Wednesday, May 27, 2015

Foundations of a General Theory of the Organization IV (models)

In my last posts devoted to the subject I gave a very high level overview of how different organizations (classified by the different ends they pursued) had evolved through History, and ended with a very condensed bibliography of the essential reading materials to be able to understand the most evolved form of organization (the multinational corporation in our current stage of capitalist development). I noticed since then I had made a glaring omission, as I forgot to mention there a very important book which helped me understand how only in the last Century a certain structure has evolved that enabled the separation (even dare I say the opposition) of interest between the owners of capital (which originally fund and finance the creation of enterprises) and the “technostructure” (the managerial class and their white collar adjuncts) who runs the companies created with that capital, and from which I drew heavily to arrive at my proposed understanding of the end of  commercial/ productive organizations as the betterment of the social recognition of their workers, regardless of what their explicit mission claims to be. That book, of course, is The New Industrial State, by John Kenneth Galbraith (by the way, although in the 90’s his ideas of vertical integration –very much in line with the mentioned Coase- seemed to fall out of favor in exchange for a focus on outsourcing and loosely coupled conglomerates that renounced to substantial amounts of the value chains which they belonged to, I can not fail to observe some resurgence in the second decade of the XXIst Century of the centripetal tendency to consolidate again “from design to retail” in most industries). I’ve already edited the last post, but wanted to give such important milestone some added visibility here.      

Now having defined the conceptual framework, completed the historical overview and provided a concise set of references we can turn our attention to the development of my own theory of the organization: how they work, what regularities we can identify in their functioning and how to steer them in the desired direction. To that end we will need to first talk a bit about models, or convenient ways to represent the most salient features of an organization that are easy to manipulate and can represent the behavior of the real thing with a reasonable degree of accuracy. Before explaining the proprietary model I’ve developed in the last decade, and to better understand what they are and how they work I’ll present some of the most popular ones out there before talking about my own.

Probably one of the better known, more widely used models of how an organization works is the one originally proposed by Harold Leavitt in 1965 known as the “diamond model”, which sees the organization as consisting of four separate aspects (and pretty heterogeneous ones at that), namely:

It is not immediately evident why those four aspects were chosen, rather than other more salient ones (like facilities, geographical locations, raw materials processed/ stocked, relationships with customers or providers, applicable laws and regulations, etc.) but specially when thinking about introducing changes in the organization it helped to consider how each of those aspects would be affected, and how the modifications in any of them would impact the rest. Indeed, the utility of these kind of models is limited to helping people (generally external consultants) think about the consequences of a certain well defined and temporally bounded change in some of them, ensuring they do not overlook and leave unmanaged any of those potentially fateful consequences. They are not that good for describing a stationary given organization (how would you convey how the “people” side looks like in a stable situation? How would you include in that what each person knows, expects, obtains, and performs for the organization? A similar problem shows if you wan to use this very simple model to reflect the “technology”, let alone the “structure” or the “tasks” of any given group).

In my first years as consultant I used a lot a modified version of Leavitt’s diamond, adapted to better fit the particular strengths (perceived or real) of my employer back then (a trailblazing consulting firm whose severe case of megalomania has been more or less validated by a continuing success in markets it significantly contributed to create):

Not that different, only technology was taken to signify a very precise part of what kept the organization running (its information systems), while strategy, people and processes were the other areas around which the services of the company were clustered (each bearing the fancy name that the marketing guru of the moment may had concocted). I still tend to find myself thinking about organization along those lines, as it became a very ingrained habit to consider those four particular aspects, subject to the same limitations as in the original inception. Although at least now I can see clearly the real use of this particular model: it helped us consultants sell more work, as after being contracted to update a client company’s technological platform (i.e. to install a new Information System) we could point to him that processes had to be redesigned to really reap all the advantages of the new technology, people would need to be assessed and most likely retrained so the new processes were used and the new technology harnessed, and if we saw the opportunity the strategy of the firm could potentially be reviewed and finessed to monetize to the hilt the new capabilities created by the combination of technology, people and processes we would unlock for him…

Which was all well and good, but then I learned that one of our niche competitors in the area of strategic consulting were doing us one better, as with the help of Tom Peters they had developed and were using in their engagements a (now famous) somewhat different model, the “7s” that identified not four, but seven aspects of organizations that had to be taken care of in order to steer them successfully through the turbulent waters of adapting to an ever changing world:

