Anda di halaman 1dari 25

Strategic Management Journal, Vol.

11, 171-1 95 (1990)

L THE DESIGN SCHOOL: RECONSIDERING THE BASIC


PREMISES OF STRATEGIC MANAGEMENT
\ HENRY MINTZBERG
Faculty of Management, McGill University, Montreal, Quebec, Canada

Among the schools of thought on strategy formation, one in particular underlies h o s t all
prescription in the field. Referred to as the ‘design school’, it proposes a simple model that
views the process as one of design to achieve an essential fit between external threat and
opportunity and internal distinctive competence. A number of premises underlie this model:
that the process should be one of consciously controlled thought, spec$cally by the chief
executive; that the model must be kept simple and informal; that the strategier produced
should be unique, explicit, and simple; and that these strategies should appear fully
formulated before they are implemented. This paper discusses and then critiques this model,
focusing in particular on the problems of the conscious assessment of strengths and
weaknesses, of the need to make strategies explicit, and of the separation between formulation
and implementation. In so doing, it calls into question some of the most deep-seated beliefs
in the field of strategic management, including its favorite method of pedagogy.

The literature that can be subsumed under are labeled the entrepreneurial school (concerned
‘strategy formation’ is vast, diverse and, since with strategy formation as a visionary process),
1980, has been growing at an astonishing rate. the cognitive school (a mental process), the
There has been a general tendency to date it learning school (an emergent process), and the
back to the mid-l960s, although some important environmental school (a passive process). A final
publications precede that date, such as Newman’s school, also descriptive, but integrative and
initial piece ‘to show the nature and importance labeled configurational, by seeking to delineate
of strategy’ (p. iii) in the 1951 edition of his the stages and sequences of the process, helps
textbook Administrative Action (1951: 110-118). to place the findings of these other schools in
Of course the literature on military strategy goes context.
back much further, in the case of Sun Tzu This paper addresses itself to the first of these
probably to the fourth century B.C. (Griffith, in schools, in some ways the most entrenched of
Sun Tzu, 1971: ix). the ten. Its basic framework underlies almost all
A good deal of this literature naturally divides prescription in this field and, accordingly, has
itself into distinct schools of thought. In another had enormous impact on how strategy and the
publication (Mintzberg, 1989), this author has strategy-making process are conceived in practice
identified ten of these. Three are prescriptive in as well as in education and research. Hence our
orientation, treating strategy formation as a discussion, and especially critique, of this school
process of conceptual design, of formal planning, can in some ways be taken as a commentary on
and of analytical positioning (the latter including the currently popular beliefs in the field of
much of the research on the content of competitive strategic management in general. Our intention,
strategies). Six other schools deal with specific however, is not to dismiss so important a school
aspects of the process in a descriptive way, and of thought, but rather to understand it better
014>2095/90/030 171-25$12.50 Received 14 Deciimber 1987
0 1990 by John Wiley & Sons, Ltd. Revised 22 May 1989
172 H. Mintzberg
and so place it into its natural context, and appeared in 1965 (by Learned, Christensen,
thereby open up thinking in the field in general. Andrews, and Guth) and quickly became the
This paper probes first into the basic model of dominant textbook in the field, as well as the
the design school, then into the basic premises dominant voice for this school of thought.*
that underlie it. That leads to a critique of this Certainly its text portion, attributed in the various
school, which gives rise to an attempt to place editions to co-author Kenneth Andrews (who
it into its own viable context-the types of also published this material separately (Andrews,
organizations and of situations most suited to it. l971,1980a, 1987)) stands as the most outspoken
In conducting this investigation we draw widely and one of the clearest statements of this school,
on the literature of this school, but use one text although claims that this school, or even the
in particular-almost certainly this school’s best concept of business strategy itself, originated with
known. In this sense this paper can also be this group at Harvard (e.g. Bower, 1986: vii) do
viewed as a rather extensive review of a book not stand up to scrutiny.
that has had a major impact on the field of Some of the basic concepts that underlie the
strategic management. design school, at least as published, would (as
we shall see) appear to have been first stated in
the academic world by a Berkeley sociologist
ORIGINS OF THE DESIGN SCHOOL named Philip Selznick, in his book Leadership
in Administration, published in 1957. Even
Ostensibly the simplest and most fundamental earlier, though less specific, is a 1955 article by
view of strategy formation is as a process of Reilley, possibly the first reflection of this
informal conception-the use of a few essential approach. Another key publication in 1962,
concepts to design ‘grand strategy.’ Of these Strategy and Structure, by historian Alfred Chand-
concepts the most essential is that of congruence ler (then at MIT), really established this school’s
or match. In the words of the design school’s concept of business strategy and its relationship
best-known proponents: ‘Economic strategy will to structure, although mention also has been made
be seen as the match between qualification of the sophisticated discussion of ‘Managerial
and opportunity that positions a firm in its Strategies’ in David G. Moore’s paper of that
environment’ (Christensen, Andrews, Bower, title in 1959. There followed an article by
Hamermesh, and Porter, 1982: 164)l ‘Capture Seymour Tilles in 1963 (then a Harvard Business
success’ seems to be the motto of the design School lecturer) entitled ‘How to evaluate corpor-
school; ‘find out what you are good at and match ate strategy’, and a textbook chapter by William
it with what the world wants and needs.’ These Newman of the Columbia University Business
capabilities or qualifications have been variously School in the same year (see especially 1963:
referred to as ‘distinctive competence,’ ‘differen- 95-98; the passage noted earlier in the 1951
tial,’ ‘competitive,’ or ‘comparative advantage’ edition of the Newman textbook might make him
(the latter more commonly used in the context the real father of the concept of business strategy
of public policy), or more simply (and broadly) in academe, although in private correspondence
an organization’s ‘strengths and weaknesses.’ with this author, Newman has expressed the
The design school has generally been associated belief that the overall ideas may have originated
with the Business Policy group at the Harvard in the McKinsey consulting practice, as reflected
Business School. That group has pursued its own in the Reilley piece of 1955 (see also McKinsey,
strategy for, as we shall see later, there is a 1932, for early suggestions of this thinking)). The
clear congruence between the view of strategy Andrews text followed in 1965, the same year
formation that it has promoted for several decades that Igor Ansoff published his highly successful
and its own pedagogical requirements in using book Corporate Strategy, based on many of the
the case study method. same concepts (but more in the spirit of the
The Christensen et al. book quoted from above, planning ~ c h o o l ) . ~
entitled Business Policy: Text and Cases, first
* Undoubtedly encouraged by the fact that in the early years
this group trained by far the largest number of doctoral
‘ Thus Lindgren and Spangberg (1981: 26) refer to this as students in business policy.
the ‘fit school.’ Porter (1981: 610; 1983: 173), a co-author in the 1982 and
The Design School 173

