Organizational Thinking

I recently attended a talk by a leader in the Workplace Learning field and was struck by his endorsement of centralized training and learning management systems. This approach seems to me at first blush to be off track with what really happens in organizations that needs to be talked about in terms of learning. However, I do see a couple of aspects of this that make sense. There are two things to consider here. First I think it makes total sense to centralize data about training. But this is an interesting notion. I think what we can do with data now is centralize it anywhere. In other words I can look at data about the whole from any point in the network (node). If I have access to some sort of 'aggregator' application and it is pointed to the information I need then I can get it. So, in a sense this makes the idea of a 'centralized data repository' kind of obsolete.

The second issue is about the functions in a company or organization that need to be 'learned'. In my view there are at least two fundamental levels of reality in any organization. I have begun to call these the 'site' reality and the 'culture' reality (these ideas are growing out of my work in ambulatory medical clinics and so the term site and culture are born of that. The word 'reality' is meant to capture the 'phenomenology' specific to that space. For example, managers and to some extent supervisors, are more attuned to what I will call 'site' issues than line workers or contractors. The 'reality' for a manager is different than the 'reality' for a line worker. Of course this argument is full of assumptions and other implications that need careful explication.).

The site reality involves the infrastructure and policies that transcend local activity to help knit together disparate people, locations, or practices. We could think about the site reality as being information about standards and more formal classification structures. I like using the word 'site' for this idea because it preserves the necessarily local flavor of how people see and interact with management and supervisory responsibilities. The 'site' links the local with the global in terms of an organization's structure. So an LA site and a Boston site are both aspects of the corporation that is headquartered in New York. The issues that are important to New York about LA and Boston are system issues and communications that are based on the notion of being able to scale standards, policies, and classifications.

Training for site realities lends itself to a 'centralized' kind of model. However, I think the term 'centralized' is misleading. A better term might be a 'standard' model. This standard model would be copied everywhere in the system (throughout the sites).

[A brief aside here on memes - memes are thought of by some as replicators on a par with genes. When we 'replicate' or copy a training throughout a system we are making an effort to direct the spread of memes through a system. To the extent that we can map a meme (I think of them as persisting units of language) to system infrastructure we are going to 'probably' see some gains. The examples that I heard in this recent talk I mentioned above bear this out. In one example there was a direct effort at 'training' dealers about a certain product and the sales strategy around the product. Because this 'site' training was around a product and around a process for selling the product the memes used stuck pretty well to the infrastructure that was on the ground (so to speak) and the central office saw some gains. Another example would be IT training for an ERP or CRM application. Because the applications are on the desktop of workers and this is what they are (forced) to work with there will be some obvious changes in behavior associated with the training experience. What's really interesting about these IT issues is the way that the culture (see below) interacts with the site mandates or standard models.]

Anyway, the site reality lends itself to standardized training and the roll up of data around that training made available via the web to either central office or site is a great thing. People can see and measure progress about the standard model. We can then connect the standard model to the business model where ever it makes sense. The fellow that I heard speak was clear about the need for workplace learning people to make sure that we somehow affect the bottom line (increase revenue, reduce costs, improve cycle times). This makes total sense.

However, there is another reality that is not much touched by the standard model propagated through the sites (by the way - you can conceive of multiple sites in one organization in one place - say a distribution organizaiton that has no satellites and no sales force - just the people in the facility. It is still full of 'sites' using my logic.) This is the zone of culture. Now, I don' t want to get into protracted definitional polemics about the word culture so suffice it to say what I mean is the very strange intersubjective reality of real people in real relationships talking, thinking, feeling, and breathing together. I am serious. We are so often in our metaphors and models about how the world should be, was, or is we forget that we are all in an ongoing kind of acid trip of 'experience'. This experience is what I mean by culture. It is us - in a Petri dish if you will, doing our thing. This is were the standard model breaks down. People subvert, diss, twist, ignore, misuse, and otherwise abuse all sorts of 'site' related messages and memes from management or central office. The amazing thing about this is that the culture reality also pervades the masters of the sites! Anyway, what is the workplace learning professional to do about this reality?

That is my question.

The first thing that I think we need to do in organizations in terms of information technology is let it rip! By that I mean the flowering of kludgy applications, work-arounds, emails, web-surfing adventures, macros, spreadsheets, powerpoints, etc. need to be encouraged and supported. This is the stuff of culture expressed at the level that stuff happens. This is where, if we really want to develop good standard models, we will get our best information. So, I say let 1000 flowers bloom. This touches on the link/node and emergent classification structures that have so much become a part of our web oriented lives (see Clay Shirky’s recent excellent little essay at:

http://shirky.com/writings/ontology_overrated.html

His use of the word ‘ontology’ maps to my general notion of ‘site realities’. But that is another story. Anyway the thing to do is listen. The second thing to do is ‘pay attention’. Paying attention means that we listen without letting our assumptions cloud our understanding. This is a harder thing to do than most of us realize. I am doing a lot of work with and thinking about dialogue and it is becoming very clear that to do good dialogue we have to be able to work on a much deeper level psychologically than many of us are used to. So what I am saying is to let the local sites thicken with local information and then listen and pay attention to what you are hearing and seeing.

