Design is the method of putting form and content together. Design,
just as art, has multiple definitions; there is no single definition. Design
can be art. Design can be aesthetics. Design is so simple; that’s why it is so
complicated.
Paul Rand[1]
I undertook a project in the spring of 1994. I had been
thinking about the future of work, lot’s of people had including many people
that I was interacting with at MIT. I started a series of discussions with Bill
Mitchell[3],
Tom Malone[4],
Joel Moses[5]
& Peter Rowe[6]
that led me to begin thinking about a new model for knowledge work[7]
(this is a lot of footnotes already, even for me…). The model I discussed with
Bill Mitchell & others cast knowledge work more as a design process &
less as an industrial (?) process. This seemed
self-evident to me, even if it didn’t to some other people. What did I (&
still do) mean by this?
Studies of the evolution of work[8]
suggest that the future (knowledge) work paradigm will have several major
characteristics: a) work will primarily consist of the identification,
acquisition and manipulation of information and knowledge, b) the convergence
of information and communication technologies will provide the basis for this
work, & c) design process will provide the framework for both doing work as
well as planning & strategizing about it. This knowledge work will be
different than the work people now do, even current "knowledge work[9]".
In order to determine what types of systems and devices need to be developed to
facilitate this future knowledge work paradigm, we must first produce and
validate, to the extent possible, a description of this work.
First, though, more on evolution. I believe that the next
ten years will be partitioned into three broad categories of work evolution.
These are shown in this table.
Current – now
|
Near-future – 3-5 years
|
Post-convergence 8-12 years
|
Mainly content based
|
Content & process based
|
Knowledge & model based
|
Focus is: problem solving
|
Convergence i.e. process & teaming
|
Design model, i.e. episodic, contradictive,…
|
Basis is: content-based deduction, individual contribution
|
Process innovation, group decision making, mixed &
modal logics
|
Personal & group intuition, induction
|
Content driven
|
Group & technology driven
|
Intuition & interconnection driven
|
The categories can be described as follows:
- Content based (current)
o
Creation, management & use of content drives
this phase
o
Work is mainly done by individuals & small
(relatively isolated) teams
o
Focus is individual & small teams working to
solve business problems
o
Feature needs revolve around integration of
content management & collaboration with organization of projects/programs,
budgets etc. also important
o
Technology is primarily program based, i.e.
based on algorithms & deductive (formal) reasoning
o
Primary impediments & push-back focus on
inadequate technology for collaboration (information & process) &
inadequate organization for collaboration as well as strong commitment to
co-location
o
Early adopters
have already moved on to the next phase
- Process based (3-5 years)
o
Deep integration of Content & Process drives
this phase
o
Work is done by distributed teams
o
Focus is on collaborative problem solving in the
work process context
o
Will require cross-organizational workflows
o
Process innovation & group decision making
become important
o
Technology enables work process &
collaboration, algorithms become less formal, more based on people’s work
process
o
Advances made in both how people work with (“Big
C”) content & collaboration as well as context-based problem solving enable
a move to the next phase
o
Primary impediments & push-back focus on
attribution for work & position of individual contribution
- Knowledge & model based (8-12 years)
o
Large-scale use of knowledge & models drive
this phase
o
Work, i.e.