Subsequently these ideas embedded themselves THE BASIC DESIGN SCHOOL MODEL
in the management literature. Indeed, by the
1980s the Christensen et al. textbook was one of In his 1957 book, Selznick wrote that:
the few left that represented them in their pure
form, most others favoring the more elaborated Leadership sets goals, but in doing so takes
renditions of the planning or positioning school^.^ account of the conditions that have already
determined what the organization can do and to
Accordingly, and given the impact that this some extent what it must do . . .
rendition of the design school has had over the
In defining the mission of the organization,
years-as well as its clarity and forcefulness of leaders must take account of (1) the internal
expression-we shall use it as a primary source state of the policy: the strivings, inhibitions, and
in the discussion that follows, referring to it as competences that exist within the organization,
the ‘Andrews’ text’. We shall draw primarily on and (2) the external expectarions that determine
the 1982 edition of this textbook, but shall also what must be sought or achieved if the institution
is to survive (pp. 62, 67-68).
reference relevant variations in its earlier editions
as well as the latest one, published in 1987,
Selznick also coined the term ‘distinctive compe-
although the changes from 1965 to 1987 were
tence’ (pp. 43ff.) and noted that ‘the task of
relatively minor.
leadership is not only to make policy but to build
’ Continued. it into the organization’s social structure’ (pp.
1987 editions of the Harvard textbook, writes of how the 62-63), an aspect of the process that came to be
ideas in the original text (the ‘LCAG paradigm,‘ after the
names of the four original authors) were ‘subsequent[ly]’ called implementation.
translated and extended by others, citing in particular Ansoffs Andrews summarizes the essence of his model5
book Corporate Strategy. In fact, Ansoff went to press with as
his similar ideas in the same year (1965) as Porter’s co-
authors’ originally did, and neither book references any work
by authors of the other (although Edmund Learned, the the intellectual processes of ascertaining what a
senior author of the first edition of the Harvard textbook, company might do in terms of environmental
did himself note the similarities in a book published with opportunity, of deciding what it can do in terms
Sproat one year later: ‘Significantly, [Ansoff’s] work offers of ability and power, and of bringing these two
numerous parallels with Harvard thinking that should not be considerations together in optimal equilibrium.
obscured by differences in terminology, definitions, emphasis, . . . what the executives of a company want fo
and coverage‘ [ 1966: 941). In the Preface to the first edition do must also be brought into the strategic
of the Harvard book. the authors write that the content of decision [as must] what a company should do.
the book ‘is the outcome of about ten years of case and
course development’ (Learned el 01.. 1965: vii). although in
the 1982 version they refer to the core idea having developed Finally, there is ‘the implementation of strategy
in the early 1960s (Christensen, Andrews, and Bower, 1982:
viii; co-author Bower is more precise in a 1986 publication:
. . . comprised of a series of subactivities which
‘The problem of corporate strategy was first phrased as a are primarily administrative’ (p. 98). The
research question in 1959 when Kenneth Andrews reported Andrews’ text of 1982 splits into two ‘books,‘
his study of the Swiss watch industry in a note and a series the first on ‘determining’, the second on ‘imple-
of cases’ (p. vi).) Ansoff published a rough version of his
approach in article form two years earlier (Ansoff. 1964), menting corporate strategy.‘
although he referred there to an initial unpublished paper of Our depiction of the basic design school
1958. Note should also be made of comments by Chester model (similar to Andrews’ own figure of the
Barnard in a 1948 book (p. 169) which seem to be in the
spirit of the design school, of the discussion of ‘administrative
development of ‘economic strategy’ (p. 187), but
strategy’ (pp. 10-18) in the Hardwick and Landuyt textbook with other elements of his discussion added (see
by that title in 1961, and of an article by Gilmore also his figure on p. 99)), is shown here in Figure
and Brandenburg in 1962 entitled ‘Anatomy of corporate
planning’. Although its detail and elaborated steps place this
1. Consistent with the attention accorded in the
last paper clearly in the planning school. underlying these text, the model places primary emphasis on the
steps is the same model as that of the design school. (Gilmore appraisals of the external and internal situations,
and Brandenburg note in a footnote that ‘we are indebted the former uncovering threats and opportunities
to Dr. H. Igor Ansoff for introducing the concept of synergy
to us and for his assistance in clarifying a number of steps in the environment, the latter revealing strengths
in our planning framework’ (1962: 61).) and weaknesses of the organization. Secondary
4 T o this could be added Tregoe and Zimmerman‘s book emphasis is placed on understanding the values
Top Management Struregy (1980). although not a textbook.
The latest Newman text (Newman, Warren, and McGill. of the management, as well as its social responsi-
1987) remains largely in the spirit of this school (in chapter
4 at least). although it also reflects increased attention to We should point out that Andrews himself rejects the word
the planning school. ’model’ (p. 12). a point we shall return to later
174 H . Mintzberg

\ Strategy I

A implementation

Figure. 1. Basic design school model

bilities. The match between these elements leads in the 1982 edition.6 ikewise, the section on
to the creation of strategies, which are then internal appraisal, ‘identifying corporate com-
evaluated, with the chosen one subsequently petences and resources,’ is brief, touching on a
implemented. Andrews does not provide exten- variety of points, such as the difficulty ‘for
sive discussion of any of these issues (the whole organizations as well as for individuals to know
text portion of the 1982 book numbers 114 pages, themselves’ (p. 183) and the idea that ‘individual
the rest of the 838-page book being cases), and unsupported flashes of strength are not as
although others have developed some of these dependable as the gradually accumulated product-
themes more extensively. and market-related fruits of experience’ (p. 185).
On external appraisal, Andrews’ section on This ties back to an important theme in Selznick’s
‘The nature of the company’s environment’ totals
20 pages, but 12 of these come from Michael The sentences immediately preceding and following this
new material (pp. 167 and 179) are identical to those that
Porter’s work on Competitive Strategy (1980), appeared next to each other in the previous edition of the
literally spliced into the Andrews’ text, initially book (Christensen et al., 1978: 251).
The Design School 175

book, that ‘commitments to ways of acting and strategy’ (1963). Andrews (pp. 105-108) followed
responding are built into the organization,’ by combining Tilles’s list of six criteria with other
intrinsic to its very ‘character’ (1957: 67). elements of the model, while Rumelt, a graduate
Figure 1 shows two other factors considered of the Harvard doctoral program in policy,
by this school to be taken into consideration elaborated these ideas in a most succinct (1980)
in strategy-making. These are organizational and sophisticated (1979) way, nevertheless
values-the beliefs and preferences of those retaining the spirit of the design school.
who formally lead the organization-and social Finally, virtually all of the literature of this
responsibilities-specifically the ethics of the school makes clear that, once strategy is designed
society in which the organization is embedded, and agreed upon, it is then implemented. We
at least as perceived by its managers. With the show implementation on the diagram as flaring
notable exception of Selznick (1957), however, out from formulation, as if the process draws on
most authors in this school accord values and a variety of data to narrow down to convergent
ethics secondary attention. Andrews offers his choice before it diverges again to ensure
two brief chapters well after he has developed implementation across the entire organization.
the framework dealing with external and internal Andrews, for example, is clear on the subordinate
appraisals. role of these elements (e.g. p. 543). Interestingly,
On the actual generation of strategies, little is here also is the one place where he becomes rather
written in this school, besides an emphasis on specific. He lists 12 steps in the implementation
this being a ‘creative act,’ to quote Andrews process (backed up by a fair amount of text), a
(1982: 186). Indeed, if this were true, what list that seems to encompass any aspect of the
more could be said, short of trying to use cognitive strategy process not considered in formulation.’
psychology to probe inside the strategist’s mind. The same tends to be true of other design school
In lieu of describing the process, however, a publications as well.
number of writers associated with this school do
try to characterize the result, in particular seeking
to distinguish some core or dominant element of PREMISES UNDERLYING THE DESIGN
the strategy (e.g. Tregoe and Zimmerman, 1980: SCHOOL
43 and Ohmae, 1982). This has an important
implication, because it replaces strategies within Running through all of the literature that we
srruregy. In effect, rather than considering an identified with this school are a number of
organization’s intentions as a set of distinct, if fundamental premises about the process of
coupled, strategies (as tends to be done in the strategy formation. Some of these tend to be
planning and even the positioning schools), it explicit, others implicit, but they are always
treats them as an integrated concept. evident. This is especially so in the Andrews’
Once strategies are created, the next step in text, although all are at one time or another
the model is to evaluate them and choose the qualified in his discussion. But it is the central
best one. The assumption, in other words- themes of a work that form the impression left
usually implicit-is that several alternative strat- with the reader, not the secondary qualifications.
egies have been produced, and one is to be Below we discuss seven basic premises that
selected. There is an ambiguity here, however, underlie this school.
because even writers such as Andrews, who
clearly view strategy formation as a custom-made Premise 1: Strategy formation should be a
process of design (for which there is evidence controlled, conscious process of thought
that organizations tend to produce only a single
solution; Mintzberg, Raisinghani and Theoret, It is not action that receives the greatest attention
1976). assume that alternate strategies (in other from the design school so much as reason-
words, alternate conceptions of the business) will strategies formed through a tightly controlled
be evaluated to select a single one (Andrews’ process of conscious human thought. Action
text, pp. 105, 109). ’ In their book Implementing Srrategy . Hrehiniak and Joyce
Tilles published first on this subject in an indeed refer to implementation as ‘all the remaining
article entitled ‘How to evaluate corporate components’ (1984: 29).
176 H . Mintzberg

