Introducing matrix management within a children’s services setting – personal reflections

Background

Matrix management was implemented within a Local Authority (LA) Educational Psychology Service (EPS) as a response to council restructuring and budget disaggregation of a centralized service into four geographical areas. The respective EPS managers and the four new multi-professional Area Teams were then required to create a practical outworking of this imposed structure.

This article reflects on the process that managers undertook to find a working solution to that implementation. It considers research methodologies that provide structure and insight to these organizational and organic processes of management change.

This opportunity and incentive to step back from ‘full immersion’ and into the role of ‘process analyst’ arose from participation in an East Midlands LA management initiative of ALICSE. This provided an opportunity for third-tier managers to attend management training, facilitated and led by local higher education institutions working with senior officers from these East Midlands authorities. The ALICSE course brought a considerable and repeated emphasis on the ethical dimensions of management. Decisions are value-laden; they impact on people’s lives; rarely is the evidence-base unambiguous or not open to challenge. Decisions are culturally specific; the narrative within which they are encompassed is rarely examined. ALISCE brought an opportunity to step back to seek the lens through which we view, which, whilst difficult, has been a helpful and revealing process.

Initially, the imposed matrix management arrangement created a tense environment: the EPS management contended that this structural change was an unhelpful step, for the following reasons:

the disaggregation of a specialist service that functions most effectively on a county-wide basis adds complexity to an already complex system;

It is against this background that the Children Services instruction was issued to develop an agreed understanding of matrix management and to deliver this system, such that it would support ‘area-based working’.

Many scholars have known for at least 30 years that successful systems-change is facilitated by working with forces within the organization which are supportive of change and improvement − ‘the healthy parts of the system’ (Georgiades and Phillimore, 1975). Whilst many organizations that ventured on the change journey have not met this criterion for systems change, Georgiades and Phillimore (1975) identify other prerequisites for successful change, many of which some organizations have met, namely that:

key personnel were involved;

Literature review

This project examined the development of function and delivery of the lead writer’s role within the local authority which was central to his work and where he was central to the project. This is likely to indicate an action research model over, for example, ethnographic/participant observation (Hammersley and Atkinson, 1983[2004]). They note that ‘all social research takes the form of participant observation…’, but also that ‘there must always remain some part held back, some social and intellectual ‘distance’, as it is in the ‘space’ created by this distance that the analytical work of the ethnographer is achieved (Hammersley and Atkinson, 1983[2004]). For the lead author, given the centrality to and investment in this process, an ethnographic approach is not the appropriate methodology.

An action research approach reflects the untidy circumstance of working with the ‘soft systems’ (Checkland, 1981) of multiple managers seeking personal clarities that make sense from their own perspective. Soft Systems Methodology (SSM) is ‘intrinsically a collaborative approach’ (Checkland and Scholes, 1990), and as such was a more promising reflection of the true context.

Unlike ‘hard systems’ analysis, which tends to address policy and procedures directly, and employee culture less so, SSM gives equal weight to both aspects, recognizing the need to begin to work at the level of unwritten rules, stories, myths and conflicts. The earlier points reflect some of these stories and beliefs held within the psychology service about the need for and purpose of matrix management.

Informal exploration with those staff involved within this process led to identifying conflicting management paradigms. Models from social care influenced the dominant management style within the local ‘area management’. These carried their own ‘rules’: ‘real managers’ challenge, intervene and direct at all levels of functioning: with casework, with professional development and with decisions at all levels. Locally, social care managers tend to be ‘full-time’ managers, holding no direct casework.

Within educational psychology, there is a culture of the front-line practitioner being, and expected to be, an ‘autonomous professional’, providing impartial and independent advice and intervention. This culture creates an expectation that the EP is able to challenge local authority decisions, thus influencing each other’s practices and management direction. As a result, the EPS achieves service development through processes of critical questioning and Socratic conversation, with an explicit focus on continual skill development of those staff within the EPS. In this context, ‘real managers’ facilitate, bringing some truth to the joke about psychologists and changing the light bulb – it takes only one psychologist to change it, but only if the light bulb wants to be changed.

Simplistically, we might characterize the former professional group’s default management style as ‘transactional’ and the latter as ‘discretionary’ (Kakabadse and Kakabadse, 2004). No value judgements are intended through these characterizations. It may well be that the safeguarding elements of the social care profession, where individuals make literal life and death judgements, drives this greater need for minimizing risk and the increased need for control, whilst the educational psychology profession’s primarily consultative practice lends itself more readily to a flatter, more collegiate organization, led in ways that we can describe as discretionary.

