by TIM CAHILL

A university research centre director asked why this is in a research strategy masterclass we were running. We were looking at a case study of a current research strategy from an Australian university, and the point was made that this could have been taken from any Australian university across the last 10 years and, with some minor variations, it would look the same.  

There are several reasons for this.  The first is that most of these strategies are not actually strategies. In our view – and how we teach strategy in our masterclasses and what we create with our clients – strategy is a solution to a problem (which comes largely through the work of Anderson School of Management’s Richard Rumelt).

This presupposes the existence of both a perceived problem and a proposed solution. Importantly, it also presupposes a set of beliefs about how the world works i.e., “we believe our proposed solution will solve our perceived problem because…”. ‘Proposed’ and ‘perceived’ are important here because they imply a learning-based dynamic that is lacking from research strategy. But, fundamentally, the first reason university research strategies all look the same is because they are not framed as the interplay of problem-belief-solution. Instead, they are often statements about a broad future ambition (goal, vision, mission…the terms, tellingly, are used interchangeably) or worse, summary statements about existing capabilities. Neither constitutes a strategy.  

In the absence of the problem-belief-solution dynamic, universities overemphasise external factors that create homogenisation, rather than identifying their unique problems and opportunities. This can result in a strategic approach that imitates rather than innovates. This includes, for example, looking at their peers and competitors when developing their research strategies. There is a tendency to imitate successful institutions or to adopt strategies that are perceived as “best practice.” This results in homogenisation, as universities feel pressure to conform to perceived norms and standards within the sector. Similarly, universities are mostly competing for the same funding sources, such as government grants. As a result, their research strategies are influenced by the requirements and preferences of funding bodies, leading to similarities in the strategies developed. Additional external drivers such as government policies, and regulations also lead to convergence in research strategies. 

If the problem-belief-solution dynamic were the basis of research strategy, by contrast, there would be a high degree of differentiation. Problems are unique in time and space, and are relative to each university’s unique configuration of existing capabilities and systems, opportunities etc. So even if the external circumstances force a kind of homogenisation of problems, the belief-solution aspects of a strategy will necessarily be different. The current homogenisation of strategy simply points out the absence of this kind of thinking. 

The second reason university research strategies all look alike is due to the lack of a learning-based approach. In almost all examples we have encountered, research strategies are effectively “set and forget” exercises. At best they might be revised annually (though more commonly every few years). And in most instances, they are compared against key targets (though generally these are very generic, a result of the “looking at others” and group think referred to above).

But this is a far cry from what is an effective learning-based strategy; that is: a problem-belief-solution strategy that is in a state of updating the underlying beliefs of the strategy on a regular basis.

Strategy in this respect is the art of making better decisions over time based on incorporating new information. It is an interplay between intentionality, on the one hand, and emergence, on the other (this is explored at length in work by Henry Mintzberg of McGill University). There are myriad ways to phrase this but, essentially, a strategy does not, upon its successful execution, look like it did at its initial conception. It is reconfigured along the way to the outcome by changing external and internal factors (like competition, policy, regulation etc.), as well as by a strategy equivalent of the build, measure, learn loop used in Lean start-up. Incorporating a learning-based dynamic into research strategies is crucial for adapting to changing circumstances and making better decisions over time. The static nature of many university research strategies can hinder their ability to respond to new information and opportunities, ultimately limiting their effectiveness. 

This is particularly important inside universities because they are highly decentralised organisations. Decision-making is highly distributed, especially when it comes to research.

Which is my third point– research strategies are not optimised for this organisational reality. Instead, they proceed top down. This omits the key role that integrating processes (a concept identified by Torben Juul Andersen of the Copenhagen Business School) play in effectively moderating institutional intentions with individual (researcher) behaviours. These overlooked processes and systems are essential to organisational performance in decentralised organisations. In their absence, universities are at best the sum of their parts with an overlay of institutional intention – or, as one university was described to me, “less than the sum of its parts”. In our work, we develop a process with our clients that provides a learning-based approach to managing the problem-belief-solution dynamic. 

The final issue is related to the process by which universities develop their strategies. I have seen two forms. The first is what I would call the “consensus” strategy. In this version a research strategy is constructed through internal consultations, with a higher or lower degree of “every kid wins a prize” in the final strategy. This is effectively a bottom-up approach, which generally results in a strategy that articulates existing capabilities. A “status quo” strategy might be a good description, or a “sum of our parts” strategy. This completely eschews the problem-belief-solution dynamic. A second approach is to outsource strategy to big consulting firms. The first phase is almost always a large-scale analysis of “current state” followed by an equally large analysis of “future state,” and a third large analysis of the gap between these two. The analysis is generic, consisting largely of looking at the internal distribution of disciplines, and the external funding performance. And, given the constraints on this and how these factors lead to homogeneity, the resulting strategy is by extension, general and homogenous.  

In large part this second approach is driven by large consultancies’ own business models. Instead, as we practice strategy with our clients (informed by Roger Martin formerly, of the Rotman School of Management, but also the collective impact sector and design thinking), the problem-belief-solution dynamic begins with an act of problem framing to drill down on what the correct problem is that we need to solve (e.g., “how might we be the leading Australian research institute in X, Y or Z?” or ‘How might we solve X large real world problem?’). This is followed by a creative act of ideating possible ways of getting there (i.e., possible solutions to our problems). And finally, this is followed by exploring the beliefs we need to hold for a particular solution to succeed (e.g., ‘we would need to have capability in A, a financial model that looks like B, a partner with capability C, for our competitors to respond like D, etc.’). Once we have these beliefs set out, we can begin to discount strategic options using analysis. For critical unknowns, as we call them, we conduct analyses to explore if these are insurmountable obstacles or not, what capabilities and systems are required, what assumptions we hold, etc.