I am siting in Singapore overlooking the straights from Sentosa Island, about to contribute to a conference. (Sorry to boast but the view is … distracting)   But still time for a blog on operating models.

Mark Lancelott of PA Consulting and I were together recently to discuss two things – how to turn strategy into something that is useful for designing operating models and what are the steps in the design process.   This blog is about the second of these two issues.

We are both of the view that you (we all) should start with value chain analysis (see recent blog on this).   For those who are not sure what I mean by value chain – then I am using Michael Porter’s concept – which is just a high level process map.  But the key is that the focus is on the prime operational steps needed to deliver value.  (There are a couple of other blogs on value chains – look under “categories”)

Value chain analysis however is best done after you are clear about some things.   First, the product/market segments you are trying to serve.  This comes from strategy.  Second, the basis of advantage that you are trying to exploit or build.  This also comes from strategy.  Third, design criteria or design principles (see previous blog).  So more about these thoughts in future blogs.

In theory, one should define the value chain needed to deliver value to each of the product/market segments.  Unilever apparently has 300,000 product market segments – so, if it takes 10 minutes to do each value chain, the analysis would take nearly three years of work.  Hence I say, in theory!   In practice it is usually possible to bundle product/market segments to a manageable number.  But, always be ready to go into more detail.

So, lets assume there are 6 product/market segments with enough difference to be worth drawing a separate value chain for each.   Lay out these six value chains and look for value chain elements in common, elements where economies of scale are possible and elements where similar work is being done that could be standardised.  So the next step in the analysis is to decide where similar elements could be combined into one (centralised) or done in the same way (standarised) or linked in some way (linked).  These choices are really strategy choices.  So it is important to refer back all the time to the “basis of competitive advantage”, and to check whether any decision that is being taken is likely to reinforce (good) or undermine (bad to the point of don’t do it) the basis of competitive advantage.

Having worked through these centralise, standardise or link elements, the next step in the analysis is to assemble the elements into an organisation model, with reporting lines and, if needed, matrix structures.  There will be a number of different ways of doing this.  So the process is one of laying out the options and then choosing the best, given the design principles that you have developed as one of the outputs of “strategy”.

Initially the organisation model will only have the operating work defined. It is then helpful to add into the organisation model the supporting functions that will be needed – finance, HR, IT, linking functions, standardising functions, etc – as well as the governance committees and decision powers.

After this, proceed as per my earlier blog – describe each box on the chart as a capability (best to do this when laying out the value chains).   Think through what is needed to deliver each capability – people, process, technology, buildings, locations, IT systems, culture, governance, etc.  Start with whichever of these is most important to the work you are doing – but remember that they all need to align.  So just focusing on IT systems for example may leave your thinking unbalanced.

And, hopefully in less than three years of work, you will have a target operating model …. simples! (for those familiar with the meerkat advertisement)