Learning from and applying the SSM approach leads to parallel activities, one following a stream of cultural analysis, and the other a logic-based stream of analysis. The issues highlighted above indicate an aspect of the cultural analysis. As the lead manager of this process, without recognizing these, perhaps subliminal, influences, the lead writer would otherwise have found it difficult to identify the source of tensions and possible blockages to the conversations. The logic-based stream of analysis was visible, representing the overt focus of discussion, such as a focus on describing and redistributing existing management processes.

SSM application within the debate about matrix management results in an expectation that we have identified examples of change that an EPS can implement and that these also represent an accommodation between different interests. These were exemplified in the present study through an acceptance that ‘both/and’ provided a more positive environment, rather than the opening expectation of an ‘either/or’ approach: not every discrete activity identified needed to be allocated to either one manager or the other, but some could legitimately be carried by both. This resulted in a significant easing of the three-way dialogue or rather ‘critical polylogue’ – the ability to conduct multiple conversations simultaneously (Kakabadse and Kakabadse, 2004) through the construction of a shared meaning as to one aspect of the nature of matrix management.

Subsequent development of the SSM research model moved from a two-stream model to an identification of four activity streams. In this local context, the first and fourth elements of SSM (‘finding out about a problem situation … [and] … taking action … to bring about improvement’ [Checkland, 1999]) were subsumed into the involved managers’ general actions, in contrast to being explicit, additional or different actions such as those that an external consultant would require.

The delivery of this project employed an iterative process of ‘cycle and review’, primarily through meetings and dialogue (spoken and written), with the main participants being managers and union representatives. This process of collaborative action inquiry develops results where ‘change and group learning occur through cycles of action and reflection … with [an] emphasis on constructing shared meaning’ (Kakabadse et al., 2007).

Another early development in our mutual constructions of shared meaning which eased this process towards delivery was the agreement for there to be ‘matrix management’ − there needed to be more than one manager. The initial, donated arrangement identified one ‘area manager’ and ‘one central lead’. This phrasing either represented something other than matrix management, or if not the intended outcome, reflected insensitivity to language. The titular change to describing both roles as ‘manager’ significantly facilitated the conversation about the detailed implementation.

Activity Theory (Edwards et al., 2009; Leadbetter, 2008) provides another model of structuring our understanding and examination of change and development within systems. This model has built primarily on Lev Vygotsky’s (1962) learning theories, among others, but identifies six points of analysis, usually depicted graphically in triangular formation and linked by lines of interaction. These are the subject, object and outcome, where the object is acted upon by the subject through mediating tools, such as language of instruction, in order to achieve the outcome (learning, change etc.). These are influenced by the socio-cultural-historical context, structured or analysed through the concepts of rules, community and division of labour. Leadbetter (2008: 215) describes it as providing ‘an interactionist, socially and culturally embedded model, through which activities can be viewed, analysed and worked upon. It is practical and action oriented’.

Of particular relevance to the current study is the emphasis within this approach of the importance of the ‘community of multiple viewpoints with differing interest and tradition’, the history of the development of the systems, and that contradictions within related systems are a central force for change, as ‘participants may question established patterns of working’ (Leadbetter, 2008). This study also applies original Activity Theory (Leontev, 1947/1981) to work environments, underpinning an approach known as ‘Developmental Work Research’ (DWR) (Engeström, 1999). DWR includes a four-step approach within its methodology:

observation of contemporary behaviour;

Whilst the authors did not apply this model in its full sophistication to the present study, nevertheless, these elements influenced both the programme work delivery and the management process. Data gathering included noting:

management behaviours already enacted by key players in the matrix − for example, notification of sickness absence and approval of travel forms;

The authors also considered the Appreciative Inquiry model to help inform the project’s delivery (Cooperrider et al., 2003). Appreciative Inquiry (AI) seeks out the best of existing practice to support the collective visioning of a desired future. It assumes that every system or organization has something that works and is valued, and that a key to implementing change is to identify this central core in such a way that it can inspire action for change. It is a solution-focused approach, based in social constructionism, construing that ‘knowing takes place through interaction with and within a social system … [where] dialogue … is necessary to determine the nature of things’ (Cooperrider et al., 2003: 13, 14), and where the language used creates the perceived reality. It deliberately turns away from a problem-oriented, deficit model.

It outlines a four-step cycle: ‘discovery’ (of the best of current practice); ‘dream’ (of what might be); ‘design’ (co-construction of future functioning); and ‘destiny’ (finding innovative ways to move towards the ideal). Through a process of cycle and review, using a series of key questions, the organization identifies those aspects of its current practice that are vitalizing and enhancing to the work process, and with a relentless future-oriented practice supports the development of both a vision and the means to deliver it.