follows, once the strategies have been fully strategy may suddenly be rationalized to mean
formulated. This theme runs through all of something very different from what was originally
Andrews’ writings, for example in the comment intended because of the opportunism which at
that managers ‘know what they are really doing’ the beginning of this book we declared the
only it they make strategy as ‘deliberate as conceptual enemy of strategy’ (pp. 828-829).8
possible’ (1981: 24), or more simply, in reference
to his ‘thesis’ about ‘conscious strategy’ that Premise 2: Responsibility for that control and
should be ‘consciously implemented’ (Andrews’ consciousness must rest with the chief executive
text, p. 543). But this is perhaps made most clear officer: that person is THE strategist
in his comments that, while the model may be
simple, it is not necessarily natural (p. 185)-it To the design school, ultimately there is only
must be learned, formally (e.g. Andrews’ text p. one strategist, and that is the manager who sits
6). at the apex of the organizational hierarchy.
Andrews is careful to position his view of this In Hayes’ terms, ‘this “command-and-control”
process clear of intuition on one side (non- mentality allocates all major decisions to top
conscious thought) and emergent strategy on management, which imposes them on the organi-
the other (where action drives reflection). On zation and monitors them through elaborate
intuition, for example, Andrews comments that planning, budgeting, and control systems’ (1985:
‘if [strategy] is implicit in the intuition of a strong 117).
leader, the organization is likely to be weak and Again, the origins of this can be found in
the demands the strategy makes upon it are likely Selznick: ‘it is the function of the leader-
to remain unmet’ (pp. 105-106). Likewise, he statesman-whether of a nation or a private
writes of the need to change ‘intuitive skill’ into association-to define the ends of group exist-
‘conscious skill’ (p. 6). And as for his view of ence, to design an enterprise distinctively adapted
emergent strategy, which means pattern in action to these ends, and to see that that design becomes
over time that is not driven by central intention a living reality’ (1957: 37). Once more Andrews
(Mintzberg, 1978; Mintzberg and Waters, 1985)’ reiterates the point most clearly. On page 3 of
while Andrews presents a definition of strategy his text, he associates the whole field with the
in his 1982 text that makes reference to pattern ‘point of view’ of the ‘chief executive or general
(p. 93), in a new passage in his 1987 text, he manager’; on page 19, he entitles a section ‘the
makes clear that he means pattern among and president as architect of organizational purpose’
across ‘goals and policies,’ not over time (p. 15; (hence Zand (1981: 125) refers to this school as
see also Premise 5 ) . the ‘rational architect’ model); and on page 545
In this edition, as in all the others, Andrews he writes that ‘the general manager is principally
clearly means to associate strategy with intention- concerned with determining and monitoring the
ality. Corporate strategy, for example, ‘defines adequacy of strategy, with adapting the firm to
products and markets-and determines the com- changes in its enviroment, and with securing and
pany’s course into the almost indefinite future’ developing the people needed to carry out the
(1987: xi). In fact, in his 1982 text, Andrews strategy or to help with its constructive revision
equates emergent strategy with ‘erosion:’ or evolution.’ As we shall soon discuss, in the
1987 text Andrews widens the participation
Strategy will evolve over time, no matter what. of others in the strategy formulation process,
It will be affected by the consequences of its especially in the ‘innovative’corporation, but not
implementation. But the elucidation of goals can at the expense of the chief executive’s central
transcend incrementalism to make it a series of role.
forays and experiments evaluated continuously
against stated goals to result in the deliberate It might be noted that this premise not only
amendment of strategy or in the curtailment of relegates other members of the organization to
strategic erosion (pp. 553-554)
* Andrews’ words are reminiscent of those of Selznick: ‘When
institutional leadership fails, it is perhaps more often by
Likewise, Andrews contrasts ‘purpose’ with default than by positive error or sin. Leadership is lacking
when it is needed; and the institution drifts, exposed to
‘improvisation,’ ‘planned progress’ with ‘drifting’ vagrant pressures, readily influenced by short-run opportun-
(p. 20). At the end of his text he claims that ‘a istic trends’ (1957: 25).
The Design School 177

subordinate roles in strategy formation, but it vengeance, the other, formal analysis of certain
also precludes external actors from the process of its components).
altogether (except for the directors, who Andrews We have already noted Andrews’ stand on
believes must review strategy (1980b, 1981)). intuition; on planning he writes that ‘this book
This is most clearly reflected in Andrews’ . . . virtually ignores the mechanisms of planning
discussion of the ethics discussion in terms of the on the grounds that, detached from strategy,
social responsibility of the managers rather than they miss their mark’ (p.
the sheer power of outsiders. This, in fact, is Of course, if elaboration is the problem, then
just one aspect of a larger problem associated even theory and research can pose a threat.
with the design school-the relegation of the Thus, Andrews adopts a position in the text that
environment to a minor role, input to strategy is not just atheoretical but decidedly anti-theory .
formation but not an intrinsic part of the process, For example, all of the research on organization
to be accounted for and then navigated through structure is dismissed with the comment that ‘the
but not interacted with. literature of organization theory is by itself . . .
of very little use in managing a live orgnaization’
(p. 554).
Premise 3: The model of strategy formation
The introduction to the first edition of the
must be kept simple and informal
textbook contained the following comment:
The Preface to the Harvard textbook contains a
quotation by Andrews that ‘the idea of corporate A considerable body of literature purporting to
make general statements about policy-making is
strategy constitutes a simple practitioner’s theory, in existence. It generally reflects either the
a kind of Everyman’s conceptual scheme’ (p. unsystematically reported experience of individ-
14). Later, he adds in the text that this ‘is not a uals or the logical projection to general manage-
“theory” attended in the rigorous sense by ment of concepts taken from engineering,
elegance and vigor’, nor is it ‘really a “model,” economics, psychology, sociology, or mathemat-
tcs. Neither suffices (Learned et al., 1965: 6).
for the relationships designated by the concept
are not quantifiable;’ rather it serves as an
‘informing idea’ (p. 12), or, as Rumelt puts it, On the latter Andrews added: ‘The disciplines
‘a set of constructs’ (1984: 558). cited have much to do with business, but their
Fundamental to (what we nonetheless prefer purposes are not ours. Knowledge generated for
to call) the model is the belief that elaboration one set of ends is not readily applicable to
and formalizatih will sap it of its essence. This another’ (p. 6). The text went on to note that
premise in fact goes with the last: one way to ‘research has been for some time under way, but
ensure that strategy can be controlled in one is not yet advanced enough to make more than
mind is to keep the process simple. As Andrews a modest claim on our attention. . . . the most
writes: ‘When the variety of what must be known valid literature for our purpose is not that of
cannot be reduced by a sharply focused strategy general statements but case studies’ (p. 6). These
to the capacity of a single mind and when the comments survived virtually intact to the 1982
range of a company’s activities spans many edition, the most significant change being that
industries and technologies, the problems of research now ‘begins to make a claim on
formulating a coherent strategy begin to get out our attention’ (p. 6). Moreover, added is the
statement that ‘the books referred to [in the
of hand’ (p. 182).
footnotes] comprise a relevant but incidental
This premise, together with the first, forces
source of knowledge’. (It is, in fact, instructive
Andrews to tread a fine line throughout his text,
between nonconscious intuition on one side and to consider these references. In all. there are 39
of them to theoretical works in the footnotes of
formal analysis on the other, a position he seems
the 1982 edition of the text, of which 31 are by
to characterize as ‘an act of judgment’ (p. 108).
This also seems to differentiate clearly the design 9 Interestingly. in so dismissing at this point.
school from the entrepreneurial school on one Andrews resurrects intuition: ‘All the knowledge. professional
side and the other prescriptive schools of planning attitudes and analytical and administrative skills in the world
cannot fully replace the intuitive genius of some of the
and positioning On the Other (One emphasizing natural entrepreneurs you will encounter in this book‘ (p.
elaboration of the same basic model, with a 10).
178 H . Mintzberg

