The matrix structure is said to provide the best of both worlds – i.e., knowledge management benefits in terms of functional specialization and efficiency benefits in terms of cross-functional project teams. Is this reasoning justified?
AbstractKnowledge management or the leveraging of organisational know-how in new and innovative ways is at the heart of success in today’s knowledge economy. As such it is a key challenge for the organisations of today to structure themselves in such a way to exploit their collective know-how to the maximum extent whilst simultaneously keeping costs under control through efficient operation of their businesses. This article looks at whether or not the matrix structure can justify the claim that it provides the best of both worlds and as such is suitable to lead organisations into the information era. Specifically we look at the supposed knowledge management benefits in terms of functional specialisation and efficiency benefits related to the use of cross-functional project teams comparing and contrasting to more modern organisational forms. Due to the fundamental flaws present in the matrix which undermine its key benefits and its tendency to fail under stress we recommend organisations take decisive action moving towards forms such as the cellular and hypertext structures which will be able to support their businesses well into the 21st century.
Introduction
The matrix structure is arguably the most widely used organisational form of the present day. On the surface its preservation of traditional vertical functional hierarchies appears to be neatly balanced with horizontal cross-functional project teams facilitating change. By adopting such a structure it promises that an organisation can ‘have their cake and eat it’, supporting both the businesses ongoing viability whilst rapidly responding to changes in the market place, simply by overlaying project management disciplines across the existing structure – all without having to undergo deep, transformational change.
Clearly this is an attractive proposition for any organisation, however is this line of reasoning justified? In assessing whether or not the matrix structure (hereafter referred to as ‘matrix’) truly represents the ‘best of both worlds’ this article aims to answer the following questions:
- What are the fundamental characteristics of the matrix structure?
- How does the matrix deliver both knowledge management and efficiency benefits?
- What other factors affect the overall delivery on the promises above?
- Do any other organisational forms better address these challenges?
Firstly we explore the nature and origins of the matrix in order to give it gestalt necessary to support further discussion. Next we delve into the specifics of how the matrix delivers on the central promises of superior knowledge management and efficiency benefits. Following on from that we consider what other factors and/or weaknesses intrinsic to the matrix can impact the delivery of the aforementioned promises. Subsequently we will look to other organisational forms such as network, cellular and hypertext structures and contrast them against the matrix to gain a wider perspective.
Characteristics of the Matrix Structure
A matrix structure is an organisational form that can be generally described as a compromise between a traditional hierarchical or functional structure and a project structure. That said considerable effort has been expended attempting to describe the exact nature of the matrix structure since its rise to prominence in the 60s and 70s. Much of the difficulty in providing a concrete definition lies in the flexibility that the structure provides and room for interpretation of the theory itself. There is however some agreement in the literature on the hallmarks of the matrix. Ford & Randolph (1992) in their review of matrix literature agree with Davis and Lawrence (1977) that a matrix can be defined as ‘any organisation that employs a multiple command system that includes not only a multiple command structure but also related support mechanisms and an associated organizational culture and behavior pattern.’ (p. 3). From this definition they go on to clarify that the two key characteristics of a matrix structure are:
- A traditional vertical hierarchy that is aligned along functional lines overlayed by a ‘lateral authority, influence or communication’ (Knight 1976, p. 113).
- Dual reporting lines to both a functional line manager and a lateral project, product or business manager.
Following these characteristics there are generally thought to be three distinct variations or biases that can be applied to the matrix (Payne 1995):
- Functional bias. In a functionally biased matrix the old hierarchy asserts its dominance over the project based, lateral lines of communication and authority.
- Project bias. A project biased matrix places greater importance on the lateral lines of communication and authority, largely at the expense of functional line management.
- Balanced. The balanced matrix attempts to evenly distribute the importance and authority of both functional and project lines favouring neither.
Origins of the Matrix Structure
Today, the matrix structure has been in existence for approximately forty years. Its origins can be traced to military and government contracts in the 1960s (Atkinson 2003) wherein government research and development grants would only be considered for organisations that exhibited some form of project management system. Reform led by the United States Government acknowledged that the functional structures of the day weren’t providing the necessary accountability for meeting costs and deadlines leading to budget blowouts and late or non-delivery of projects (Knight 1976). Seeking to introduce project management disciplines organisations of the time, such as NASA, quickly realised the challenge; how to implement project management systems with respect to their existing organisational structure?