Then we can begin to both collect data about whether the more subtle goals of the standard model have been met and gather information about how we can develop better strategies that may or may not lend themselves to propagation via the standard model(s) that exist throughout the sites. blah blah blah.

So what I am trying to say is that ‘training’ does not work at the culture level. Learning happens and a good friend, co-worker, mentor, customer, friend, boss, or whatever can help the process through simply showing up and hanging around.

This learning professional that I listened to mentioned ‘Communities of Practice’ in response to a question I asked that was kind of around this whole topic. He mentioned the activity around this notion and I really appreciated that. However, I am skeptical. I have read the work of Jean Lave and Etienne Wenger and am really taken with the participatory and embodied notions that the concepts imply. But, these ideas don’t lend themselves to scaling via standard models.

They are complex and emergent phenomena and require a different set of tools. More on that later no doubt. I am currently working on getting lots of data on trust into a framework that is suggesting to me that we may be able to map behaviors as attractors in what I am increasingly calling the mother (phase) space of trust. Needless to say - I really enjoyed this person's presentation.

It is very interesting to be alive in these times. The paradox of our various affiliations is palpable. We (the collective we of humankind) seem to hunger deeply for connection and meaning yet our practices continue to keep us separate and alone! Using standard forms or practices (conferences, speeches, workshops, rally's, etc.) don’t seem to make a difference. Most of our jobs systematically deny our need for rich complex interactions.

My question is: What practice does make a difference? It occurs to me that like Gandhi, Mandela, & King the first step is to change ourselves. After that I think the circle radiates out in local and meaningful ways. After that maybe a revolution - but probably not. I think what I so appreciated about teaching at GU this past spring was the opportunity for a sustained conversation - but it does not escape me that the conversation was not completely voluntary - there is 'coin' involved.... the struggle we all make to survive and get the tickets (degrees) or money to meet our goals.

So how do we get around this? I think that our individual commitments is a good place to start. To meet and talk and maybe write a bit. To begin slowly to organize around a core idea (what is the idea?) and then let things happen. Community is important.

One idea I am having is about complexity - I am increasingly interested in working with complexity models to help make sense of collective phenomena. I have been working with some relatively large groups (n = 500 & n = 150) and doing qualitative analyses of a Trust construct. I am then doing frequency counts of coded data (16 coded behaviors) and then displaying the data in a color coded frequency chart. Very simple stuff but very interesting from the perspective of complex systems.

I am seeing the Trust construct as a 16 dimension phase space. The frequency counts suggest a possible interpretation as attractor basins within the space. This means (to me) that there is a tendency in the organization in question for people to behave in particular ways in reference to trust that are consistent with the attractors (big assumption here - BAH). These attractors tend to 'pull' people along their gradients in ways that are mostly unconscious (also BAH). When we can 'see' the basins (in my imagery these are whirlpools of similar memes, behaviors, and thoughts) we can then design both strategies and interventions that may help move or perturb the attractors into more or less significance.

The meta-pattern revealed by the frequency data is compelling. The salience of individual behavior frequencies was preserved across polarizing questions (the narrative data on Trust that was initially coded). So the behavior (for example ‘Tells the Truth’ was consistent across questions but particularly salient for one question (e.g., What is breaking trust?). The next step is to triangulate this frequency data with other available survey data (Likert data).

This next paragraph is pure BAH. From an individual psychological perspective I think that increased consciousness and awareness bring an acute sensitivity to these complex social topographies. I think our hearts and minds are able to recognize when we are being 'pulled' or directed toward a way of being that is either consistent or inconsistent with our intentions (intention is the other half of this equation). Have you ever started a new job and felt that strange sense of ‘wrongness’ when asked to participate in a standard organizational rite? I think this sense indicates a sojourn into an attractor basin (if not the whole phase space) that is inimical to some core tacit value. Of course it is important to remember that values are not absolute and are, in fact, learned and ‘picked up’ in the contexts we inhabit. So, that’s why after a few weeks in the new job it is hard to remember you even felt weird. Let alone why.

So, the analysis of the relevant 'phase spaces' of an organization or a culture can help surface into collective awareness the patterns that are driving our behavior. What is important, in my view, about Trust is that it is one of our very basic background, tacit, and ubiquitous contexts - it is one of the 'mother spaces' that we inhabit. Again, my imagery sees us living in a multidimensional world of nested contexts - each with its own flavor and gradient.

In terms of Follett and participatory democracy I think it is interesting to note that the boundary crossing that seems to be necessary for participatory democracy in the large sense is made difficult by the strength of the attractors of individual experiences. Which is interesting in light of the wide spread recognition of the reduction in 'meta-narratives' or narratives that create meaningful practices (attractors) across wide populations.

So, one of the questions that I have is: do we need to create new meta-narratives (or are they being created as we sit?)? Is a meta-narrative the post-hoc, or after the fact conception of the shape of a lot of years of individual moiling around in states of relative chaos, incoherence, and separation? Is our job now in the creating of narratives? Or watching and becoming aware of their collective creation and then speaking them? Or both?

A Framework of Effects of Information Systems on Organizations

(a paper I have been working on (off and on) for a while)

Language defining and describing information technology in general and specific applications in particular (e.g. the Windows interface) lead us to believe that we understand the use of information tools and that their application in work environments is relatively unproblematic.
The buzz words and hyperbole surrounding the ‘information’ revolution that we are in the middle of sometimes make if difficult to see what is right in front of us. The simple facts resulting from the introduction of distributed information systems into time constrained and space constrained work environments are important to consider.