faculty members or doctoral students at the As a result of this premise, the design
Harvard Business School.)’(’ school says little about the content of strategies
It might be noted that this treatment of theory themselves, but instead concentrates on the
extends even to the work of a co-author whose process by which they should be developed. And
ideas are contained in the same book. As noted that process above all should be a ‘creative act’
earlier, Michael Porter’s views (centrally located (Andrews’ text, p. 186), to build on distinctive
in the positioning school) were literally spliced competence. Writing in support of the positioning
into that portion of the text on assessing the school, Hofer and Schendel refer to what we are
environment. The text on either side of it, calling the design school as the ‘situational
however, questions assumptions fundamental to philosophy’ (1978: 203), at one extreme of
Porter’s work (1980, 1985). To take just one the field, in contrast with the ‘principles of
obvious example: management’ approach at the other.
The choice of objectives and the formulation of
policy to guide action in the attainment of
objectives depend upon many variables unique Premise 5: Strategies emerge from this design
to a given organization and situation. It is not process fully formulated
possible to make useful generalizations about
the nature of these variables or to classify their As noted in passages cited above, this school
possible combinationsin all situations (Andrews’ offers little room to incrementalist views or
text, p. 5). emergent strategies. It is the big picture that
Thus the Porter graft does not take, and, more results from the process-the grand strategy, an
importantly, other theory has not been allowed overall concept of the business. This is no
to infiltrate the model. Darwinian view of strategy formation but the
Biblical version, with strategy the final concep-
tion! There is, in other words, a strong implication
Premise 4 Strategies should be unique: the best that strategy as perspective appears at a point in
ones result from a process of creative design time, fully formulated, ready to be implemented.
As suggested above, it is the specific situation How else could Andrews have assumed that the
that matters, not any system of general variables. process reduces to ‘choice,’ a word he uses often,
It therefore follows that strategies have to be referring also in the Preface to the 1987 edition
tailored to the individual case: ‘In each company, to ‘chis decision,’ ‘the strategic decision,’ ‘the
the way in which distinctive competence, organi- entrepreneurial decision . . . [once] identified’
zational resources, and organizational values are (p. xiv, italics added). In other words, the
combined is or should be unique’ (Andrews’ text, assumption is that the strategist is able to line
p. 187). Stronger words are offered on page 109: up alternative strategies before him to be
‘sometimes the companies of an industry run like evaluated so that one can be definitively chosen.
sheep all in one direction’ although imitation
‘does not constitute the assurance of soundness.’
Premise 6: These strategies should be explicit
‘OFor the record. these are all the references found in the
and, if possible, articulated, which also favors
text portion of the book, as well as the Preface. References their being kept simple
to cases, or references within the cases themselves, were not
included. References were counted rather than sources, so While Andrews accepts various reasons for not
that in a few cases the same source was referenced more articulating strategy (such as confidentiality or
than once. A source was considered to emanate from Harvard difficulty of updating (see pp. 96-97)), he clearly
if at least one author was on the staff or was a doctoral
student there. Lest this criticism be extended unfairly to all views these as necessary evils. In common with
of the co-authors, or even the claims about the literature virtually all the writers of this school, he believes
itself, it should be noted that Edward P. Learned, the senior that strategies should at least be explicit to those
author of the original edition of the textbook, in his book
published together with Sproat in 1966, and entitled who make them and, if at all possible, articulated
Organization Theory and Policy, contained perhaps half the so that others in the organization can understand
amount of text, yet twice the number of references, only a them: ‘The unstated strategy cannot be tested or
small proportion of those emanating from the Harvard
Business School. The 1987 edition of the Andrews’ book contested and is likely therefore to be weak . . . .
contains 24 such references, 18 from Harvard. A strategy must be explicit to be effective and
The Design School 179