Whilst pure project structures had been previously described, they did not sit well with the status quo. Implementing a project structure would require transformational change to the deep structure of the underlying organisational (Fenech & Dovey 2005). Managers within the existing hierarchies perceived that this would cause them to lose influence and degrade their relative importance within the organisation (Payne 1993). Indeed their fears were justified insofar that project structures in their purest form make functional line management irrelevant. As such, organisations began to hastily overlay project management across their existing hierarchies thus obviating the need for complete restructure. Thus the matrix was born.
Simultaneously the larger business community was moving out of the era of standardisation and into the era of customisation. Corporations found they could no longer rely upon their industrial might, cost efficiencies brought about mass production and technology to maintain competitive advantage (Miles et. al. 1997; Dovey & Fenech 2007). Henry Ford’s long standing maxim; ‘Any customer can have a car painted any colour…so long as it is black.’ (1922, p. 71), which characterised the preceding era, no longer applied as organisations sought to leverage their know-how in ever increasing ways to differentiate their products and services (Miles et. al. 1997). Moreover organisations such as Texas Instruments identified the need for project management as a strategy to manage, coordinate and control the increasingly frequent change throughout their businesses (Bernasco et. al. 1999). Many saw the matrix as an attractive way to address the information sharing and processing deficiencies present within their current hierarchies whilst avoiding the alleged inefficiencies of project structures.
Knowledge Management Benefits of the Matrix
Knowledge management or the leveraging of organisational know-how in new and innovative ways is at the heart of success in today’s knowledge economy (Miles et. al. 1997; Dovey & Fenech 2007). As such it is a key challenge for the organisations of today to structure themselves in such a way to exploit their collective know-how to the maximum extent. In its optimal configuration operational and investment intelligence as well as adaptation capabilities should be universally available at the top, middle and lower tiers of the organisation (Miles et. al. 1997). For knowledge to become accessible throughout the organisation the structure must address its capture, storage and subsequent dissemination.
Capture of knowledge both explicit and tacit is well served by the matrix. Ford and Randolph (1992) state that the underlying functional structure of the matrix allows employees to remain grouped by areas of expertise, which in turn promotes greater collaboration between staff in their respective technical disciplines fostering technical excellence. There is some evidence to suggest that providing subject matter experts (SME) with a common home features heavily in creating the necessary motivation for knowledge capture to occur. Payne (1993) observes that maintaining strong technical departments create a sense of pride and commitment whilst preserving a sense of identity and focus that project structures often struggle to provide. On the other hand, matrix structures suffer from imbalance within reward structures, which may negatively impact motivation for knowledge capture within functional areas where a project bias exists (Bishop 1999). Regardless, this ability to capture knowledge must be equally complimented by an ability to store and utilise it.
Storage and dissemination of explicit knowledge is not specific to the matrix structure and can easily be facilitated by technology such as email, document repositories, wikis and forums and made tractable when combined with powerful search engines (Taylor 2005). Where the matrix stands apart from traditional hierarchies is disseminating tacit knowledge, which it achieves via cross-functional project teams. Tacit knowledge, unlike explicit knowledge, does not lend itself to codification but rather relies on face-to-face interaction and shared experience. In a multi-project environment these temporal groupings of specialists picked from across the organisation permit cross-pollination of ideas. Upon completion of the project these specialists return to their ‘sentient home’ taking their new experiences with them rather than remaining trapped in its functional silos (Knight 1976), or project structures which disband immediately following project completion and the knowledge is therefore lost.
Efficiency Benefits of the Matrix
Efficiency through the use of cross-functional teams appears within the literature to be a key defence of why organisations adopt a matrix structure. Espoused efficiencies related to cross-functional teams generally fall into two areas: resource utilisation and duplication of effort. In a single project environment the occurrence of either factor is probably unavoidable, though since approximately 90% of projects by value are undertaken in the multi-project context (Payne 1995) we will address it from this perspective.
In multi-project environment project structures are often criticised for their inefficient use of resources. This commonly manifests itself where a project requires only a portion; say 75%, of a specialist skill set. If the project exists in isolation it is likely that it will have to foot the whole bill for the specialists time or risk losing the individual. The matrix provides a solution to this by retaining the linkage to the specialist’s home team. The functional manager in this situation can look to allocate the latent capacity (the remaining 25%) of a given resource to other projects within the wider project portfolio (Knight 1976). This is not without problems though. Eskerod (1996) describes a situation where project managers may compete against one another for priority within the multi-project environment. Savvy project managers that succeed in capturing the interest or imagination of the specialists in their employ can surreptitiously ‘steal’ more than their allotted time (for example shifting the balance to 60%-40%) which in turn impacts the overall effectiveness and integrity of the model.