First I want to define what a ‘distributed information system’ is. Simply stated it is a system of communication and data storage that is networked via personal computers and work stations. It is a system that, from a technical standpoint, is centrally coordinated, and from a social standpoint is not centrally coordinated. Distributed information systems are hybrid designs that blend two historically unconnected organizational phenomena.

This is important to consider. The engineering know-how and the technical expertise behind the rational development and implementation of information systems, software application, and physical infrastructure are consistent with traditional organizational theory and practice. The work groups and professional assumptions of engineers and designers effectively and rationally produce the information products that we use today. These systems are also the first to take advantage of the collaborative networked work systems of the future. The reason for this is mostly attributable to the strong professional identities that emerge from legitimized competence in a technical field as well as the relatively higher salaries that these information professionals command.

The general effect and lack of centralized coordination of advanced information technology and distributed systems on the wider social environment of organizations is another matter. Here we see many non-professional workers with relatively weak professional identities expected to assimilate and adapt to the new environment that is, on its face, a very technical and rational system that purports to describe and specify a workers function and process. Now, by itself, this state of affairs would be little different than that which obtained with the advent of Fredrick Taylor’s ‘scientific management’ technique that swept organizations over the last 80 years. Workers were forced to adapt to the model or they lost their jobs. The behaviors required for adaptation was generally of a routine sort and required relatively simple training.

The difference is two-fold. First the ‘system’ of scientific management was static. Once the job’s parameters were described and specified – and once the control systems were in place the process went on with only slight variations. The static nature of the system allowed for the hierarchical control of subsystems in an overall rational manner. The worker need only be trained. With dynamic systems workers must be encouraged to learn – that is – to absorb and assimilate new knowledge that translates to new behavior continuously!

Second, the control tools were not under the control of the workers. The rationalization of the technology was opaque or invisible to the average worker. It was only through the ‘special’ knowledge gained by the expert that the reasons why the control mechanisms worked and their relationship to corporate goals were made clear. Workers had only to behave – understanding was not required, in fact, it was generally discouraged.

That state of affairs is now changing. Our distributed information networks are dynamic and the ‘user’ is expected to understand the nature of the system, their place in it, and their function’s connection to overall corporate goals. We expect stock clerks and production workers to behave like professionals.

The problem here is that these workers generally do not have the psychological or behavioral skills to support this new work. What is generally happening instead is that the dynamic feedback effects of distributed information systems (especially e-mail) are creating a hyper-grapevine or hyper-informal system of communication and meaning making that is more often than not at odds with the goals and practices of the formal rational system designed by senior management and their professional process designers.

There is both a problem related to the generic description of the ‘technical system’ as opposed to the ‘social system’1 as well as a problem within the social system that relates to the creation and sustenance of a ‘professional’ identity. The loyalty and allegiance typically associated with professional identities is generally missing in the rank and file work force2.

The solution to this state of affairs is to be found in the design of the systems that workers are expected to adapt to. It is not enough to simply ‘train’ a worker to use a new system but to build into or design the system features that direct the user toward a fuller utilization and understanding of the meanings and purposes behind the design itself. This means a fuller participation in the design stage, an ongoing link with designers and decision makers in the beta testing stage, and a permanent place in the ongoing management of the system itself.
This is nowhere more evident in the information technology/work process interface that occurs in all ERP systems. Users of the IT are active participants in the processes that the IT models, constrains, and monitors.

Workers must be encouraged to collaborate, reflect, and inquire as a regular part of their job. This posture signals a radical change in most organizations. Prerequisites for this type of behavior minimally include a work culture that is noted for its open and honest communication climate and the development of trust across all levels of responsibility.

The difference in the whole systems approach advocated here (as opposed to the typical approach developed by functional specialists) is that the creation of new behaviors and attitudes is not to be found in training or exhortations toward open and honest communication but through the systematic development of structures and systems that afford open and honest communication. The most important aspect of this shift in focus is in the developing of a holistic and coherent design plan for the organization. Numerous models3 exist to aid in this design that also help direct strategy for the penetration of the model throughout the entire organization.

On a more specific level this requires (for example) a rethinking of document design and routing, archiving of data, and an opening up of organizational meetings. Probably of most importance, however, is the active participation of workers at all levels in the development of goals & objectives and the criteria through which they are both understood and measured.

This approach requires a fundamental shift in management’s approach and abilities related to leadership. Leadership cannot be conceived independent of the particular context it is being exercised in. Each organization is different and each individual will come to their position of influence in different ways and for different reasons. Effective leadership is not developed through coaching alone but also through the development of a sound theoretical framework within which the dynamics of the particular organization makes sense. This ‘sense’ making does not come out of a book or through the development of greater skill at manipulating others. It comes through experience and the honest and straight forward engagement of the organization as a whole.

1.Goguen, Joseph A. "Toward a Social, Ethical, Theory of Information." In Social Science, Technical Systems, and Cooperative Work: Beyond the Great Divide, edited by Geoffrey C. Bowker, Susan Leigh-Starr, William Turner and Gasser Les, 27-56. Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers, 1997.

