The shape of an organisation has been a hot topic in organisation design for a number of years with claims and counter-claims from different camps about which shape is best and how to achieve this nirvana. The shape of an organisation is a direct consequence of:
- the number of layers; the number of rows on an organisation chart, which determine how “tall” an organisation is.
- the spans of control; the number of direct reports, or columns, there are under each manager of people in the organisation, which determine how “wide” the organisation is.
These two parameters combine to produce a myriad of shapes for an organisation; from the tall, thin pyramid with many layers and small spans of control; through the hourglass with layers of middle managers with small spans of control; or the diamond where middle managers have large spans of control; to the flat and wide organisations with few layers and large spans of control.
Each shape has its admirers supported by wisdom presented to support the beauty of their desired shape. “Best practice says spans of control of 7-12 are best,” “Seven layers is ideal,” “Flatter is faster,” “Large spans of control and self-managed teams in the operation lead to better productivity,” “Holacracy is the answer,” and so forth. The unfortunate truth is that despite there being no end of snapshot case studies that demonstrate, with a sample of one, that improvements have ensued as a result of changes made to the shape of an organisation. There is no evidence base to support the long-term benefits of any one model over another. Each has its benefits in relation to the strategy it is enabling.
Redesigning An Organization: Don’t Start With Shape
If sufficient thought and effort goes into designing an organisation to deliver its strategy, basing design choices and decisions on solid foundations, design criteria, parameters and constraints, then the resulting shape will prove right for the organisation at that point in its history. Decisions about the shape of the organisation serve as an output of the design process, not an input to it. Any ongoing changes can be made as part of planned strategic reviews and will be driven by a need to keep the organisation aligned to its strategy. Not by deciding that changing shape would be a good idea and looks easy.
Despite a lack of compelling evidence, organisations and their CEOs continue to feel attracted by the idea of starting redesign efforts by addressing the shape of their organisation. When starting at this point, nobody ever seems to say that they want their organisation to be taller and thinner; they always seem to be aiming for flatter and wider. In most cases, this amounts to accountancy-driven thinking — a belief that stripping out layers of management and increasing spans of control will lead to a more efficient organisation. There’s a compelling and simple logic to the belief that reducing management layers leads to increased profitability — or is there?
The reducing layers of management reduces costs myth.
Without redesigning the operating model and thus redesigning the management work, simply stripping out management layers does not significantly reduce the amount or complexity of management work. What we often see in larger organisations is that, as a layer is taken out, two things happen. First, the remaining managers have larger teams to manage. Second, as there’s no investment in capability development, the complex work carried out by the departing managers always moves upwards. Both actions make the remaining management jobs ‘bigger’ in simple Job Sizing terms and, as a result, they become higher paid and the organisation experiences grade drift. Taking out a layer of management will lead to any financial benefit of headcount reduction eroding over quite a short period of time by an increase in the cost of the remaining managers.
Changing layers v. levels.
One of the best-researched, validated and documented bodies of evidence in the field of management science relates to the levels of management work.* This body of evidence shows us that there are up to seven levels of management work required to deliver any given strategy. But this work is often simplistically misinterpreted to be saying seven layers of management is the ideal. This body of work also provides us with a guide on how to design these levels into management role. The organisation must then decide how to organise these management roles into jobs in hierarchical layers.
With no change to strategy, it’s possible to take out layers. But the management levels will remain the same and the work must still happen. You can’t take out management jobs without reassigning the management work to other jobs; potentially all management jobs will change as a result. However, if the reason for addressing the layers of management is because the strategy has changed or there’s a belief that the management work is wrong, then a more fundamental redesign effort will be required. Starting from the current work will be insufficient; you must start at strategy and redesign from there, as “form follows function.”
- Don’t start a redesign effort by trying to change spans and layers; this is just restructuring for the sake of it.
- Taking out a layer of management requires the management work to undergo reassignment. This will drive grade drift as well as increased costs, which will erode any potential benefits over time.
- There is no magic algorithm for spans and layers. Rather, the right answer for the chosen strategy will emerge through holistic design efforts.
- Decisions about shape serve as an output of the design process, not an input to it.
Peter Turgoose is a Senior Consultant at ON THE MARK.
To learn more about redesigning an organization, contact OTM.
On The Mark’s experience and passion for collaborative business transformation that’s supported by pragmatism, systems thinking, and a belief in people is unparalleled. OTM has been in business for 28 years and is a global leader in organization design.
We understand that you may want more information on how OTM works with clients, our terms and further information from our extensive case-study library. Contact us about careers, more information or how we work here.
* Requisite Organization: A Total System for Effective Managerial Organization in the 21st Century (2nd Edition) – Elliott Jaques (1996)