Another common criticism of project structures, which the matrix seemingly provides a solution to, is duplication of effort. Often times within the multi-project environment there exist synergies between two or more projects. For example Payne (1995) discusses two separate capital works projects, both dealing with ground-air communication, being carried out on two airports in the Middle East. Between these two projects there were clear similarities from a technical point of view, as well as the industry and the geography. The matrix again aims to address this by exploiting the linkage back to a functional home for specialist resources. By maintaining communication with other specialists in his or her area, the matrix in theory provides a means for solutions to commons problems to be distributed across projects. Whether or not this actually occurs is largely dependent on motivational factors, which we will discuss next.
The Best of Both Worlds?
When viewed in isolation the knowledge management and efficiency benefits offered by the matrix make it an appealing proposition. However it’s the overall effectiveness of the sum of the parts, which ultimately gives legitimacy or illegitimacy to the form. Unfortunately the ability of the matrix to delivery on its stated benefits is subject to many other factors.
The first of these factors is without doubt the characteristic dual reporting lines. This intersection of power and authority is flashpoint for major conflict within matrix organisations as it creates and environment where ‘power struggles are inevitable’ (Davis & Lawrence 1978, p.134). This has led some to comment that the matrix form is ‘inherently unstable’ (Payne 1993, p.240). The constant jockeying for authority between functional line managers and project managers can be quite intense and disruptive at times have unintended consequences on the previously described advantages provided by the structure.
As alluded to earlier, knowledge management, a supposed strength of the matrix is often hampered by partisanship perpetrated by members of cross-functional teams. During a case study on the implementation of a new process control system Alsène (1999) noted that engineers acted principally in the interests of their staff team as opposed to the project. In doing so they actively disregarded key project objectives such as user-friendliness in pursuit of technical excellence. This is hardly surprising given that typically reward structures within hierarchical organisational forms such as the matrix tend to lend themselves to functional objectives as opposed to organisational goals (Payne 1995). Moreover Bishop (1999) finds that project accomplishments are rarely rewarded, if at all, especially in cases where functional manager feels his or her authority is threatened by the project manager. Referring back to the case study above one may conclude that although the organisation in question had obviously been effective in capturing knowledge, its application was not served by the structure making it of questionable value.
Challenges to authority can also interfere with the efficiencies of the matrix. Earlier it was noted that three possible biases proliferate in the matrix: functional, project and balanced. These conflicts commonly arise within the matrix when bias and therefore roles are not adequately defined, a consistent approach agreed across projects and are universally understood by functional and project managers (Knight 1976; Davis & Lawrence 1978). In the experience of Dovey & Fenech (2007) during their case study of a large financial institution there was little consensus or true support for what should have been a project biased matrix by senior management. The outworking of this situation was such that when crisis hit functional managers reasserted their former authority, stripping the project managers and PMO of its power to implement change. Research done by Gobeli and Larson referenced by Payne (1993) found that even though most conflicts could be resolved by shifting the bias one way or another (such as occurred in the case of Dovey & Fenech), shifting the bias toward the functional end of the spectrum had a negative effect of up to 30% on technical performance, a clear indictment against the stated efficiencies of the matrix.
The matrix structure also suffers from a number of other maladies related to the sharing of authority between line and project managers. Lack of accountability for negative outcomes is identified as a significant problem where lines of responsibility are unclear within the matrix (Knight 1976). Even if responsibility can be apportioned fairly Bishop (1999) observes that the ‘communication protocol of a hierarchical organisation accentuates the importance of rank within the organisation and keeps the focus on what position a person has attained rather than what he or she actually does’ (p.7). It can be seen that a direct consequence of this phenomenon is that the greater the seniority of those at fault, the greater the organisations willingness to sweep the matter under the rug (Alsène 1999), therefore accentuating the imbalance of power between project management and the functional hierarchy. Sadly this can also persist to the level of specialists where the transitory nature of assignments to projects can affect levels of commitment and blur the boundaries of responsibility (Payne 1995). Ironically lack of accountability is one of the original issues the matrix was designed to solve.