2. Eraut, Michael. Developing Professional Knowledge and Competence. London: The Falmer Press, 1994.

3. Rummler, Geary, A. & Brache, Alan, P. Improving performance: How to manage the white space on the organization chart 2nd Edition. San Francisco: Jossey-Bass Publishers, 1995

In reading Burke on organizational change I am struck once again by the need for clear epistemological foundations in thinking about organizations. For me the foundational questions to ask are: what is an organization and how does our role as 'observer' affect this perception?

What is an organization? And to answer this question we have to ask questions about observation, objectivity, and the possibilities for reasonable interpretations. So it seems we have to settle on two matters. First - what of the 'ontological' fact of an organization? Second, what is the role of the observer in this regard? In my view a great place to start is with the work of Maturana and Varela. However, their interpretation is best applied in the realm of biology and there is much to consider in translating it to the world of organizations in general.

The central point here is that we think about the organization as problematic in terms of its 'objective' existence. As individuals we see the organization through the lens of a story or narrative. I use these terms with much more confidence than I have in the past as I am more familiar with the idea that our perceptions and ideation about the world is an ongoing process that keeps itself going through a self-referentially grounded narrative. We talk (and think or feel) our way through the complexity of our ongoingness.

This then gets us into the realm of the 'observer'. As storytellers we take a stance toward our experience (of which the organization at whatever level is a part) and cast and recast it as an object or 'story element - character or setting' in our story. This narrative is partial as we can only take our perspective. Now it is true that experience and the absorption of others' experience through reading and other media can 'fill out' our perspective and help us a get a more ecologically valid view of our story element but it still remains partial.

This brings me to the issue of how we learn about, for example, a theory of change in organizations. We typically read about or hear this 'story' and may or may not take it as a definitive or valid model of that which we want to know. This makes me think of the whole notion of dimensions, levels, and categories. The literature on organizations and on change are full of 'models' that contain 'levels' or 'dimensions' that purport to represent some crucial aspect of an organization’s structure (e.g., organization chart) or process (process flow map). This literature (narratives) and the narratives flowing out of it have been until relatively recently grounded in realist and empiricist ontology and an epistemology holding the observer as unproblematic (as long as he or she conducted their experiments appropriately). That is changing however but not quite at the level of most of our individual stories about the world. Most of us are still comforted by models that contain dimensions, levels, and categories that map nicely to theories of causation in a straightforward (and I mean linear!) way. Myself included. It is mostly the way I see the world.

So what does this mean for scholarship or the exercise of curiosity?

Clearly the narrative is changing - and the change (oddly) is coming out of the very bastions of conservative thought (e.g., Harvard, Stanford, MIT, Yale, etc.) whose rank and reputation have been built on the edifice of these same ontological and epistemological certitudes. However, I may just be projecting here. The point I am making is that the 'new' science is in the air and beginning to have it affect on individuals. To wit my emphasis on Maturana and the 'Observer' issues.

What is required to conceptualize the 'organization' (in my view) is a relatively strong background in both the natural and the social sciences as well as a huge capacity for ambiguity and confusion. This is so because of the need to translate certain techniques from the domain of the 'hard' sciences to the 'new', and still 'soft', social sciences. So instead of importing Cartesian & Newtonian science methods into a social sciences framed as a machine or mechanism we now are understanding that non-linear dynamics modeling organizations as 'organisms in an ecology' gives us a better read on what is happening at work! And of course within this new social science the notion of the person and of subjectivity has become a central preoccupation. We participate in the story that we tell!

So, the issue for me is to come to grips with this to the extent that I can tell and hear the story as interesting, entertaining, and meaningful.

What I didnt' get into here and I probably should have is the question of the 'model'. Right off the bat I think of a representation or a Jamesion 'percept' - or what Bohm wrote of as a thought disguised as a perception. So our 'models' present to us what is familiar and reasonable as a likely picture of what things are really like. And the model 'model' does that before we even know that we are doing it! And this is changing too... our models of models have to change.

Systems Systems Systems

What a world.

Sometimes I am toubled by the thought that people don't honor the past in understanding the present - that so much 'new thinking' becomes another escape from really thinking about and trying to understand. Discipline and practice are vital. Systems Thinking is the hardest type of thinking we have yet cannonized in the tradition of western thought worlds.

Seeing Beyond Technical Solutions
Efficient and effective management of the system development lifecycle is the standard against which most IS design efforts are compared. However, the effective management of technical solutions does not adequately ensure that the technical solutions crafted at the strategic and business area levels will ‘take’ at the applications or performance level that the system is ultimately dependent upon.

Typically there is a fundamental mismatch between systems as idealized by designers and systems as used by front line workers. This problem is further complicated by the increase in cooperative work being done at the performance level. This cooperative work is the result of numerous factors but chief among them is the pressure on firms to work with a leaner staff and at the same time remain competitive and continuously increase productive capacity. Global markets and a general trend toward rapid adaptation to market trends indicate that this state of affairs is not likely to change anytime soon.

