Reorganisations: Method in Madness?
Re-organising in response to external triggers is intuitive. Reorganising proactively in response to changes within the organisation is not; but both are essential
The conventional wisdom about re-organisations is that they are self-importantly announced by (new) CEOs, gleefully anticipated by consultants and dreaded by employees. There is a measure of truth in each of these statements. But recent research about re-organisations by my colleagues and I suggests there are also several things missing from the picture above.
Re-organisations are inevitable for most organisations. This is because no organisational structure is perfect – every structure emphasises certain kinds of interactions between people but de-emphasises others. Functional units help gain critical mass and economies of scale- but impede cross-functional coordination. Divisional units (i.e. product or customer or geographic divisions) are good for cross-functional coordination, but create duplication and impede cost-efficiency. Therefore, no organisational structure will solve all problems- and if the problems the organisation confronts changes, so must the organisation structure.
We can think about two broad classes of triggers for reorganisations: Internal and external. Major changes in technology, in regulation, or in economic conditions are obvious external triggers. Think of the internet and the clicks and mortar problem of the late 1990s, the repealing of the Glass-Steagal Act and the impact on banks in the US or of recessions and the need to cut costs as examples of each of the above. It’s the internal triggers that are less obvious.
The research that I have done over several years with my friends and collaborators - Ranjay Gulati (Harvard), Marlo Raveendran (UC-Riverside) and Freek Vermeulen (London Business School), suggests a surprising internal trigger for reorganisations - the fact that people within them adjust too well to existing organisational structure. Paradoxical? Yes, but nonetheless true.
Consider the fact that divisional boundaries inevitably become silos. While boundaries foster collaboration within them by creating accountability, incentives and a sense of identity, exactly the same things impede collaboration across divisional boundaries. Similarly, resource allocation typically mirrors the organisation structure. What this means though is that opportunities that fall between units or are too low within units to receive senior management attention will struggle to gain resources. This is one important reason why it is essential to periodically reorganise. In fact, my colleagues and I have argued elsewhere for the virtues of “Change for Change’s Sake”. 1
Both effects illustrate the surprising fact that as people within an organisation adjust to its structure, this will inevitably make it necessary to change the structure. Why? Because no structure is perfect. Complaints about “Silo Syndrome” basically highlight the fact that the divisional boundaries are not perfect at enclosing dependencies, as clearly some seem to be cutting across them. This may be the logic at work when employees see reorganisations even though neither the strategy nor the competitive landscape has changed.
Another statement often heard is that CEOs reorganise after taking over, more or less to announce “Hey, I am here”. I wonder if in such cases we have considered that a new CEO may have been called in precisely because the organisation needed to change? There may be method in the madness after all.
For a future column, I want to tackle the topic of “new forms of organisation”. One can’t look around today without hearing of crowdsourcing, contests, hierarchy-free organisations and other such radical concepts. How new are these really? Can we learn anything of relevance from them for our more “conventional” Let’s find out! Take the short survey “Radical Organisations” (Click here) to highlight what you consider to be radical new forms of organisation in your sector. I will investigate these, summarise and discuss the results in my column.