specific enough to require some action and appropriate structure,’ writes Andrews (p. 551).
exclude others’ (pp. 105-106). as if the existing structure does not bear on the
If strategies are to be so articulated, it also new strategy.
follows that they have to be kept rather simple:
to the point, easily stated to be easily understood.
Andrews’ qualifications
‘Simplicity is the essence of good art,’ writes
Andrews, ‘a conception of strategy brings sim- While these seven premises are clearly evident
plicity to complex organizations’ (p. 554). To in Andrews’ text, as noted earlier he does qualify
him, strategy helps organizations make better virtually all of them, tucking into his text here
decisions ‘by reducing the world of detail to be and there either nuances that soften their
considered to those central aspects of external character or else comments that acknowledge
environment and internal resources that affect the unfortunate reality, as compared with the
the company and bear on the definition of its preferred ideal. A number of them have also
business’ (p. 835). been added to the 1987 text, for example:
Of particular interest to Andrews is the role
of outside directors in strategy formation: he False hope, oversimplification, and naivete, as
well as zest for power, have often led . . . to
believes that they must be actively involved at the assumption that the chief executive officer
least in the evaluation and review processes. But conceives strategy single-mindedly, talks the
this can only happen if strategies are explicit, so board of directors into pro forma approval.
that they can be articulated to the directors, and announces it as fixed policy, and expects it to
be promptly executed by subordinates under
simple, so that they can be understood by people conventional command and control procedures
who have only brief time to devote to the (P. 82).
organization: ‘The power of strategy as a simplify-
ing concept enabling independent directors to Andrews rejects this view for all but ‘the
know the business (in a sense) without being in entrepreneurial startup stage,’ while we see it as
the business will one day be more widely tested a not unreasonable caricature of his own text!
at board level’ (p. 834). In his 1982 text Andrews wrote that ‘strategy
formulation is itself a process of organization,
rather than the masterly conception of a single
Premise 7: Finally, only after these unique, full-
mind’ (p. 827), in other words, at least in
blown, explicit, and simple strategies are fully
‘technically or otherwise complex organizations,’
formulated can they then be implemented
‘an activity widely shared in the hierarchy of
We have already noted the sharp distinction management’ (p. 828). In 1987 he even contrasted
this school makes between the formulation of ‘constructive engagement’ with ‘archaic notions
strategies on one hand and their implementation of authority , responsibility, hierarchy, stat us, and
on the other. Consistent with classical notions or centralized decision making ’ (p. 86). Yet his
rationality-diagnosis, prescription, then action- major justification for this seemed to be the
the design school clearly separates thinking from generation of commitment to the strategy emanat-
acting (see Bourgeois and Brodwin on their ing from the apex of the organization’s hierarchy.
‘change model.’ 1984: 246). (It is of interest that ‘Commitment . . . is a simple reason for the
the word used is ‘implement,‘ not ‘achieve,’ the involvement of whatever number of people is
assumption being that given proper implemen- required to make a success of whatever is
tation, achievement is a foregone conclusion.) intended’ (p. 120; see also pp. 55-56, 59).
Central to this distinction is the associated Andrews also accepted that ‘in real life the
premise that structure must follow strategy. processes of formulation and implementation are
As Andrews puts it: ‘Corporate strategy must intertwined’ (exist in a ‘reciprocal relationship‘
dominate the design of organizational structure (1987: 853)), that ‘the formulation of strategy is
and processes’ (p. 543). The assumption seems not finished when implementation begins’ even
to be that each time a new strategy is formulated, though the cases in the book have been arranged
the state of structure and everything else organi- around these two topics ‘for the sake of orderly
zational is to be considered anew: ‘Until we know presentation’ (1982 text, p. 541). He also acknowl-
the strategy we cannot begin to specify the edges that ‘we should look first at the logical
180 H . Mintzberg
proposition that structure should follow strategy themselves as, of course, is natural if he is not
in order to cope with the organizational reality to undermine his own position.
that strategy also follows structure’ (p. 99), in The fact is that the premises of the design school
other words, that ‘the structure and processes in combine to form their own tightly integrated
place will in fact affect the strategy’ (p. 552). strategy-the whole thing really is a ‘model’ after
In the 1982 text Andrews acknowledges emer- all. By ultimately remaining true to its premises,
gent strategy as well (e.g. p. 553), going further Andrews positions the design school in its own
in the 1987 text to discuss ‘a balance between niche, distinguishing it particularly from the
focus and flexibility, between a sense of direction planning and positioning schools on one side,
and responsiveness to changing opportunities which by elaborating the model shift it from the
. . , . Corporate strategy need not be a realm of judgement to that of analysis, and the
straitjacket. Room for variation, extension, entrepreneurial school on the other, which by
and innovation must be provided’ (p. 84). Yet mystifying the whole process locks it into the
he is careful to avoid association with what he inaccessible (and unteachable) realm of intuition.
calls extreme incrementalism, understoo -Ias The outstanding question is how large is that
reactive improvisation, muddling througli , or niche: how much of the viable strategic behavior
following one’s nose’ (p. 83). And elsewhere in of organizations, whether for purposes of descrip-
the 1987 edition, new sections also make clear tion or prescription, is it reasonably able to
the continuing commitment to deliberateness: ‘It encompass?
is not only possible but also essential to plot a
course into a future that cannot be foretold . . .’
(p. xiii, see also the comments on Japanese CRITIQUE OF THE DESIGN SCHOOL
management on pp. vi-vii).
Adding all these qualifications together, one The writings of the design school can be critiqued
can easily come up with quite a different model on a number of levels. In perhaps the most
of strategy formation. But of course, no reader general sense, the school has denied itself the
can doubt for which model Andrews stands, and chance to adapt. Research results that have put
not just ‘for the sake of orderly presentation’!” parts of it under suspicion were not considered;
One obvious question that arises from a number indeed, there was no reason to, if the model
of these qualifications is why practice seems to could not be elaborated upon. This problem is
differ from the prescribed model, at least some well illustrated by the book by Norman Berg
of the time. Andrews does not address the (1984), also of Harvard, who provided, two
question-research of course does, but he pre- decades after the original Andrews’ text, almost
cludes the results of research from his text. Nor the same chapter headings (save an application
does he pursue these qualifications at any length. at the end to the divisionalized firm), with the
Most are presented as asides or afterthoughts, same points made in the same ways, including
while his real commitment remains to the premises even the same qualifications (see for example,
pp. 28-33).
I’ Or ‘temporary conceptual convenience’, as Andrews put As Andrews so keenly argued, the source of
it in a memo to his colleagues in response to comments this data and inspiration for the model was to be the
author made in a talk given at the Harvard Business School
in 1976: concrete case-the description of one firm in one
whatever our preferences, let us avoid the allegation that situation. Ironically, however, his beliefs about
the central conceptualization of Business Policy as a field theory kept him from using even this rich data
separates formulation from implementation for anything base to build better theory. Certainly after 1965,
except temporary conceptual convenience. The inter-
relationships of a complex interdependency cannot be if not before, if there was a relationship, then
intelligently discussed all at once. What is being related the model had to drive the writing of cases (for
can usefully be stopped and examined before reinstalling example, if you want to find out how strategy is
it ConcePt’JalIY in a dynamic Process (Andrews, 1976: 4).
made, go interview the strategist), not vice-versa,
In a personal reply to Andrews, I concluded that with respect since the model has barely changed since then.
to this dichotomy, ‘Aside from headings, we may be doing
the same thing. The question is: do headings matter?!’ We Of course, its supporters might ‘Ontend that
shall return to this later in our discussion. the model was good enough in 1965 and remains
The Design School 181

so today. We shall not promote this contention, to propose was a framework. Leaving aside the
however, its contribution has been profound, as ambiguities in Andrews’ own writings on this
we shall point out later, but it has never been point,12even leaving aside the fact that an author
good enough, indeed no one model can be. It must be interpreted by his central thrust rather
describes but one approach to strategy formation, than his secondary qualifications, in the most
and even that one sometimes exhibits a level of fundamental sense the two interpretations are
generality and a tone of inevitability that seems not really different. Both are underlaid by some
overly simple in places and, at times, dogmatic. powerful assumptions, a critique of which will
Indeed one sometimes wonders whether, like the underlie our own argument. These concern the
testament of a religious prophet, it comprises a central role of conscious thought in strategy
set of profound truths subtly buried in simple formation, that such thought must not only take
prescriptions, or else if ‘the whole idea is just precedence over action but must precede it in
one big fat platitude’ (as the Harvard authors, time, and correspondingly that the organization
to their credit, quote a critical company executive must separate the work of thinkers from that of
in the final case of their 1982 edition (p. 821)). doers. In our view, these assumptions often prove
In the Preface to this 1982 edition the authors false, both descriptively and prescriptively. In
wrote that ‘our teaching focus then [when the other words, often not only don’t organizations
core idea of the book was developed in the early do these things, but by all accounts, they should
1960~1,as today, emphasizes the determination not. This suggests that while the design school
of corporate strategy (Book One) and the framework, if not the model, may never go out
implementation of corporate strategy (Book of date, it can easily go out of context.
Two). This format has stood the test of time’ (p. We develop our critique by considering specific
viii). But it has not, even in their own school, aspects of the model-first, the belief about the
as attested to by recent difficulties and changes need for a conscious assessment of strengths and
in the Harvard MBA Policy course. That course weaknesses, then the assumed sequence of
still splits into two on these lines, but formulation strategy followed by structure, after that the
has been moved into the first year of the program, premise that all strategies should be made explicit,
and refashioned in the spirit of the positioning and finally the assumed dichotomy between
school as Porter articulates it, while the second formulation and implementation. We shall con-
half on implementation remains (at the time of clude the critique by considering the relationship
this writing) in a state of flux after several years between the design school model and case
of searching for a new formula. study teaching, before closing the paper with a
Strategy can locate a system in a niche, but in delineation of the contexts we believe to be most
so doing narrow its perspective. This is what appropriate for this school. The reader is asked to
seems to have happened to the design school bear in mind that although the other prescriptive
itself with respect to strategy formation. As schools of planning and positioning have broken
noted, the premises of the model deny certain with certain of the premises of the design school
important aspects of strategy formation, including (notably in keeping the process simple and
incrementalism and emergent strategy, the influ- strategies unique, to a lesser extent also in
ence of existing structure on strategy, and the introducing the planner and analyst into the
full participation of actors other than the chief process alongside the chief executive), the fact
executive. We wish to elaborate on these short- that they have accepted the most basic ones
comings in this critique, to indicate how they renders most of the following a critique of those
narrow its perspectives to certain contexts (as schools as well.
indeed, do the premises of most of the other
schools).
One point should be made before we probe
into the details. Andrews might well argue that I2Or even his support for the specific model: for example.

we are interpreting his writings too literally, that ‘the text is dispersed throughout the book so as to permit a
step-by-step consideration of what is involved in corporate
it is unfair to take apart a model-a specified strategy and in the subactivities required for its formulation
sequence of prescriptive steps-when all he meant and implementation‘ (p. 11; italics added).
182 H . Mintzberg