The basic nature of this split is illustrated by trends in the field of Human Computer Interaction (HCI). The HCI community is increasingly bedeviled by the fact that both settings and the users in the settings are heterogeneous in both their understanding of systems and the applications that the systems are put to (Bannon, 1997). This problem has seen the increase (at least in Europe) of new frameworks of cooperative design that entail a greater degree of open ended design work with a greater degree of dependency on work place variables. Nonetheless development continues apace in the U.S. using, for example, old models for the development of GroupWare products that hit the same user ‘wall’ that many standard IT products have encountered.

One response in the emerging field of Computer Supported Cooperative Work (CSCW) is the use of a new set of tools and techniques in the development of systems oriented toward relatively large scale implementation of complex information systems (Project Report - CSCW Symposium, 1996). Among these new methods are the practice of ethnographic research, system wide coordination of change initiatives, using the Intranet capabilities to funnel just in time user requirements information to system developers, and the development of ongoing feedback loops (also utilizing Intranet capabilities) connecting front line practice with strategic leadership and various IS/IT and production based modeling efforts.

Ideally a firm would use a variety of tools and techniques along a continuum ranging from the technical and abstract to the social and context embedded for the design, implementation, and ongoing use of information technology and systems applications. Unfortunately the capacity of organizations in general and the IS/IT field in particular to see the organization and its practices as a whole is compromised by political infighting and/or disciplinary specialization and compartmentalization. These patterns are difficult to change. But change they must – the pressures on firms to respond quickly and effectively to external perturbations as a whole are increasing.

A continuum outlining the range of possibilities is captured below and reflects the increased focus on context in general and the development of tools oriented toward making informal aspects of the work place explicit in the service of system design.



Assuming an organization can both accommodate and connect elements along this continuum then that firm will find an enhanced capacity to adapt as a whole to both external and internal organizational changes. From this perspective it would be fair to say that the IS/IT and organizational system as a whole functions in a manner similar to the human nervous system. There are both afferent (feedback to the strategic center) and efferent (direction to the distributed functions and processes) impulses continuously coursing through the firm. This, however, can only occur if the firm embraces the task of incorporating the informal and context dependent performance culture with the technical and rational aspects of system and organizational design.

A Design Framework for Reconfiguring IS/IT Functions
Assuming that the organization is in pursuit of holistic solutions to organizational problems a coherent framework for linking both real and virtual aspects of the organization is needed. The framework proposed here has its roots in the organizational science literature focusing on organizational structure and the management of knowledge (Jensen & Meckling, 1996; Nonaka, 1997; Schein, 1997), organizational and information systems dynamics (Goguen, 1997; Hall & Moss, 1997; Senge, 1990), and the theoretical foundations of social learning (Agre, 1997; Eraut, 1994; Maturana & Varela, 1987; Lave, 1988; and Piaget, 1971). The common thread of this design foundation is a more or less universal recognition that systems thinking (in the sense that dynamic wholes must be considered) is an absolute prerequisite for the development of tools and techniques to help understand and mange the 21st century organization.

Organizational Structure
Traditional organizational structures are function based and hierarchical. This is a plain fact and the transition from this traditional structure to the ‘flatter’ organization popularized in management books and articles is probably a long time in coming. This is true for several reasons. First there is an historical habit and pattern in organizational life that will change slowly if at all and only as new and progressive experiences are gradually incorporated into the overall culture. Second, the coordination, technology, and basic competencies required in a ‘flat’ organizational structure are difficult and complicated issues to engage (this document being an example of the difficulty in moving toward whole systems design). Consequently the design model used in organizational change initiatives should reflect the somewhat intractable nature of hierarchical organizations but also include processes and mechanisms that will sow the seeds of gradual but progressive change toward a distributed, process based, and highly adaptable organization. The work of Rummler and Brache (1995) is important in this regard for they include in their process based model a structural template with a lot of face validity and enormous explanatory power in both modeling and participating in the organizational as a whole. Their conception is central to the model outlined graphically below.



A key task for organizations to take up is the translation of the horizontal informal communications (knowledge) within work groups to the verticle formal channels that ideally flow across functions. Most of our current technology and expertise is devoted to working with the verticle flow of information.

For information systems work the next graphic relates the key categories of the Rummler/Brache model to recent work in IS/IT design and knowledge management.



What is required to begin the development of a design strategy as illustrated above is a commitment to seeing the whole organization and doing the necessary work in the performance realm taking full measure of the informal work culture and the complexity of situated work. The key features of this effort are the work place research components and the management structures to support the inclusion of this informal but highly meaningful knowledge into the strategic and process functions of the organization. Work of this sort will bear fruit if the leadership in the organization is prepared to make the long term commitment to develop a holistic management and process model for the firm. Eventually all the work of the organization will be done in reference to the cybernetic and virtual information and knowledge flows that this model suggests.