what people do, takes on characteristics of the design work model
o
Work is done by a range of individuals,
collaborative teams & collaborations among teams, location not an
impediment
o
Focus is on knowledge & context-based
problem solving to address business issues
o
Technology provides means of creating &
maintaining context (organizational memory), large-scale knowledge modeling
capability, problem solving assistance & ability to deal with very
large-scale (Big C) content, use of ultra-large data sets for predictive
analysis
o
Models of work & business problems create
context for problem solving & business decision making
o
Primary impediments & push-back focus on
intuition & contradiction based reasoning rather than standard deduction
& induction
Before I move on to a description of a design work model,
I’d like to focus on the impediments to this evolution for a minute. In the
content-based phase, impediments to this evolution are mainly centered on the
opinion that technology is inadequate to support the level of collaboration
necessary. This is already inaccurate as technology today more than adequately
supports most forms of collaboration. The more serious impediments, that are
relevant to the next phase as well, are that organizational forms are not
adequate to support collaboration & that a preference for individual
contribution still exists in many organizational cultures. Many organizations,
even those in technology companies, have not made the changes in composition
&/or leadership needed to facilitate larger-scale collaboration that may
have essential elements that are not co-located. These changes include:
- Distributed & decentralized management & decision making
- Formation of semi-autonomous to independent project & program teams
- Nonhierarchical communication structures
- Nonhierarchical reward structures
Most organizations still adhere to either a push hierarchy
(standard command & control structure) or a pull hierarchy[10] (centralized management & decision-making
with management pulling opinions, strategy & tactics from lower levels in
the hierarchy), & these types of organizations have considerable difficulty
in making the changes required for real collaboration.
In addition, there continues to be a bias towards individual
contribution, especially in highly technical areas. The assumption is that a
single person (or sometimes a single small group) is the only one with the
knowledge or experience to create a design, architecture or program that will
be effective & productive. The problem with this bias is that it is
sometimes true &, at least in my experience, often much more productive
than current egalitarian collaborative models.
Another issue that becomes more relevant in the
process-based phase is that of attribution. Often group & individual reward
structures, including the opportunity to continue to work on strategic &
interesting problems, is tied to attribution. This is an issue if multiple
groups can legitimately claim responsibility for the success of a project.
Often the solution is to reward “everyone”. This can be less appropriate &
much less appreciated by the people that actually affected the success of the
project. There are organizational structures that may ameliorate this issue,
but the cultural issues are harder to address.
Finally in the knowledge & model-based phase, the
impediments are largely due to the real differences in how work is
accomplished. People, except for some types of scientists & engineers, are
not (yet) accommodated to working from models or from deep knowledge bases. In
addition, during this phase, the decision-making norm will also be quite
different – more biased towards guided intuition & non-deductive reasoning.
The use of extremely large amounts of data for predictive analysis is also not
yet well understood or accepted except in some specific & limited segments.
It will take time for these norms, & for the design work model to take
hold, therefore the 8-12 year timeframe.
There are a number of examples of knowledge work extant that
provide a good start for the description of work & work process in the
knowledge & model phase. I believe the best example of this is the general
design process. Design is a knowledge-based component of much work, but in the
future, most knowledge work will have the characteristics of design work. Rowe
(1987), Norman (1990, 1992) and others have described design work that can be
summarized as follows:
- episodic - context is preserved across a number of temporally distinct work periods during which different viewpoints or potential resolutions may be explored
- knowledge-based - deep and broad historical, technical and contextual knowledge is necessary
- eclectic – uses a wide range of problem solving techniques
- contradictions - fueled and/or motivated by the maintenance and resolution of tensions or contradictions including:
- ill defined/well defined task definition
- underconstrained/overconstrained
- intuitive/nonintuitive
- textual/nontextual
- "logical/nonlogical"
- theoretical/pragmatic
- content/context
- structure/function (organization/technology)
- time/completeness
- complexity/simplicity
- strategic/tactical.
The recognition, maintenance and resolution of these
contradictions are the major characteristics of this new work paradigm. A
contradiction occurs when some statement or fact is asserted to be both true
and false. In the sense used here, a contradiction is a true work condition (s)
or fact(s) whose assertion mutually excludes another true work condition or
fact. The presence of unresolved contradictions creates tensions in the work. These
tensions are present, to some extent, in our work today. The focus of our
current work, however, is technical contents. This is true regardless of what
the "technology" is that we are working with (VLSI, accounting
practice, manufacturing process, etc.). Our near future work will be
increasingly focused on process and teaming. Once the technical support exists
to support diverse process and teaming models in real-time, this will become
less of a focus. This should take 3-5 years for early adopters and perhaps as
much as 10 years to come into general use. Future knowledge work (5-10 years
out) will encompass these process and teaming models, but will more and more
come to be structured by the tensions and contradictions listed above.
