I have in a previous blog post argued that universities should be structured as a matrix organization to allow for interdisciplinary collaboration. That blog post was written after I had been Head of the Department of Musicology for four years. At the time, I was slightly frustrated about the challenges we faced with “silos” within the university. Many initiatives were halted due to the university’s organizational structure, which is divided into faculties and departments. My thought was that a matrix organization could be a way to solve the problem.
Four years later, I have co-directed the establishment of RITMO as a new research centre at the University of Oslo. Here, I will reflect on why I no longer believe that a matrix is the best organizational structure for a university. That does not mean that I want to keep a hierarchical organization. Instead, I am now thinking about how we can create a web-based university organization.
Challenges with Matrix organizations
My argument for considering a university matrix organization was based on the observation that several top universities already operate in this manner. For example, Cambridge and Oxford, with their organizations into both colleges and departments/faculties and where faculty members are affiliated with both a college and a department. There are also traces of matrix-like organizations in North American universities, where teaching is often structured into colleges/schools while research activities are organized within a traditional departmental and faculty structure. Such structures look fairly similar to corporate matrix structures, such as this one:
An example matrix organization (credit: Msilva83, CC BY-SA 4.0 via Wikimedia Commons).
Unlike the hierarchical, linear structure at UiO, a matrix organisation is established to handle both “horizontal” and “vertical” activities within the organisation. This is particularly relevant for those who don’t really “fit” into the vertical structures, for example, various types of multi- and interdisciplinary research and teaching. These are, by definition, fundamentally based on horizontal structures.
While matrix management may solve some problems, it may also lead to new ones, as Christopher A. Bartlett and Sumantra Ghoshal argues:
In practice, however, the matrix proved all but unmanageable—especially in an international context. Dual reporting led to conflict and confusion; the proliferation of channels created informational logjams as a proliferation of committees and reports bogged down the organization; and overlapping responsibilities produced turf battles and a loss of accountability. Separated by barriers of distance, language, time, and culture, managers found it virtually impossible to clarify the confusion and resolve the conflicts.
This is from a corporate perspective, however, things may look different at a university. After all, universities are full of people who deal with complex things and don’t want to be organized into simple structures. Still, my reason for disregarding a matrix organization is primarily conceptual.
Adding a “horizontal” beam
RITMO is, in many ways, structured as a matrix-like organization. It is based on bringing together researchers from three departments within the university: Psychology, Informatics, and Musicology. My matrix-like dream has almost come true, at least in the sense that everyone working at RITMO is affiliated with both RITMO and a department. RITMO can be viewed as a “horizontal” structure overlaying the “vertical” structure of departments and faculties.
Also within RITMO, we have a matrix-like structure, with four research clusters and a multitude of projects that connect people in different ways. Even RITMO’s thematic areas and empirical areas are organized as a matrix:
This all looks quite logical from the outside, but it is more “messy” on the inside. Fortunately, getting funding for a Norwegian Centre of Excellence opens many doors that would otherwise be closed. We have managed to “hack” UiO’s linear organizational structure to accommodate RITMO’s horizontal placement. While it works, we also see that “hacking” a linear organization is not the same as making a real matrix organization. For example, the economic system, HR processes, IT system, etc., are not rigged for handling a horizontal organizational element like RITMO.
Through the years, we have seen a shift towards focusing more on interdisciplinary research at UiO and elsewhere. “Everybody” talks about interdisciplinarity these days. I always try to remind people that a lot of what is going on is usually multidisciplinary, that is, people working on a common topic from multiple disciplinary perspectives. We also conduct multidisciplinary research at RITMO, but we work towards interdisciplinarity: the integration of various separate disciplines.
The advantage of interdisciplinary research projects (small scale), centers (medium scale), and university-wide initiatives (large scale) is that they aim to cross the borders of departments and faculties. The more this happens, the more we also challenge the rigid linear structure of our organization. Over the last couple of years, I have been involved in several different working groups within UiO that have concluded with the need for organizational changes to support interdisciplinarity. I am pleased to see that many people, although not everyone, agree that this is a necessary change. The challenge, then, is how to make it happen.
Towards a web-like organization
One of the reasons I have given up the matrix as my dream organizational structure is its fixed and rigid structure. True, matrix-like organizations do not necessarily need to be entirely grid-based; however, metaphorically speaking, a matrix is more rigid than I would like.
The various types of interdisciplinary activities I dream of—in both research and education—come in all shapes and sizes. Furthermore, many such initiatives should be dynamic; they should come and go as needed. I actually like 4-year-long projects. That is sufficient time to delve into a specific problem, after which new issues will arise and can lead to new projects. Developing a lab, on the other hand, takes more time and is something that needs to happen in parallel. Rolf Inge Godøy and I started the fourMs Lab back in 2005, and it took at least a decade before things were really up to speed. Running a time-limited centre like RITMO is yet another type of organizational structure. Meanwhile, I am also involved in the long-term planning of educational activities, such as the MCT programme. All of these activities (research projects, lab, centre, master’s programme) have different timelines, needs, and resources. They also involve different numbers of people from various departments.
At the moment, we try to make it work by “hacking” our linear, hierarchical university model as best we can. However, placing a grid-based structure on top would not necessarily solve the problem. My new dream is to have an organization that allows people to connect in a multitude of ways. Just in the same way that a web-based network structure allows for dynamic connections. Is this possible within a university?
Building on network theory
A web-based university structure is radically different from both linear and matrix-like organizational structures. I haven’t given it much thought so far, so this is my first attempt at putting my thoughts into words.
Starting from the perspective of network theory, one could imagine that such a web-like structure would be built around nodes. This could be people, research groups, or even a centre. These nodes should be connected in one or more ways, particularly more ways. Universities are complex organizations, and university people deal with complex things. That is why, in my opinion, creating a complex organization is preferable to a simple one.
In network theory, nodes are connected with edges. These edges have weights and directions. I think these edges are critical for how one could imagine that a web-like organization could work. Imagine that decision power and budgets could be based on the size and directions of nodes and edges. That would create a different dynamic within a university.
Supporting complexity
Is it possible to develop a web-based university organization in practice? In many ways, universities already function like a web. Teaching is usually organized strictly into programmes and larger courses. Fortunately, many exciting developments are on the horizon for micro-courses and related micro-credentials. If implemented wisely, such micro-educational elements can serve as a web-based framework for education.
Research activities are already much more free-floating within the organization. Researchers collaborate in many constellations, and external research funding allows for some of the flexibility that I am looking for. Still, there is too much rigidity when it comes to moving money between departments and faculties. This makes it tricky to solve buy-outs, for example. The reason for these challenges is that we have a rigid system that seeks to maintain simplicity and structure. However, in 2021, it should be possible to come up with something more flexible. What I envision is actually how Kevan Hall summarizes a successful matrix organization in Making the Matrix Work: How Matrix Managers Engage People and Cut Through Complexity:
- Context - ensure that people understand the reasoning behind the matrix
- Cooperation - improve cooperation across the silos, but avoid bureaucracy and having too many people involved
- Control - avoid centralization, build trust, empower people
- Community – focus on the “soft structure” of networks, communities, teams and groups
Summing up
This is my first attempt at writing up some of my thoughts on a web-based university structure. It is not particularly well-thought-out, but can serve as the starting point of a thought process. I am curious to explore the thought process further and consider how it could work in a real-life scenario. If anyone has any suggestions, please let me know.