References

Agre, P. (1997). Computation and human experience. Cambridge, UK: Cambridge University Press.
Bannon, L. J. (1997). Dwelling in the "great divide": The case for HCI and CSCW. In G. C. Bowker, S. Leigh-Starr, W. Turner & Gasser Les (Eds.), Social Science, Technical Systems, and Cooperative Work: Beyond the great divide (pp. 355-379). Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers.
Eraut, M. (1994). Developing professional knowledge and competence. London: The Falmer Press.
Goguen, J. A. (1997). Toward a social, ethical, theory of information. In G. C. Bowker, S. Leigh-Starr, W. Turner & Gasser Les (Eds.), Social Science, Technical Systems, and Cooperative Work: Beyond the great divide (pp. 27-56). Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers.
Hall, D., & Moss, J. (1997). Helping organiztions and employees adapt. Organizational Dynamics, 26(3), 311-332.
Jensen, M. C., & Meckling, W. H. (1996). Specific and general knowledge, and organizational structure. In P. S. Myers (Ed.), Knowledge management and organizational design (pp. 17-39). Boston: Butterworth-Heinemann.
Lave Jean. (1988). Cognition in practice. Boston, MA: Cambridge University Press.
Maturana, H., & Varela, F. (1987). The tree of knowledge; The biological roots of human understanding. Boston, MA: Shambhala.
Nonaka, I. (1997). A new organizational structure. In L. Prusak (Ed.), Knowledge in Organizations (pp. 99-135). Boston: Butterworth-Heinemann.
Piaget, J. (1971). Biology and knowledge. Chicago: University of Chicago Press.
Project Report - CSCW Symposium. (1996). Systems development and co-operative work: Methods and techniques. WWW, http://comp.lancs.ac.uk/computing/research/cseg/projects/ambleside.html.
Rummler, G. A., & Brache, A. P. (1995 2nd Ed.). Improving performance: How to manage the white space on the organization chart. San Francisco: Jossey-Bass Publishers.
Schein, E. (1997). Three cultures of management: The key to organizational learning in the 21st century. WWW, http://learning.mit.edu/res/wp/three.html. Cambridge, MA: MIT Sloan School of Management.
Senge, P. (1990). The fifth discipline. New York: Doubleday.

Turning organizational and management theory into practice seems to result in a consistent misunderstanding relating to accountability, responsibility, and the location (or co-location) of power in the organization. The way this misunderstanding often plays itself out is in the disconnect between management and the regular employee. It is often the case that management spends a lot of time and energy working on ideas, models, and plans at the organizational or process levels only to find them unaccountably shipwrecked on the rocks of daily practice and performance. This brief essay will explore this persistent difficulty and offer a possible solution in the context of an overarching organizational design framework.
The overall framework referred to here is the application of a three part model relating organizational, process, and performance phenomena. While much of the language used in articulating this model comes from the important work of Rummler & Brache (1995) the underlying logic and validity of the model shares a long history in work devoted to organizational design, the social sciences, cognitive science, and biological theories of learning and growth.
It also needs to be noted that the analysis of problems related to 'workplace buy-in' of management initiatives as explored here assume an existing and ongoing effort at the organizational and process levels that aims to clarify goals, identify key processes, and pay attention to the management of the same (this implies the existence of feedback loops built into the overall design process at the organizational and process levels). Absent this kind of effort (including the appropriate managerial disposition) any efforts to clarify the key relationships at the performance level as they affect the actual realization of idealized processes will be in vain.

That Old Paradox
One of the most difficult issues for organizational theorists, managers and executives to understand is the paradox of the bottom up and/or the top down organization. While this seeming paradox may be usefully addressed at the managerial level and above through the use of teams, retreats, frequent communication, increased awareness and knowledge, variable compensation, etc. it becomes quite a bit more difficult to craft relationships and mechanisms that connect the same two way linkage at the performance or practice level.

Example #1
For example, if you are in a firm of professionals (say an engineering consulting firm) where the entire product of your efforts is the result of the coordinated work of people just like you (i.e., professional engineers) it is relatively easy to coordinate bottom up processes (the actual work of engineering) with top down fiduciary and strategic goal setting. In this case we find that the owners or executives in the engineering consulting firm share so much in common with their performers that communication and feedback, buy-in and cooperation, and the host of other culture related issues are more easily met.

Most members share a similar language with similar meanings attributed to similar experiences. These people also share similar activity patterns, decision making practices, information sources, and educational experiences. The commonality of these qualities cannot be underestimated for it is in the similarity of these types of qualities that alignment to goals, coordination of practice, and similarity of interpretations are found.
The only real limitation to proactive change and the road to a quality organization in this example is the imagination and leadership ability of the firm along with the normal constraints of a competitive and dynamic marketplace.

To the extent that this hypothetical engineering firm has a large support staff they will probably be afflicted with misalignment and 'buy-in' problems roughly proportional to the size of the support staff, the criticality of its work, and the longevity of its members. Certainly the efficiencies of information technology and automation in general have reduced the overall proportion of these types of positions and allowed firms like this a bit more flexibility in designing, implementing, and managing their strategic change initiatives.

Example #2
The scenario described above, however, is not applicable or relevant to many firms and organizations. The typical manufacturing, service, or infrastructure-building organization is composed of many different cultures and ways of seeing the world. The typical organization is stratified along lines familiar to us all with relatively few people at the top in positions of control and authority over increasingly specialized and functionally defined processes below. This stratification is determined by both knowledge and power. Typically there is a higher level of general and specialized education the higher up in an organization you go. There is also an increase in power as determined by fiduciary responsibility and/or control and competence.
These firms are generally based on a top-down theory of management and supported in this by a tradition and general business infrastructure that strongly supports it. It is also true that firms like this are under increasing pressure to perform in ways similar to the less stratified and more responsive firms as illustrated by the first example. This is where we run into the paradox problem.