What
does it mean to say that knowledge work will be based on our ability to
recognize, maintain and resolve contradictions and tensions? Recognition is
defined as the awareness that something perceived has been perceived before.
Maintenance is defined as the act of retaining or preserving some item or idea
and resolution is the act of providing a solution to a problem[11]
often through the resolution of one of the contradictions. These three
processes applied to the list of tensions represent the essence of future
knowledge work. This work will be characterized by: 1) the need to be aware
that the tensions and contradictions in both the context and content of the
work are necessary and that these contradictions provide the dynamic that
drives the work, 2) that part of the work may require maintaining these
contradictions and exploring the consequences of all potential possibilities,
and 3) that eventually a solution or set of solutions must be derived from this
tension.
The
formal presentation of the recognition, maintenance and resolution of logical
contradiction helps us understand the role of these tensions in the description
of a work paradigm. This formal presentation is outlined in Appendix A.
Let’s look at a near-future example of knowledge work in
light of this work process description. Natalia is a 28-year old employee of a
large public hospital. She has been tasked with determining the most prevalent
disease comorbidities in the hospital’s patient population, determining the
statistical characterization of the cost of treating those comorbidities &
making recommendations for reducing the cost of that treatment. This is an
issue in all healthcare organizations today that is only just at the edge of
our ability to analyze & make anything other than anecdotal recommendations
for. Here is a “design work process” description of this task:
- Episodic – This is not Natalia’s only work task, so she has to balance making progress on it with work on a set of other tasks. This means that she must take it up at intervals (in episodes) & that at the start of each episode, she must recover both the context & the content in the state that she left them & in a manner so that she can begin working on this task again with a minimum of redundant effort.
- Knowledge-based – Natalia will need a broad range of knowledge including: general medical knowledge, hospital clinical procedure, hospital financial procedure, operation of hospital clinical (electronic health record, practice management) software & financial management (cost accounting, billing) software, reporting software, analysis & business intelligence software, development, utilization & interpretation of financial models & other areas of knowledge just for this one task.
- Eclectic – Natalia will need to be able to access a variety of clinical & financial databases as well as create a financial model for determining revenue & cost associated with the prevalent comorbidities. She will need to be able to interpret the results of this modeling & translate those results (cost per comorbidity, revenue per comorbidity) into recommendations as well as be able to create a way to accurately & understandably present her results & recommendations.
- Structured & motivated by contradictions – Some of the contradictions that affect this task are:
o
The task definition is ill-defined. This causes
Natalia to have to interpret the intention of the task & how its results
& recommendations will be used. Her interpretation could not be in line
with management expectations & this tension requires Natalia to create a
specific definition for the task that she is working on a solution to.
o
The task is under-constrained. Again, this
causes Natalia to focus on developing a description of the boundaries of both
the task & the solution; i.e. “what
to leave in & what to leave out”.
o
Interpretation of the results & development
of recommendations is somewhat intuitive rather than empirical. This is related
to theoretical/pragmatic. Natalia must make recommendations that are feasible
& that in turn will lead to results in finite time. The tension here is
extreme as there are already many recommendations reported in the literature
& the public press in this area as well as many expectations of what the
“right” solution(s) may be. Natalia must be able to focus on interpretations
& recommendations that are relevant & feasible for her organization
& not based on general expectations or ideas in the public sphere.
o
The solution(s) may require structural
(organizational) changes as well as functional (partly technological) changes.