Assessment of strengths and weaknesses: under the title ‘Defining corporate strengths and
thinking vs. learning weaknesses’. Starting out with a conventional
design school view of these (see p. 53), Stevenson
Our critique of the design school revolves around asked managers to assess their companies’
one central theme: its promotion of thought strengths and weaknesses in general. Overall,
independent of action, strategy formation above ‘the results of the study brought into serious
all as a process of conception, rather than as one question the value of formal assessment
of learning. We can see this most clearly in a approaches.’ In general, ‘few members of manage-
fundamental step in the formulation process, the ment agreed precisely on the strengths and
assessment of strengths and weaknesses. weaknesses exhibited by their companies’ (p. 55).
How does an organization know its strengths The overall impression left by this study is
and weaknesses? On this, the design school that the detached assessment of strengths and
is quite clear-by consideration, assessment, weaknesses may be unreliable, all bound up with
judgement supported by analysis, in other words, aspirations, biases, and hopes. In fact, Stevenson’s
by conscious thought. One gets the image of managers seemed to understand the problem,
executives sitting around a table discussing the their ‘most common single complaint’ being that
strengths, weaknesses, and distinctive com- strengths and weaknesses have ‘to be defined in
petences of an organization, much as do students the context of a problem,’ or to quote one of
in a case study class. Having decided what these his subjects, ‘As I see it, the only real value
are, they are then ready to design strategies. in making an appraisal of the organization’s
Some writers offer specific lists of potential capabilities comes in the light of a specific deal-
strengths and weaknesses for all organizations, the rest of the time it is just an academic exercise’
while others, though not offering such lists, do (Pa 65).
assume that types of strengths and weaknesses Every strategic change involves some new
exist in general. Andrews, on the other hand, experience, a step into the unknown, the taking
would only associate strengths and weaknesses of some kind of risk. Therefore, no organization
with a particular organization-its competences can ever be sure in advance whether an established
are distinctive to itself. But does even that specify competence will prove to be a strength or a
them precisely enough? weakness. In its retail diversification efforts, a
In his article on ‘strategic capability’, Lenz supermarket chain we studied (Mintzberg and
(1980) critiques the use of an ‘organizational Waters, 1982) was surprised to learn that discount
frame of reference’-usually based on some stores, which seemed so compatible with its food
abstract ideal or a comparison with the situation store operations, did not work out well, while
of the past-with an external frame of reference. fast-food restaurants, ostensibly so different, did.
In other words, internal capability has to be The similarities of the discount store business-
assessed with respect to external context. But as how products were displayed, moved about
we have already mentioned, there is a tendency by customers, and checked out, etc.-were
in the design school to slight the environment in apparently overwhelmed by the subtle but differ-
favor of a focus on the organization itself (which ent characteristics of merchandising-styling,
may manifest itself in a tendency to overstate obsolescence, etc. On the other hand, the
strengths and under emphasize weaknesses (e.g. restaurants may have looked very different, but
Katz, 1970: 350; see also Dimma, 1985: 251). they moved simple, basic, perishable, commodity-
But the problem of assessing strengths and like products through an efficient chain of
weaknesses may go deeper still. Might com- distribution much like the supermarket business
petences not also be distinctive to time, even did. The point we wish to emphasize is: how
distinctive to application (Radosevich, 1974: 360; could the firm have known ahead of time? The
see also Hofer and Schendel, 1978: 14&150)? discovery of what business it was to be in could
And can any organization really be sure of its not be undertaken on paper, but had to benefit
strengths before it tests them, empirically? from the results of testing and experience. (And
The point we wish to make came out most the conclusion suggested from such experiences
clearly, if inadvertently, in a study carried out at is that strengths generally turn out to be
Harvard by Howard Stevenson (1976), published far narrower than expected and weaknesses,
The Design School 183

consequently, far broader (see Mintzberg and formation. No ongoing organization ever wipes
McHugh, 1985)). the slate clean when it changes its strategy. The
Nowhere does this come through more clearly past counts, just as does the environment, and
in practice than in attempts at related diversifi- the structure is a significant part of that past.
cation by acquisition. Obviously no organization Claiming that strategy must take precedence over
can undertake such activity without a prior structure amounts to claiming that strategy must
assessment of its strengths and weaknesses. Yet take precedence over the established capabilities
the vast majority of experiences reported in the of the organization, clearly an untenable prop-
popular press and published research suggests osition. By overemphasizing strategy, and the
that related diversification is above all a learning ability of the stategist to act rather freely, the
process, in which the acquiring firm has to make design school slights, not just the environment,
a number of mistakes until it gradually learns but also the organization itself. Structure may be
what works for it, if it ever does (see Miles, malleable, but it cannot be altered at will just
1982; also Quinn, 1980: 28). And in the writings because a leader has conceived a new strategy.
of academe, the problem is perhaps best illus- Many organizations have come to grief over just
trated by Levitt’s (1960) popular ‘marketing such a belief.
myopia’ conception, that firms should define We conclude, therefore, that structure follows
themselves by broad mission rather than narrow strategy as the left foot follows the right in
product or technology (e.g. transportation instead walking. In effect, strategy and structure both
of railroad). The idea was enticing, but in many support the organization. None takes precedence;
applications too easy, a cerebral exercise that each always precedes the other, and follows
could detach managers from the realities of the it, except when they move together, as the
businesses they managed. What, in a few words organization jumps to a new position. Strategy
on a piece of paper, would enable railroads to formation is an integrated system, not an arbitrary
fly airplanes? Levitt, a marketing professor but sequence.
here arguing in the spirit of the design school,
wrote that ‘once it genuinely thinks of its business Making strategy explicit: promoting inflexibility
as taking care of people’s transportation needs,
nothing can stop it from creating its own Once strategies have been created, via the
extravagantly profitable growth’ (p. 33; our conscious assessment of strengths and weaknesses
italics). Nothing except the limitations of its own among other things, then the model calls for
distinctive competences! their articulation. While recognizing some reasons
for not making strategy explicit, this school
Structure follows strategy . . . as the left foot generally considers an unwillingness to articulate
strategy as evidence of fuzzy thinking, or else of
follows the right
political motive. But there are other, often more
While the design school tends to promote the important, reasons not to articulate strategy,
dictum, first articulated by Chandler (1962), which strike at the basic assumptions of the
that structure should follow strategy and be design school.
determined by it, in fact its model also accepts the The reasons generally given for the need to
opposite. Since the assessment of organizational articulate strategy are, first, that only an explicit
strengths and weaknesses is an intrinsic part of sttategy can be discussed, investigated, and
the model, a basic inpur to strategy formulation, debated (e.g. Andrews, 1981: 24); second, that
and since structure is a key component of only by making strategy explicit can it serve its
this, housing the organization’s capabilities, then prime function of knitting people together to
structure must play a major role in determining ‘provide coherence to organizational action’
strategy too, by constraining and conditioning it (Rumelt, 1980: 380); and third, that an articulated
as well as guiding it. strategy can generate support-can rally the
While this may be an obvious point, hardly troops, so to speak, and reassure outside influ-
disputed even within the design school, it does encers. These all sound like excellent reasons for
have a broader implication, an important one in articulating strategy. And they are-so long as
our critique of this school’s model of strategy all the conditions are right. The most important
184 H. Mintzberg