Resolving the Paradox
In using the firm of our first example as an ongoing point of comparison we see that the most significant fact affecting its overall success is in the similarity of work cultures. This similarity allows for information and decision making to flow quickly and effectively through the organization.

The problem for the traditional organization is to develop a common work culture that binds and links those at the top with those at the bottom. This is the challenge of organizational design and development.

Obviously an assumption implicit in this analysis and proposal is both the belief and empirical claim that it is in some sense possible to 'engineer' cultural alignment. It is also assumed that before the performance level is engaged in the process of change there has to be a significant degree of alignment at the managerial level and above.

As indicated above the solutions proposed here rest on the assumption that the organizational and process levels of management are somewhat aligned and on the road to a greater shared understanding of the entire organization. If this is the case then we can focus on the relationship between the practice or performance level with its various constituents. Those constituents being the levels above as well as external relationships with vendors, contractors, customers, and other information providers.

Approaching Performance Through Processes
Generally speaking we are what we do. We are known through our behavior. This principal is central to industrial psychology, human factors analysis, and management science in general. It is through our daily practice and performance that our work is done and that our lives take on meaning. Consequently it is through an understanding of our practice that we are able to change it. This is a key point. If we don't know what we are doing we will not be able to do what we want to do (or what we are told to do). In many ways our daily life and habits are opaque to us. We act and then are surprised if we see ourselves act (think, for example, how strange it is to see yourself on video-tape or hear your recorded voice).

The reason that the consulting engineer firm in example#1 above is successful is because most members unconsciously share the same mental frameworks and very similar ongoing reinforcing experiences. If their world is disturbed or shocked by an external force (say a major recession) they are generally able to quickly become conscious of their practice to make the necessary changes to adapt to the new circumstances.

However, in a typical organization we have multiple cultures (or multiple groups with multiple mental frameworks) acting in mostly unconscious ways in their daily practice or performance. It is also generally the case that anything that shocks or perturbs these shared mental frameworks is instantly perceived as an external force threatening the status quo.

It is in the interpretation of the external perturbation that we can see the consequences of interventions of the work groups or cultures that we are working with. In the case of the consulting engineers the perturbations are interpreted in roughly the same way. This allows the system as a whole to adjust and integrate the disturbance in a widely similar fashion.
For the work culture that is a part of the whole but acts and interprets its existence as if it were the whole we have a problem. The problem is that this sub-group acting like a whole group reacts to external stimulus in mostly unconscious ways and tends to act in a manner suggesting an avoidance of the forces causing the disturbance. This avoidance will occur along different channels but for similar reasons. For example, as the grapevine (informal information system) transmits the nature and perceived causes of the shock or disturbance there come into play historical justifications for acting that may or may not be actually related to the disturbance at hand. This behavior is seen by those external to the work culture as a sort of pathology that then becomes the focus of the change effort. The problem is that when the externally perceived pathology is targeted by the change agents (or management) the work culture responds to disturbances here in much that same way as they did with the initial disturbance. That is, the perturbing factor and its effects are displaced and dealt with more or less automatically through previously experienced reactions.

These reactions are mostly unconscious and this general process can go on ad infinitum. So that each time the focus of the external force changes - the work culture's reaction changes so as to displace it from the actual practice that is going on. And this really must be the case for if we stopped our activity every time an external forces disturbed our system properties we would never be allowed to perform our activities with the seamless and effortless wonder that we generally perform them.

A good example here is to notice that when we are driving a car we are most effective when we are not consciously reacting to external factors that disturb or potentially disturb the functioning of our driving skills. The myriad of 'things' going on while we are driving is truly amazing. Yet we bracket most of these things and put them into the background to be dealt with unconsciously and in historically proven effective ways. For example, the way that we deal with pedestrians or bicyclists is interesting. After we become relatively skilled at driving we only pay attention to them if they exceed some threshold of aberrant behavior. When we are reminded or 'perturbed' that there is a potential problem with a pedestrian or cyclist we generally act quickly and decisively to extricate ourselves from the situation so as to resume or automatic and efficient driving behavior. I might add that we are generally annoyed when we have to make these adjustments as well! This example equates the driver with her set of skills to the subculture in the organization and the pedestrian or cyclist to the person or group creating or suggesting organizational and/or behavioral change. Becoming conscious and aware of these dynamics is important in the development of skilled behavior.

The Practice of Self-Awareness
The ability to coordinate complex social behaviors rests on some fundamental assumptions of human behavior. Chief among those assumptions is trust. Trust is a comprehensive category that includes a belief that those around me are accountable for their behavior (with accountability understood as the result of acting in good faith on the basis of good feedback to strengthen existing networks that support our common enterprise). When there is trust there is also - generally - honesty. And honesty used in this instance means an ability to see and interpret one's own behavior using the same standards and criteria used when seeing and understanding another's behavior. Simply put - it is the practice of the Golden Rule.
The problem, then, is how we can develop in our organization of organizations the practice of honest self-awareness. The first step is simply to listen and hear what people in the work culture we are focused on are saying. This involves a strategy of active listening at the level of a work culture. As indicated above the ability to engage in active listening presumes a level of trust as well as a degree of sophistication on the part of the listener as to the importance and value of the information that is being heard. There is nothing so discouraging than answering someone's question when you know that the asker is not interested in your answer. This is probably the single most important element of leadership as exercised on a regular basis - that is - that the person in a position of authority has to be genuinely and skillfully interested, engaged, and open to the people that he or she is in authority over. If the person is not so engaged the people instantly and automatically know it. This creates the absolute requirement that those in authority have to walk the talk - have to model the behavior that they are interested in fostering in their subordinates. We also know that this is probably the most difficulty practice leaders engage in – if it were easy certainly more leaders would practice it!