There is a tension in the balance between these that must be addressed.
o
Her manager has given Natalia two weeks
(elapsed) to accomplish this task. Natalia has already told her manager that
this is not enough time for anything other than a set of guesses, but there is
a big strategy meeting coming up &,… There is usually this tension between
time & completeness (that is often a proxy for accuracy/correctness).
o
As with most work tasks today, something that
seems simple can actually be quite complex. Just count all the patients that
have hypertension & figure out what it costs to treat them… simple. Not…
First of all what is a patient? What definition do you use for hypertension (or
any clinical condition)? What do you mean by cost? Does the cost accounting
system even have patients in it? The more you peel this back the more complex
it becomes. This means Natalia, once again has to be explicit about definitions
for every important term as well as for each calculation. This means she can at
least say that for this definition of these conditions & comorbidities
& for this definition of the calculation of costs, this is the median cost
in 2014 of treating this comorbidity for the 2,146 patients recorded with it.
o
This task can be interpreted at a tactical
level: what are the relatively small changes that can be made to our clinical
workflows & financial processes that will enable us to reduce costs? It can
also be interpreted at the strategic level: what are the organizational &
large-scale technology changes that can be made that would allow us to
systemically reduce costs? As with all of these contradictions, there is a
balance to be had here as well.
o
Finally, in healthcare there is a purely
strategic & almost unmentionable contradiction that is very well
illustrated by this task – that is cost reduction versus improvement in patient
outcome. These are not necessarily mutually exclusive, but the trade-offs made,
especially when resource constrained (think dollars) can make it seem as if
they are. Does reducing the cost of treatment on a per patient or per condition
basis improve or worsen patient outcomes. Certainly we do not want to have
worse outcomes at the “cost” of cost reduction, but it often seems as if this
is the choice. There is enough fat & waste in the healthcare system that
these do not have to be mutually exclusive, but this kind of general statement
does not work on a per patient basis. The tension here can be extreme.
Natalia’s work on this task does appear to be very aligned
with our description of design work process. Can we learn something from how
designers work to make her work more effective, or even more efficient? Also, I
hear you say, “what about strategy, I thought this was about strategy.” Right,
lessons learned first…
Several relatively recent sources have looked at these
lessons from different perspectives these include a 2007 report on how global
brands use the design process to create leverage & competitive advantage[12]
& several Harvard Business Review articles[13].
A primary lesson described in both reports is that design
must be tightly tied to & aligned with the end user experience – not the
technology used, general design trends or management expectations. This
requires both knowing who the end users will be & creating an experience
that is compelling & useful for them in both functional & emotional
ways. This is the case whether you are designing new communication devices or,
as Natalia is, designing a solution & recommendations for strategic
decision-making.
A second lesson is that models & prototypes are
essential in the design process, & that they do different things. Models
allow you to describe & examine the problem space; while prototypes allow
you explore the solution. Models & prototypes should be readily available
to people who are using or evaluating the work & feedback loops must be
part of the process so that criticism & suggestions from feedback can be
incorporated.
Design processes also tend to be optimal as collaborative
processes. Teams may be co-located or distributed, but that cross-functional
collaboration improves results & outcomes of projects. In Natalia’s case,
she is working as an individual contributor but may, in fact collaborate in
areas that she is not as experienced in such as development of financial
models.
Finally, several challenges have also been incorporated as
lessons:
- Larger amounts of ambiguity must be accepted by both end users & peers of the designer
- The same is true of risk
- Expectations must be set appropriately – design process is effective at reducing complexity, facilitating innovation & identifying future trends. It can also be effective at producing solutions to complex problems, as in Natalia’s case. It is not so good at actually operating a business. Even design companies like frog[14] & ideo[15] have business processes for program management, financial management etc.
OK, you say, but you promised strategy… There are two
aspects to this. The first is that a strategy is a designed object & needs
to be developed through a design process. The second is that design process
itself is a strategy that can be adopted by organizations. I’ll deal with both
of these.
As a designed object, an organization’s strategy needs to
have certain characteristics. The primary one of these is that it is focused on
a user end-state. What does the organization want to accomplish. What
experience is it trying to create for the users of its product or service?
Often organizational strategies, especially those of for-profit companies, are
specified with respect to how much growth they need to achieve, what industry
segments they’ll penetrate & by how much, how much revenue & profit
they’ll generate. These are not a strategy but rather the results of a strategy.
Strategy can be designed by collaboratively iterating descriptions &
tactics (prototyping) & continuously modifying the strategy according to
criticism & feedback. As we have seen, such a process must incorporate the
acceptance of higher levels of ambiguity & risk in order to succeed. Once
such a strategy is developed & decided upon, tactics & business
processes must be developed to initiate & evolve the strategy.
