Blogs

DIAGNOSIS – From 30,000 ft

Harvard Professor Clayton Christensen wrote a worthwhile and readable article on the topic of management theory for the September 2003 Harvard Business Review. In it he offers a metaphor which is very relevant to the topic of diagnosis. The story goes like this.

A patient walks into a doctor’s surgery with alarming symptoms. When she is ushered into the doctor’s rooms she begins telling the doctor of the symptoms. Before she can finish, the doctor hands the patient a prescription and says “here, take these, they worked for the last patient.”

The situation Christensen wrote about in this anecdote is clearly unacceptable when it comes to our health. The question posed by Christensen and in this blog is, is this same situation acceptable when it comes to our organisations?

When it comes time to make changes to how the organisation works there is no shortage of recommendations on how to go about it. These recommendations are usually based on how one solution is purported to have worked for another organisation. We see this general advice offered through a plethora of books and courses on topics like leadership development, innovation, change management and strategy. Some consulting services follow a similar, broadly applicable approach too. And while general advice may offer some help, the notion that it right for a large proportion of organisations, in a large variety of situations, is not convincing.

If we were to translate this broad approach using a health metaphor, it would be akin to recommending paracetamol or aspirin as an intervention. And that’s ok for a mild condition where the intervention is well understood as symptomatic relief, but unsatisfactory for anything more. To put this another way, the assumption built into the logic of this general approach is the same as in Christensen’s doctor-patient anecdote; that one organisation and its situation is the same as any other.

We are not advocates of a general approach as we are of the view that each organisation, and its current context, are different. Therefore, we advocate that each change effort is undertaken through a process of diagnosis. Further, our nuanced view of diagnosis is a little different from the commonly held view which positions diagnosis as an early phase of a linear program of change. Instead, our view is of diagnosis running the entire course of a change program, altering its focus as the change progresses. Let me explain.

In the early stage, diagnosis points teams towards opportunities for improvement and transformation and in doing so helps build support for embarking on a change strategy. Because the focus is both towards the external environment as well as internal, initial diagnostic data helps the organisation understand the full implications of the environmental context as well as helping to establish expectations about the prospects of change and the effort required to achieve outcomes.  The diagnosis process surfaces data that guides decisions on where to start a change process, as well as guiding the design of interventions. This is the common use of diagnosis.

We think of diagnosis extending through the change process and beyond so as to monitor the effects of interventions and to derive data relative to the desired, expected and actual outcomes. We do this because organisations are human systems which, obviously, don’t comply to designs in the same way as mechanical systems do. We fully expect to be surprised, in some way or other, by how the human system responds to changes in management systems, structures, polices, processes, technology, new capabilities or any other form of intervention undertaken. We also expect stakeholders in the external environment to respond as the organisational systems change, and as the organisation in turn responds to the marketplace. This idea is represented by the squiggly line in the adjacent diagram. The best metaphor I have is that of an aircraft which spends most of its time working out where it is relative to an ideal course, and making adjustments accordingly. You might label our view of diagnosis ’emergent’; representing the organisation relative to its journey towards its strategic purpose.

For diagnosis to be useful it has to draw data from multiple perspectives so as to more accurately reflect the whole system. For example, it would be valuable to assess the content and design aspects of the strategy, various systems and processes including scorecards, the structure, leadership and management capability and to understand the effects these produce. Effects play out in performance, climate and culture with insights accessible through the views of stakeholders the organisation serves. As previously noted, diagnosis also needs to incorporate change; forces for change or status quo, effectiveness of interventions, intended and unintended outcomes relative to the goal. The actual list and detail is of course more comprehensive, but better served by future blog posts.

Collecting data for diagnostic purposes is part art and part science. Some data is quantitative and sourced from data stores such as financial, inventory, and manufacturing systems. Some is qualitative and comes via interviews, conversations, and feedback systems. All of these are pieces of the whole system, none complete. As a consequence, sense has to be made of that data, and remade as new information comes to light.  Sense making can be even more artful as the process often benefits from diverse interpretations of what may appear to be ‘obvious’ data and trends. Our inclination, then, is to encourage broad and integrated conversations around both the raw data and interpreted diagnosis. It’s our view that functional interpretations are less helpful and often miss time-sensitive strategic opportunities.

If we are to accept the notion that organisations and their context are unique, and that decisions relating to why, when and how to change are critical to ongoing success, then the case for a robust approach to diagnosis is clear. Does your organisation have a suitable cross-functional diagnostic process built into its design?