of these is that the strategist is sure-knows Another reason not to articulate strategy is
where he or she wants to go, and has few serious that pronouncements of it, often necessarily
doubts about the viability of that direction. In superficial, can engender a false sense of under-
other words, the design school implicitly assumes standing. Andrews argues that ‘a conception of
conditions of stability or predictability. But strategy brings simplicity to complex organiz-
organizations have to cope with conditions of ations’ (p. 554). True enough. But at what price?
uncertainty too. How can Andrews’ company The potential danger of a little knowledge
come ‘to grips with a changing environment’ needs to be recognized: the possible trivialization
when its ‘strategy is [already] known’ (1981: 24)? and distortion of the subtle needs of a complex
And how can its managers promote the necessary organization. As Wrapp has noted, sometimes it
changes when its own board of directors uses is impossible to articulate direction ‘clearly
that articulated strategy ‘to prevent the company enough so that everyone in the organization
from straying off its strategic course”( 1980b: 32)? understands what they mean’ (1967: 95). And
Our point is that organizations must function the problems can magnify when outsiders are
not only with strategy, but also during periods involved in the process, even board members.
of the formulation and reformulation of strategy, Perhaps that is why Andrews finds such strong
which cannot happen instantaneously. ‘It is managerial resistance to the inclusion of outside
virtually impossible for a manager to orchestrate board members in strategy-making.
all internal decisions, external environmental To summarize, the problems of making strategy
events, behavioral and power relationships, tech- explicit essentially bring us back to the need to
nical and informational needs, and actions of view strategy formation as a learning process, at
intelligent opponents & that they come together least in some contexts. Sure strategies must often
at a precise moment’ (Quinn, 1978: 17). Indeed be made explicit, for purposes of investigation,
sometimes organizations also need to function coordination, and support. The questions are:
during periods of unpredictability,’ when they when? and how? and when not? There is
cannot possibly hope to articulate any viable undoubtedly a need for closure at certain points
strategy. The danger during such periods is not in an organization’s history, moments when the
the lack of explicit strategy but exactly the process of strategy formation must be suspended
opposite-‘premature closure,’ the reification of temporarily to articulate clear strategies. But this
speculative tendencies into firm commitments. need should not lead us to believe that it is
When strategists are not sure, they had better natural for strategies to appear fully developed
not articulate strategies, for all the reasons given all of a sudden, nor should it allow us to ignore
above. the periods during which strategies must evolve.
Moreover, even when it makes sense to
articulate strategies, because they appear to be
Separation of formulation from implementation:
viable well into the future, the dangers of doing
detaching thinking from acting
so must still be recognized. Explicit strategies,
as implied in the reasons for wanting them, are The formulation-implementation dichotomy is
blinders designed to focus direction and so to central to the design school-whether taken as
block out peripheral vision. Thus, they can a tight model or a loose framework. But is the
impede strategic change when it does become distinction a valid one for conceptual and
necessary: to put this another way, a danger in analytical, even pedagogical, purposes? In other
articulating strategy is that while strategists may words, should people concerned with strategy
be sure for now, they can never be sure forever. (including students learning about it) think, let
The more clearly articulated the strategy, the alone behave, in terms of formulation and
more deeply imbedded it becomes in the habits implementation?
of the organization as well as the minds of its How can anyone really question this distinction,
strategists. There is, in fact, evidence from the or even the assumption that formulation must
laboratories of cognitive psychology that the precede implementation? After all, this is just
explication of a strategy-ven having someone another version of the basic form of rationality
articulate what he or she is about to do anyway- that underlies western thinking-in its simplest
locks it in, breeding a resistance to later change form, that to act you must first know what you
(Kiesler, 1971). want to accomplish. Think first, then do.
The Design School 185

The organizational form that corresponds to In recent years there has been a spate of books
this dichotomy is the classical hierarchy, what we and articles on implementation (such as Hrebiniak
prefer to call ‘machine bureaucracy’ (Mintzberg, and Joyce, 1984). Noting that few intended
1979). It above all emphasizes the distinction strategies are successfully realized-the figure
between the few people on top who are allowed cited by Fortune writer Walter Kiechel (1984:
to think and the many below who are supposed 8) that ‘fewer than 10 percent of American
to act. Machine bureaucracy is common in mass corporations’ implement their intended strategies,
production and the mass provision of services. It was deemed ‘wildly inflated’ by Tom Peters!-
dominates thinking in the consulting profession they call for more attention to the implementation
(most of whose techniques promote this form of process. ‘Manage culture,’ executives are advised,
structure), in big business (outside of high or ‘pay more attention to your control systems.’
technology), and in government, including the If one side of the formulation-implementation
military. dichotomy does not work, then effort must be
In his article on the dysfunctions of traditional invested in the other.
military organization, Feld (1959), noted the Majone and Wildavsky point out that to study
sharp distinction that is made between the officers implementation is to raise ‘the most basic question
in the rear, who have the power to formulate about the relation between thought and action:
plans and direct their execution, and the troops How can ideas manifest themselves in a world
on the front, who, despite their first-hand of behavior?’ (1978: 103). As they characterize the
experience, can only implement the plans given ‘planning-and-control model of implementation,’
them. These ‘organizations place a higher value which sounds to us like the design school model
on the exercise of reason than on the acquisition in the public sector, ‘good implementation is the
of experience, and endow officers engaged in the irresistible unfolding of a tautology,’ or to
first activity with authority over those occupied translate their terms into ours, the transformation
by the second’ (p. 15). This ‘is based on the of intended strategy into realized strategy through
assumption that their position serves to keep a ‘suitable . . . “production function” ’, meaning
them informed about what is happening to the goals, plans and controls, ‘and-to take care of
army as a whole . . . [which] is supported by the the human side of the equation-incentives and
hierarchical structure of military organization indoctrination’ (p. 106); ‘the perfectly pre-formed
which establishes in specific detail the stages and policy idea . . . only requires execution, and the
the direction of the flow of information’ (p. 22). only problems it raises are ones of control’ (p.
This, in fact, is the assumption fundamental to 114).
the formulation-implementation dichotomy: that All that would be fine were only the world
data can be aggregated and transmitted up the cooperative. Unfortunately, often it is not, in
hierarchy without significant loss or distortion. It many cases for good reason, whether the resist-
is an assumption that fails often, destroying ance to the intended strategy comes from the
carefully formulated strategies in the process. environment in which it is to be implemented,
To use a quotation Feld (p. 15) meant for the the organization that is supposed to do the
military, in how many contemporary organi- implementing, or even from the strategy itself.
zations do ‘the conditions most favorable to Sometimes the ‘implementors’ who make up
rational activity, calm and detachment, stand in the rest of the organization are perfectly willing
direct antithesis to the confusion and involvement’ to proceed as directed from the center, but the
of the factory floor, the salesman’s call, the environment simply renders the strategy a failure.
government clerk’s service? In how many does It may change unpredictably, so that the intended
detached formulation render the organization strategy becomes useless, or it may remain so
ineffective? In how many is critical information unstable that no specific strategy can be useful.
ignored because it is deemed ‘tactical’? Speaking Despite implications to the contrary, the external
from Japan, Ohmae goes so far as to suggest environment is not some kind of pear to be
that ‘separation of muscle from brain may well plucked from the tree of external appraisal, but
be a root cause of the vicious cycle of the a major and sometimes unpredictable force to
decline in productivity and loss of international be reckoned with.
competitiveness in which U.S. industry seems to In other cases it is not the environment but
be caught’ (1982: 226). the implementors within the organization who
186 H. Mintzberg
resist. They may, of course, be narrow-minded first, that the formulator can be fully, or at
bureaucrats too wedded to their traditional ways least sufficiently, informed to formulate viable
to know a good new strategy when they see one, strategies, and second that the environment is
or small-minded ones who do not understand the sufficiently stable, or at least predictable, to
new strategy, or bloody-minded ones who prefer ensure that the strategies formulated will remain
to go their own way (e.g. Thoenig and Friedberg, viable after implementation. Under some con-
1976 and Scheff, 1961). But sometimes they are ditions at least, one or the other of these
right-minded people who do what they do to assumptions proves false.
serve the organization despite its leadership. In an unstable environment, or one too complex
They may resist implementation because they to be comprehended in a single brain, the
know the intended strategies to be unfeasible- dichotomy has to be collapsed, in one of two
that the organization will not be capable of ways. If the necessary information can be
realizing them or, once realized, they will fail in comprehended in one brain, but the environment
an unsuitable external environment. is unpredictable-or perhaps more commonly,
Implementational failure can also occur without takes time to figure out after an unexpected shift-
inhospitable environments and resistant organi- then the ‘formulator’ may have to ‘implement’him
zations. The problem can lie in the strategy itself; or herself. In other words, thinking and action
indeed, in part at least, it almost always does. must proceed in tandem, closely associated: the
For one thing, no intended strategy can ever thinker exercises close control over the actions.
be so precisely defined that it covers every The leader-here Andrews’ one strategist (or a
eventuality. Moreover, while the formulators may small group)-develops some preliminary ideas,
be few, the implementors are typically many, tries them out tentatively, modifies them, tries
functioning at different levels and in different again, and continues until a viable strategy
units and places (Rein and Rabinovitz, 1979: emerges, much as Quinn (1980) described the
327-328), each with their own values and process, or continues to act even if one does not.
interpretations. They are not robots, nor are the Such close control of a leader over both
systems that control them airtight. The inevitable formulation and implementation is characteristic
result is some slipping between formulation and of the entrepreneurial mode of strategy-making,
implementation. where power is highly centralized in a flexible
‘Slippage’ is a term used in the public sector organization (Mintzberg, 1973). But, as noted
to mean that strategic intentions get distorted on earlier, that mode, because it is rooted in the
their way to implementation; ‘drift’ is another vagaries of intuition, tends to be dismissed by
term used there for realized strategies that differ the design school. True, it may sometimes be
from intended ones, but within their context ‘opportunistic,’ as Andrews claims, but such
(Majone and Wildavsky, 1978: 105; Kress, opportunism can be necessary, perhaps in and
Koehler and Springer, 1980; Lipsky, 1978). Here, of itself or more productively perhaps, as a means
however, we would like to take a position beyond to experiment and learn. Pascale (1984) provides
both concepts. a marvelous example of the latter in his descrip-
Certainly much formulation is ill-conceived, tion of how the Honda Motor Company execu-
just as much implementation is badly executed. tives in the United States backed their way into
But often the fundamental difficulty lies not in their highly successful motorcycle strategy of the
either side, but in conceiving a distinction between 1960s, in contrast to the Boston Consulting
formulation and implementation in the first place. Group’s (1975) inference of that strategy as
Behind the premise of the formulation-- brilliantly deliberate. (In the 1987 textbook (p.
implementation dichotomy lies a set of very vi), Andrews comments that ‘Japanese manage-
ambitious assumptions: that environments can ment appears to be more truly strategic than
always be known, currently and for a period well improvisatory.’ Perhaps for him, as for the Boston
into the future. in one central place, at least by Consulting Group, believing is seeing.)
the capable strategists there. To state this more Where there is too much information to be
formally, by distinguishing formulation from comprehended in one brain-for example, in
implementation, the design school draws itself organizations dependent on a great deal of
into two questionable assumptions in particular: sophisticated expertise, as in high-technology
The Design School 187

