NGOs. As highlighted above, these organisations face unique challenges and cannot simply copy and paste the insights generated for the private or public sector. We hope to see more research focused on this area, especially due to the potential benefits IT can offer to these organisations and their work.

8. Conclusion

From a critical perspective, we examined international NGOs’ IT governance with a series of 27 interviews – a topic always presumed to be neutral. In the interviews, we found that centralised and security-focused IT governance structures were the most widespread. Especially in the light of the vulnerable communities many NGOs work with, such a focus on data security has merits. However, we also found that these decision structures systematically devoice the subaltern and allow paternalistic tendencies to persist. More interestingly, we also saw mimicry patterns, ambivalence, and hybridity emerge. Our findings suggest that we must acknowledge that IT goals and the organisational context shape ITG. The crucial aspect is that postcolonialism also shapes these goals and (at least in international organisations) the internal and external organisational context. We hope our work is only the start of a larger discussion in IS around this topic, as artefacts and the surrounding social systems within IT enable or restrict empowerment and emancipation.

Acknowledgements

An earlier version of this paper, titled “Is Centralization Good IT Governance or Post-colonial: Insights from NGOs,” was presented at ICIS 2022. We would like to thank the editors and reviewers both of the European Journal and the International Conference for their support in developing the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/0960085X.2024.2325358.

Notes

1. In this article we use terms such as global north, global south, and western. While these terms are often used, they have their ownshortcomings. Following Pansera (2018) and Jimenez et al. (2022), we use the term global south to subsume countries that used to becalled “developing” or “less developed”. Next to this impliedeconomic classification, we would like to emphasize the correspondinggeo-political order manifested in “an arrangement of powerrelationships that dominate the relations between the former dominantcolonial empires and the dominated colonies” (Pansera, 2018). Global north or western thus refers to the former dominant colonialempires. We acknowledge that these terms ignore particularities ofsingle countries in the respective areas.

2. As an interims manager this interviewee has no fixed organisation. He has served organisations with 1.000–40.000 employees in the last three years.