Alternative Organisational Forms
Since the aforementioned negative forces plague the matrix structure, can it really lay claim to being the best organisational form to deal with rapid change whilst leveraging organisational know-how and executing projects in a resource efficient manner? For the purpose of this discussion we will concentrate on three alternative structures: network, cellular and hypertext.
Network organisational structures according to Miles et. al. (1997) are next on the evolutionary path after matrix structures. Network structures are able to quickly respond to changing market conditions like the matrix but go a step further by focusing on the entire value chain rather that being limited by the confines of the organisation. This expanded focus allows organisations to identify and reprioritise areas of their business where assets and know-how can be utilised most effectively, whilst at the same time outsourcing peripheral business activities to other complementary organisations better suited to the task. While this reduced focus on the core business may help organisations decrease the size of their project portfolio it still critically doesn’t address the central concepts of project management as it applies to internal projects, in fact by introducing vendors it may even complicate things further, a view consistent with the authors experience.
Cellular organisational forms are perhaps the epitome of adaptive organisations; best placed to leverage explicit and tacit knowledge at all levels without succumbing to fierce internecine conflict so characterised by the matrix form (Dovey & Fenech 2007). Cellular forms eschew complexity, size and deep hierarchies in favour of flat, nimble, highly empowered cross-functional ‘project’ teams. Entrepreneurialism is rewarded as well as business-like behaviour (Miles et. al. 1997) sharply contrasting the large monolithic organisations of today; this is not to say that the cells are disparate. Miles et. al. (1997) goes on to state that each cell has an entrepreneurial responsibility to support the objectives of a larger organisation, however its relationship with said organisation is not exclusive. In other words the cell is free to seek business opportunities that align to its goals and objectives wherever they might exist. Should the demand for a particular cell’s expertise disappear, the cell must either repurpose itself or disband. In this way a cellular organisation addresses resourcing inefficiencies prevalent in project organisations in a rather novel, yet pragmatic manner.
Knowledge management is also handled in a different manner to the matrix. Cells are nominally independent from each other and their small size ensures the easy transfer of ideas and information between team members. Fleishman (2005) in her discussion of Cellular Terrorist Organisations (CTOs) such as al-Qaida notes that only limited information transfer is necessary between cooperating cells working toward larger objectives. The type of knowledge is predominantly codifiable and is therefore well suited to indirect, technology-based collaboration such as the web and email. Where face-to-face knowledge transfer needs to occur so-called boundary spanners can assist but are recognised as non-essential. This of course fundamentally contradicts the reasoning behind the ‘sentient homes’ proposed by Knight (1976). In reality cells share a lot in common with a management by projects approach where ‘cross-functional information sharing is paramount’ (Fenech & Dovey 2005, p. 8). Importantly these benefits are easily applied outside of terrorist organisations, on local and global scales, as witnessed by the success of computer graphics firm TCG and computer manufacturer Acer (Miles et. al. 1997).
Hypertext. Though not mentioned by Miles et. al. (1997) or Fleishman (2005) the author believes the so-called hypertext organisational structure proposed by Nonaka and Takeuchi is worthy of note. The hypertext structure is comprised of three separate layers; the first operational, the second project based, and the third knowledge based (Alsène 1999). A structure of this kind supports the continuity and ongoing viability of the business via the operational layer, whilst having the flexibility to employ a fully-fledged management by projects approach at the second layer thus avoiding the conflict inherent in the matrix. Learning is assured through via the orthogonally positioned knowledge layer that aims to capture both project and operational knowledge equally. Consequently the hypertext organisation may provide an important familiar middle ground for large organisations looking to transition to more modern, flexible and adaptive structures but are unable to muster the resolve and/or consensus at the senior management level to transition to a cellular structure.
Conclusion
The matrix has faithfully served corporations for approximately the last forty years. In this time much light has been shed on the exact nature of the matrix, its benefits and its shortcomings in the literature. During this time other organisational forms have sprung up and sought to challenge the fundamental thinking behind organisational structures. In comparing and contrasting the new versus the old we have mainly focused on the knowledge management and efficiency aspects of the organisational forms in question.