What we are talking about is using the design work process
to develop the organization’s strategy, so that the strategy is a designed object.
We can also see, through this effort, that the design work process itself can
be a strategy that is applied to many of the efforts & work tasks that an
organization undertakes. This was shown in Natalia’s work example above. Using
the design work process as a framework:
- Episodic & iterative approach so that multiple tasks are worked on at once
o
Use of models for problem description &
prototyping for solution exploration
- Classification & utilization of deep & large volumes of knowledge
o
Requires ways of searching, storing, analyzing
& utilizing such knowledge
- Use of a broad variety of modeling, prototyping & problem-solving methods
o
Provides the best possibility of interpreting
& presenting actionable results
- Allowing for the recognition, maintenance & resolution of the contradictions inherent in work efforts (as described above)
o
Recognition of contradictions allows for
important aspects of a work task or problem to be identified & described
o
Maintenance of such contradictions allows for
appropriate levels of ambiguity & risk to be maintained so that solutions
are not tracked into & collapsed prematurely, &
o
Resolution of such contradictions allows
solutions to be derived
These general characteristics outline a framework for
addressing work tasks, up to & including strategy development that
organizations can productively use. By looking at both the development of
strategy & the accomplishment of work tasks as design problems, people
(working in organizations or independently) can use a highly flexible model of work
process to provide leverage in design, development, modification &
measurement of work & work process. When I say “organizations will need
to…”, I actually am talking about the people in those organizations who will
need to be more tolerant of ambiguity & risk, understand that their work
will be accomplished interatively & in episodes rather than in a single
effort, be able to work with large amounts of information & knowledge &
apply a broad set of problem-solving tools, applications & ways of thinking
to creating solutions. Organizations that encourage & facilitate this
approach to work & strategy, & that are made up of people that can
utilize the design work model will be very well positioned to compete
successfully as how we do our work, & even what our work is evolves over
the next 5, 8, 10 years.
[1] Professor of Graphic
Design, Yale University, 1956-1969, 1974-1985.
[2] M.C. Escher. Drawing Hands,
1948
[3] 1944-2010, Former Dean of
the School of Architecture & Planning, MIT
[4] currently Patrick J.
McGovern Professor of Management, Sloan School of Management, MIT
[5] Former Provost & Dean of
Engineering, currently Institute Professor, MIT
[6] formerly Dean of the
Graduate School of Design & currently Harvard University Distinguished
Professor, Harvard University
[7] Knowledge work can be differentiated
from other forms of work by its emphasis on "non-routine" problem solving that requires a combination of convergent, divergent, and creative thinking. https://en.wikipedia.org/wiki/Knowledge_worker
[8] Amazon.com lists 16,547
books under this topic ranging from The Critique of Pure Reason. Immanuel Kant.
1781, to The Future of Work – Human Value in a Digital World. Marcus Clarke.
February, 2015. Accessed 18 August 2015.
[9] Knowledge work can be
differentiated from other forms of work by its emphasis on
"non-routine" problem solving that requires a combination of convergent, divergent, and creative
thinking. https://en.wikipedia.org/wiki/Knowledge_worker.
Accessed 15 August 2015.
[10] D. Lavoy. 2014. Is
collaboration limited by organizational structure. http://www.cmswire.com/cms/social-business/is-collaboration-limited-by-organizational-structure-024450.php
CMS Newswire, March 2014. accessed 10 September 2015.
[11] all definitions are from
The American Heritage Dictionary of the English Language, 3rd Edition. 1992
[12] Eleven Lessons – Managing
the Design Process in Global Brands: A study of the Design Process. Design
Council. 11/2007.
[13] Design for Action. T. Brown
& R. Martin. Harvard Business Review. pp. 56-65
Design Thinking Comes of Age. Jon Kolko.
Harvard Business Review. pp. 66-71. 11/2015.
[14] www.frogdesign.com/
[15] www.ideo.com/