AI ensures that there is opportunity to answer the questions: ‘what works?’, ‘what should we keep?’ and ‘what is valued by client and professional?’. It provides the structure and process to secure these answers and then ensures that the opportunity and security is provided to identify alternative, positive futures and the path towards them. When using AI, the chance of promoting investment in and ownership of change by the work force is high, through its emphasis on the co-creation of possible futures.

However, this change process that introduced matrix management was not able to use Appreciative Inquiry. The edict to work within a matrix system preceded the process of identifying the practicalities of this by many months: although AI is a powerful tool, it is an approach that many undermine through presenting the outcome prior to the process, rather than identifying it as a result of implementing the inquiry.

Reflections on project delivery

For successful delivery of the project, it was essential that this should be completely embedded within the key players’ ordinary activities. As is likely with all Children’s Services, staff are generally working well above the nominal ‘contracted hours’, so there is rarely capacity for additional activity.

As such, the project was delivered through a relatively small number of meetings between three staff: two tier-three managers and a staff union representative. Prior to each meeting, the managers circulated a document detailing the previous proposals. For the first meeting, the principal psychologist circulated a preliminary document to stimulate discussions.

For action research purposes, this study might have benefited from support from a fourth person, able to engage in information gathering from each participant. Instead, participants were representatives of different constituencies, who sought information and consultation with their colleagues. As such, the observations noted within this article regarding the various agendas and interests, clashing or otherwise, have been those gathered by the lead author as one of the three key players in the process and as the author of the matrix management document ultimately agreed by area managers, senior educational psychologists and the union representative. The lead author has gathered through careful observation, careful listening and in situ questioning observations that underpin this study. With this, he has applied his skills and experience as an applied psychologist of over 25 years.

Nevertheless, this has truly been an iterative process of collaborative inquiry, where the lead author sought and achieved shared meanings from positions previously contentious and incompatible. The lead author delivered this process over a 5 month period. This time-scale facilitated opportunity for consultation with colleagues over its development, such that the participants were able to join the meetings confident they had a clear idea of their constituents’ views.

During this process, this awareness of their constituents’ viewpoints was a positive contribution to the process. In other contexts, where staff bring the role of ‘representative’ to the negotiating table, the views of their members, standing as ghosts at the shoulders of the representative, could bring a chill to an otherwise warmth and fluency of debate, where representatives may feel unable to move on a point that they had not anticipated prior to the negotiation. In this respect, we may better describe the participants in this discussion as ‘delegates’ from their constituencies, who had been given the authority by their line mangers to act on behalf of these others rather than solely to represent their views. This functioning in the role of ‘delegate’ seemed to add to the discretionary abilities that each brought to the table.

This time-scale towards ratification also enabled colleagues to identify relative priorities through real-time developments. Some elements that individuals thought were important and were ‘non-negotiable’ at the start of the process were subsequently viewed as relatively trivial and open to more debate than they first thought. The dialogue began with focus on managerial tasks, such as how the EPS distributed the small budget for ‘consumable materials’ (for example, re-ordering of assessment papers) across the four area teams, or for staff travel, subsistence or professional development, and identifying which of the matrix managers should authorize these. At the start of the conversation, participants viewed these as important symbols of the locus of power and control within the management structure.

The journey made by these forms represented significant management issues, such as identifying under whose authority one could spend the budget, and recruit, deploy and develop staff. So from an expense form, the trail leads to professional decisions as to what is or is not an appropriate task for an educational psychologist.

The process of dialogue, supported by the Activity Theory framework, helped make these links explicit. Through these meetings, managers were able to tease out wording that would ensure that professional accountability and strategy remained with the professional manager, whilst the area managers’ function would be to identify their area operational priorities.

As the extent of the budget savings that the EPS needed became clearer, there became a more existential focus, such as which services would remain after implementing cuts. This brought a realignment within those in the conversation: developing allegiances in the face of adversity, where participants found common ground by identifying management responsibilities that either manager in the matrix agreement could discharge; strategic and professional developments held by the principal psychologist; and identification of local priorities, coherent within the strategic overview, through the area managers.

A metaphor for the process of identifying the delineation of responsibilities within the matrix that emerged during these deliberations was that it would be more helpful ‘to wait to see where the footprints are, before we make the paths, than to assume we know the routes before the trails are made’. The timeline employed within the process did enable us to ‘put the path where we walk, rather than walk where we put the path’. This process permitted the organic development of the agreement and provided opportunity for appropriate modesty that none knew the ‘correct arrangements’ for this matrix prior to its delivery.

However, it also ran the risk of sustaining past practices and missing the opportunities to create new organizational realities in which to deliver services differently and more effectively. In the end, this was a risk. The matrix arrangements did not lead to radical systemic change. This may have been a lost opportunity. However, there is always another imminent reorganization within the county council, providing further opportunity to consider service delivery systems and apply the learning from this present project.