In tackling the subject presented we must question the nature of the statement itself because it assumes that ‘both worlds’ (i.e. functional hierarchy and project structures) are strictly necessary for success. This is patently untrue as witnessed by the global and local success of organisations employing cellular structure such as Acer and TCG presented by Miles et. al. (1997). Even if we are to assume that both elements are necessary, and concede that the matrix does provide some basic mechanisms to support knowledge management and efficient use of resources, the weight of evidence directed toward the inherent flaws of the matrix structure should give any management practitioner pause for thought. Fenech and Dovey (2005) go so far as to say that this line of reasoning is ‘fatally flawed…[principally] because the matrix structure exacerbates political intrigue’ (p. 8) potentially nullifying the alleged benefits it is supposed to deliver.
Therefore the author finds that the reasoning that the matrix offers the ‘best of both worlds’ is unjustifiable, especially when considering the superior alternatives put forth by (Miles et. al. 1997; Alsène 1999 & Fleishman 2005). The question remains whether senior management of large vertically structured organisations have what it takes to respond to the challenges of the knowledge economy by enacting large-scale change and in effect sponsoring their own disempowerment. Ultimately it is the author’s belief that the market will decide the outcome. This view is consistent with Fleishman’s (2005) appraisal of the unfortunate success of CTOs to date when pitted against large bureaucratic organisations such as the United States Government – surely a portent of things to come.
References
Alsène, È., 1999, 'Internal Changes and Project Management Structures Within Enterprises', International Journal of Project Management, vol. 17, no. 6, pp. 367-376.
Atkinson, P. 2003, ‘Managing Chaos in a Matrix World’, Management Services, November, pp. 8-11.
Bernasco, W., de Weerd-Nederhof, P. C., Tillema, H. & Boer, H. 1999, ‘Balanced Matrix Structure and New Product Development at Texas Instruments Materials and Controls Division’, R&D
Management, vol. 29, no. 2, pp. 121-131.
Bishop, S. K. 1999, ‘Cross-Functional Project Teams in Functionally Aligned Organisations’, Project Management Journal, vol. 30, no. 3, pp. 6-12.
Davis, S. M. & Lawrence, P. R. 1978, ‘Problems of Matrix Organisations’, Harvard Business Review, May-June, pp. 131-142.
Dovey, K. A. & Fenech, B. J. 2007, ‘The Role of Enterprise Logic in the Failure of Organisations to Learn and Transform’, Management Learning, vol. 38, no. 5, pp. 573-590.
Eskerod, P. 1996, ‘Meaning and action in a multi-project environment’, International Journal of Project Management, vol. 14, no. 2, pp. 61-65.
Fenech, B. J. and Dovey, K. A. 2005, ‘Evaluating Project Management Maturity Models: An Analysis of Business Needs’, In Proceedings of the PMI Global Congress 2005 – Asia Pacific, Project Management Institute.
Fleishman, C. 2005, ‘The Business of Terror: Conceptualizing Terrorist Organizations as Cellular Businesses’, viewed 26 March 2010, <http://www.cdi.org/pdfs/terrorist-business-model.pdf>
Ford, H & Crowther, S. 1922, My Life and Work.
Ford, R. C. & Randolph, W. A. 1992, ‘Cross Functional Structures: A Review and Integration of Matrix Organization and Project Management’, Journal of Management, vol. 18, no. 2, pp. 267-294.
Knight, K. 1976, ‘Matrix Organisation: A Review’, Journal of Management Studies, May, pp. 111-130.
Larson, E. W. & Gobeli, D. H. 1989, ‘Significance of Project Management Structure on Development Success’, IEEE Transactions On Engineering Management, vol. 36, no. 2, pp. 119-125.
Miles, R. E., Snow, C. C., Mathews, J. A., Miles, G. & Coleman, H. J. Jnr 1997, ‘Organising in the Knowledge Age: Anticipating the Cellular Form’, Academy of Management Executive, vol. 11, no. 4, pp. 7-20.
Payne, J. H. 1993, ‘Introducing Formal Project Management into a Traditional Functionally Structured Organisation’, International Journal of Project Management, vol. 11, no. 4, pp. 239-243.
Payne, J. H. 1995, ‘Management of Multiple Simultaneous Projects: A State of the Art Review’, International Journal of Project Management, vol. 13, no. 3, pp. 163-168.
Taylor, B. 2005, ‘Knowledge Management for Software Projects’, Cutter Consortium, viewed 4 April 2010, <http://www.cutter.com/content/project/fulltext/advisor/2005/apm050901.html>
