I was discussing my last post about capability maps with David Winders and we got into a discussion about the timing of work o organisation structure. David was explaining that one of the benefits of capability maps is that it keeps the analysis and discussion away from organisational structure. This makes it possible to engage all the senior people in the work without the inevitable posturing for power.
I pointed out that good value chain maps do the same job. They focus people’s attention on the work that needs to be done not the organisational units who will do the work.
But, David then said that he likes to leave work on organisational structure to last in his operating model projects, whereas I like to do this thinking straight after value chain thinking. So my model POLIST (son of PILOS) – has Processes first and Organisation second. Interestingly, David’s model CCPPOLDAT (Customer, Channel, Product, Process, Organisation, Location, Data, Application, Technology) also has organisation after process. But, in practice, he likes to leave work on organisation as late as possible to allow as much engagement as possible.
I am sharing this because I have not thought deeply about the order in which I do things. Hence David’s comments caused me to stand back and wonder whether he is right. I recognise the politics that surface as soon as you start talking about organisation. But I have usually found that my organisation modelling tool and the 9 tests of good design are powerful enough to cut through the politics. Also, “I” in POLIST, standing for IT and other Links, cannot be addressed fully without knowing what the organisation structure is: after all the links that need to be addressed are those that cut across the organisation structure. So, the value of doing structure early is that these linkage issues are exposed early.
However, I could do work on Location and Suppliers and People Model (a part of organisation) and some IT issues before focusing on structure. So I am interesting in trying this out.
Ashridge on Operating Models 
Andrew
I think we are agreed that good value chain maps and good capability maps are the same – the only difference might be the scope of the two maps, as I understand your focus to be more on the primary value stream than on all the capabilities that might be necessary for a functioning enterprise.
With respect to the “timing” issue that you have raised, there are several different issues that arise (in my experience):
a) any such maps are developed in collaboration with the existing management team where they will have a natural propensity to have an interest in what they “own” and what they are responsible for – this means that the presentation of an abstraction from organisational structure may be necessary if they are to shift their thinking and the associated constraints of the present structure
b) there are a wider range of options that can be explored if the map is not constrained by the current or even proposed structure too early in the development of the “architecture” – put the other way, determining structure too early may place an unnecessary, inappropriate and performance limiting constraint on any proposed operating model / architecture
c) in line with David’s thinking, the nature of capability mapping is to be “dimensionally agnostic” – not just organisation structure, but location, process, information system, technology – to offer the maximum degrees of freedom in which to explore the “optimum” architecture and hence, operating model
d) there is a growing recognition by McKinsey and others that as more capabilities are able to be supported by technology, there exists a new challenge in combining the “residual” components into meaningful work for the people involved. That, too, suggests that organisation structure should come later in the process.
So, my own thinking and practice is in line with David’s – leave organisation structure until later in the process – so that it does not unnecessarily constrain viable and sustainable options.
Some further input …
Just re-reading “Organizing for the future” McKinsey Quarterly, Jan 2016
There is an interesting section on – Think ‘platform’, not ‘structure’ There is a strong linkage between capability and platform (providing capabilities which are used by the enterprise). Hence, developments in these directions are also tending to defer “structure” questions until later.
You might also want to consider the earlier section – From bedrock to quicksand – which talks about companies being compelled to reassemble tasks into meaningful jobs, and even prompt the reconceptualization of the very idea of what a job is!
My view is that the nature of these changes is such that conventional approaches to operating models and operating model design may not prove adequate. Interesting times ahead!!
Good comments Peter. Because I am very comfortable working on organisation structure and because I have often done projects in the past where the focus was organisation structure followed by consequences for the operating model, I am comfortable with the political issues that are raised and with addressing these early on. But I agree that it is important to be clear about the work that needs to be done to deliver value to the customer/beneficiary and to organise in the best way to do this work.
I can also see that, if you prefer to address the organisation issues later in the process, you need a capability map as a vehicle for defining all the activities (not just the value chain activities). So I think we agree!