So while naïve us were stuck trying to sell “people” stuff (training) the whiz kids at McKinsey could sell “skills” upgrades; while we adapted new technologies they piloted “systems” work (only in this case the last laugh was on us… strategic consulting was wildly profitable, but it was truly difficult to sell something above 10 million dollars even to the biggest corporations back then, while IT work easily racked 100 millions, and before the Century’s end the first above-a-billion contracts were already being signed, although at a ridiculously low margin and based on saving estimates that would make most of them go sour very fast)… and on top of that they were probably selling “staff” work, “structure” work (I wonder if that was the equivalent of our more modest “processes”) and what I consider should be the real top of the pops, “style” work (damn, I would love to see some of the directors of my current employer trying to defend in front of the board to spend some of the company’s coin to have some recent Ivy League graduate help us change our “style”, but ludicrous as it may sound a good amount of bucks have exchanged owners, and careers have been built, on such flimflams).

Now I’m not saying such models are off the mark, or entirely useless. They were the core of many PowerPoint prodigious presentations, and many an organizational change (in Fortune 100 companies) has been hatched using them as a framework. Indeed the new century has brought us a plethora of ever more complex, ever more innovative models, like the Holonic enterprise reputedly used by the former Internet darling company Zappos, which tries to reimagine the traditional corporation as a “Holocracy” where every branch and subdivision can grow and expand in answer to market demand without loosing their creativity and potential for innovation:

There are even more abstruse (Burke-Litwon Causal model, McMillan Fractal web, etc.), but for our current purpose we have enough with what I have shown so far. Any and all of them are decent boilerplate fillers, good (lacking anything better) to grab the attention of a Director of Organization (or, even better, of a CEO or COO) and potentially make him understand what your organizational consulting firm can do for him/ her. They can even be useful providing heuristic (“hard and fast”, although also “fast and loose”) rules of the kind of things that may reasonably happen when an extraneous factor (from a new information system to a new CEO, and of course a new organizational structure with new reporting lines, new hierarchies and new responsibilities for the existing people) is introduced in an existing organization. But all of them are unsuited for my purpose of better understanding why organizations are formed in the first place, what makes them persist along time, what help them sustain changes (be more resilient) and what can be done to make them more efficient. In order to fill that gap (which started as a theoretical yearning, although as I developed it I kept on finding sometimes unexpected practical applications) I had to develop my own model, which unsurprisingly I’m calling the ADVISE model (as it is ultimately intended to help organizations of any persuasion become more Adaptable, more Dominant, more Voluntary, more Isocratic, more Simple and more Egalitarian… why? Because I will argue that such organizations achieve their ends more efficiently and can consistently deliver higher benefits to their members, although I recognize a lot of empirical work would need to be done to adequately substantiate such claim). The outlook of the model will look familiar to anybody that has accompanied me this far, and has read posts I, II and III of this series:

Taken directly from the definition I proposed at the beginning, all the elements that have to be taken into account are there: a group of people (members) situated in a certain environment (that imposes certain constraints on them –regulatory and normative, limiting what resources they can appropriate and in what kind of activities they can engage, what rights they enjoy and  what protections they can count on), pursuing certain ends. In that pursuit they may consume certain resources, and most likely will need to specialize and divide their labor, thus there will be different roles, and different relationships between them. Although roles can be repeated (there can be multiple project controllers in a company) the way each individual perform those roles is unique (so Jack the power hungry, somewhat neurotic and firmly autocratic project controller will interact with other members in a totally different way than Suzy the mild mannered, depression-prone but witty project controller), and will in turn give rise to totally different group dynamics.

My contention is that a complete description of the people, the roles they perform (what they are expected to do, how and when), the relationships between them, the environment they operate in, the resources they consume and the end they pursue would give us the kind of “thick” description (in the sense advanced in anthropology by Clifford Geertz) I’m after. Of course such a “complete” description may in the end be unattainable (for starters we are still pretty bad at “describing” actual people, and I am weary of all the efforts so far of psychometrics, and abominable constructs like the Five Factor Model and Myers-Briggs Type Indicator prove that people are always more complex and multi dimensional than our best efforts to simplify and pigeonhole them), but I believe that even an incomplete description along these lines (containing the most basic outline of what people there are, who each one interacts with, a most cursory description of their role, what they consume and what they are after) is prone to leave less loose ends, and to contain more useful hints on how that group will likely evolve in the presence of certain exogenous factors (or how, in the absence of such factors, left entirely to their endogenous impulses, they would either attain an stationary state or regress) than any of the alternative models out there.

As they say, the proof is in the pudding, so the next chapter(s) will let each reader decide to what extent that claim is founded.      

No comments:

Post a Comment