Sometimes significant underlying tension characterized this process of delivering matrix management. That there are different agendas represented within the meeting is obvious from the roles brought to the table. The union representative, at worst, can bring power without responsibility; the professional manager can bring responsibility without power; and the operational manager can bring power without knowledge and with uncertain responsibility.

Although something of a caricature, these elements remained as an underlying source of friction, particularly during the initial stages of the process. In later stages, individuals sought to realign their role to less dissonant positions, such as: the operational manager seeking clarity over responsibility and greater knowledge; the union representative seeking to take a greater lead in the conversations and with it a demonstration of greater responsibility; and the professional manager seeking to (re-)establish power with the responsibility. Concurrent with this, there seemed to be subliminal influences, such as: the professional manager seeking to avoid passing knowledge to the area manager, because ‘knowledge is power’; the operational manager seeking to take wider responsibilities that will bring greater control; and the union representative presenting as if without an agenda other than that of ‘honest broker’.

These are speculative comments. Some are based on observations during and outside the meetings; some are informed by past experience in other contexts; and some are perhaps based on the personally held views from the second author, who was an ALICSE tutor.

Personal role and identity and previous management experience and expectations are elements that impact on the dynamics of this process. Professional identity or socialization into the profession can be an unspoken backdrop to such conversations (or perhaps ‘critical polylogues’ [Kakabadse and Kakabadse, 2004]). The frameworks of investigation for this project supported consideration of these ‘hidden force fields’.

Recommendations

Organizations should cautiously use and introduce with care matrix management. As a means of bringing coordinated serviced delivery to a complex and sophisticated arena, it is neither a sufficient nor essential mechanism to achieve such coordination.

Matrix management can be an emotive and potentially difficult structure to deliver. It can result in confusion for staff and managers alike, leading to a lack of staff confidence that the managers who have been identified for certain tasks are managers who have the skills to deliver to those tasks. Prior to detailed clarification of roles and functions, situations can arise where those managed make assumptions as to their managers’ roles, and raise the not always unspoken question, ‘but what do you know [about my job]?’. The risk of introducing matrix management swiftly is the loss of relationship between professional strata. Front-line staff may assume that managers require front-line knowledge, whilst managers assume that ‘management’ is a transferable skill for delivery in any context. Whilst there may be some truth in both positions, the introduction of matrix management in a manner measured by careful deliberation, perhaps including a period of ‘shadow management’, is likely to be an opportunity to test both assumptions.

As such, matrix management is at its weakest when it has been imposed without sufficient consultation and deliberation as to whether or not it is the correct management framework for the correct teams or services.

As with any structure or decision imposed upon staff, it risks alienation of the staff whose cooperation is required to deliver high quality services for highly vulnerable people under highly stressful conditions. A Children’s Services environment needs to be able to clearly demonstrate that it is a reflexive, learning organization where the qualities needed to deliver services to clients – empathy, sensitivity, high skill levels, sophistication – are the same qualities visible to the work force from management. Appreciative Inquiry provides a clear methodology that would otherwise support such an introduction.

Where these conditions are apparent, matrix management may be useful in bringing a breadth of management expertise into a complex mix of need and skills. Matrix management has not been a necessary step to bring these qualities together. Research from Edwards et al. (2009) shows that a particular management structure is not one of the essential aspects of successful multi-professional working. What makes a success more likely are the ‘softer’ qualities of working, such as confidence in one’s own identity as a professional, the degrees of freedom to act that one brings to a setting, the ability to work across professional boundaries, where needed, and the ability to form and re-form temporary, ‘virtual teams’ to create ‘knot-working’, where colleagues ‘tie-together’ for a specific client-facing purpose, and then ‘undo’ as individuals identify other tasks.

Matrix management seeks to be one answer to fostering ‘seamless, multi-professional working’. However, debate about a range of models to achieve this outcome can be hard to find within the Children’s Services. Instead, scholars and practitioners may present models as if there are no other options that can achieve a particular desired outcome. Projects such as this and development opportunities such as the ALICSE programme facilitate broader thinking. They facilitate the recognition that if decision-making and organizational structures were to be evidence-based, or at least evidence-informed, then an indication of this might be healthy debate about these decisions and structures.

Postscript

This LA retained matrix management for its EPS for just under 2 years. Following further management restructuring, the LA identified that the EPS’s focus was not suitable to deliver through a disaggregated service across four area teams. Since autumn 2012, the EPS has returned to an authority-wide service structure, where the principal educational psychologist holds the budgetary, operational and strategic management as a single-service manager.