The First Element in the Performance Strategy
Assuming that the local leadership is so disposed the first element in the strategy to align the performance culture with the other organizational pieces is carried out using a methodology similar to cultural ethnography. Ethnography is a method of collecting, analyzing, interpreting, and evaluating information from the culture itself. This information is collected through observation, interview, focus-groups, surveys, and a collection of artifacts typifying the specific culture. The transformative part of the process entails the culture itself becoming the chief interpreter and evaluator of their own cultural information.

There are two important functions embedded in the use of this tool. First of all the surfacing of the work culture's actual practice will occur during the process and this information will form the basis for the second element - the process mapping phase - which is discussed below. Second, the process of going through an ethnographic analysis opens the work culture to a conversation and practice of looking at its own functions. This is the self-awareness part. It is important to understand that the primary purpose of the workplace ethnography is to inform the participants themselves of their shared meanings and to become primary material for the development of the process mapping phase. A secondary purpose is to share the final ethnographic document with leadership at both the process and organizational levels.
The provision of this information to the process and organizational level is extremely significant in that it provides vital feedback to decision makers as to the possibilities of change and the disposition of their most significant resource (the job performers). This feedback is presented in narrative form (with some matrix analysis) in order to further connect the story telling culture at the executive and management levels with the story telling culture at the performance or practice level. It is important to not undervalue this narrative information. Embedded in these narratives are the existing metaphors, symbols, and mental frameworks that characterize the organization's daily practice. A wise leader is one that understands this system of cultural meanings.

Allowing the ethnographic phase to occur in the manner uniquely determined by the work culture, however, is difficult for managers and executives using traditional management models to function. The process can be perceived as involving a loss of control, be seen as subversive to organizational goals, or be too painful or disturbing to the organization over all. It is important to have the right people in place to assist in the process to ensure sensitivity to political realities, historical hot spots, and general institutional norms. It is also critical, as argued above, that this process occur in a context of already developing change at the process and organizational levels.

The Second Element
After the investigation into the work culture is completed (finished narrative(s)) an appropriate team(s) structure is devised within which to micro-analyze work processes and job designs. This activity is the nexus around which organizational and process level management initiatives meet with actual workplace performance. The process/work design teams form both the forum and the future shape of how performance will change to ensure quality outcomes. It is at this point that the work of the process and organizational level teams is articulated to the process/work design teams.

Those fundamental goals, standards, and process parameters developed and understood by the managerial and executive functions are now to be animated by performance level teams. The artifacts provided to these process/work design teams should include organizational relationship maps, strategic plans, and a description of the management plan envisioned to guide the development of increasingly rational and transparent business processes.
What the ethnographic process produces for the process/work design teams to build from is an analysis of job design that specifies values and beliefs, a description of actual work activities, an analysis of how and by whom decisions on the ground are really made (specification of decision criteria), and a listing and analysis of information sources utilized in carrying out the performance/activities.

These two data sets are then made to fit through the facilitated process mapping exercise.
This element of making these data sets fit is the critical activity for the entire project. It is this activity that resolves the paradox of top down and bottom up practice.

General References:
Brown Seely, J., & Duguid, P. (1999). The social life of information. Boston: Harvard Business School Press.

Darrah, C., N. (1996). Learning and work: An exploration in industrial ethnography. New York: Garland Publishing, Inc.

Lave Jean. (1988). Cognition in practice. Boston, MA: Cambridge University Press.

Maturana, H., & Varela, F. (1987). The tree of knowledge; The biological roots of human understanding. Boston, MA: Shambhala.

Nardi, B., A., & O'Day, V., L. (1999). Information Ecologies. Cambridge: MIT Press.

Nonaka, I. (1997). A new organizational structure. In L. Prusak (Ed.), Knowledge in Organizations (pp. 99-135). Boston: Butterworth-Heinemann.

Piaget, J. (1971). Biology and knowledge. Chicago: University of Chicago Press.

Rummler, G. A., & Brache, A. P. (1995 2nd Ed.). Improving performance: How to manage the white space on the organization chart. San Francisco: Jossey-Bass Publishers.

Schein, E. (1997). Three cultures of management: The key to organizational learning in the 21st century. WWW, http://learning.mit.edu/res/wp/three.html. Cambridge, MA: MIT Sloan School of Management.

Activity theory and social practice. (1999) (S. Chaiklin, M. Hedegaard, & U. J. Jensen, Eds.). Aarhus, Denmark: Aarhaus University Press.

Strauss, A. L. (1987). Qualitative analysis for social scientists. Cambridge UK: Cambridge University Press.

Strauss, A. L. (1993). Continual permutations of action. New York: Aldine De Gruyter.

Wenger, E. (1998). Communities of practice; Learning, meaning, and identity. Cambridge: Cambridge University Press.