firms, hospitals, and universities-then the strat- McHugh (1985) on a ‘grass roots model,’ and
egy may have to be worked out on a collective Mintzberg (1987) on ‘crafting strategy.’)
basis. Here, then, the dichotomy collapses in the And then, perhaps most common, are a whole
other direction: the ‘implementors’ become the range of possibilities in between-‘implemen-
‘formulators’ (Hardy et al., 1984; Mintzberg and tation as evolution,’ as Majone and Wildavsky
McHugh, 1985). As Lipsky puts it, implemen- (1978) put it-where there is thought, then there
tation is ‘turned on [its] head’ (1978: 397), and is action, this produces learning which alters
actions in good part determine thoughts, so that thought, followed by adjustments to action, and
strategies also emerge. so on. Intended strategies exist, but realized
Both situations-‘formulators’ implementing strategies have emergent as well as deliberate
and ‘implementors’ formulating-amount to the characteristics. Here words like ‘formulation’
same thing in one important respect: the organi- and ‘implementation’ should be used only with
zations are learning. Andrews’ great mistake was caution, as should the design school model of
dismissing organizational learning by considering strategy formation.
it opportunism. Even though he recognized the To conclude this critique, this seemingly
intertwining of formulation and implementation in innocent model-for Andrews, just an ’informing
practice, his making of the distinction conceptually idea’-in fact contains some ambitious assump-
led him to underestimate the important role tions about the capabilities of organizations and
of such learning, individually and especially their leaders, assumptions that break down in
collectively, over time, in strategy formation. whole or in good part under many common
More generally, the design school, by implicitly circumstances.
assuming that strategic learning somehow takes
place in one head for a limited period of time
and then stops, so that strategies can be articulated
and implementation can begin, denied processes THE DESIGN SCHOOL AND THE CASE
that have often proved critical to the creation of STUDY METHOD
novel and effective strategies.
Out of all this discussion comes a whole range We believe that the relationship between the
of possible relationships between thought and design school model of strategy formation and
action. There are times when thought does, and the traditional method of case study teaching
should, precede action and guide it primarily, may help to explain why there has been so much
so that, despite some inevitable slippage, the reluctance in certain quarters to adapt the model
dichotomy between formulation and implemen- to other views of strategy-making. The design
tation does hold up, more or less. In other words, school model matches perfectly the pedagogical
while it may be true that ‘literal implementation requirements of the case study method, as
is literally impossible’ (Majone and Wildavsky, Andrews and his colleagues note repeatedly. The
1978: 116), sometimes what is achieved is close students are handed a document of 20 or so
enough. And here is where we might expect pages that contains all the available information
viable application of the design school model. on the organization in question. They study it
Other times, however, especially during or the evening before class (alongside the other
immediately after a major unexpected shift in cases they must prepare for that day), and then
the environment, thought must be so bound up appear all ready to argue what it is that General
with action in an interactive and continuous Motors or the John F. Kennedy High School
process that ‘learning’ becomes a better label, should do.
and concept, for what happens then is Bear in mind that time is short: the external
‘formulation-implementation’. The organization environment must be assessed, distinctive com-
may be groping its way toward a new strategy, petences identified, alternate strategies proposed,
or may simply be coping until things settle down and these evaluated, all before class is dismissed
so that it can then do so. (For models of strategy- in 80 minutes. Two days later it’s on to Xerox
making as a learning process, see Quinn (1980) or Texas Instruments. Here is how the process
on ‘logical incrementalism,’ Burgelman (1983) is described by the senior author of the Harvard
on ‘corporate entrepreneurship,’ Mintzberg and textbook:
188 H. Mintzberg

how do those of us interested in management What effect has such case study teaching had
education strive to contribute to the development on practice, on the generations of managers who
of future general mangers? We do this first by have graduated from schools that rely on this
disciplined classroom drill with the concept of
strategy. Drill in the formal and analytic-what pedagogy? If that has left managers with the
is the current strategy of the firm? What are its impression that, to make strategy, they can remain
strengths and weaknesses? Where, in the firm’s in their offices with documents summarizing the
perceived industry, are profit and service oppor- situation and think-formulate so that others can
tunity? And, how can those corporate capacities implement-then it may well have done them
and industry opportunities be effectively related?
and their organizations a terrible disservice,
Moreover, this analytic classroom process encouraging superficial strategies that violate the
focuses attention on a key administration skill- distinctive competences of their organizations.
the process of selecting and ordering data so To quote Livingston (1971), a Harvard professor
that management asks the critical questions at the time himself, in his classic article ‘The
appropriate to a particular situation (Christensen,
in Christensen er al., 1982: ix-x). myth of the well-educated manager,’ the problem
of management education is its ‘secondhanded-
But how can a student who has read a short ness:’ ‘Managerial aspirants are lequired only to
resum6 of a company possibly know these things? explain and defend their reasoninr:, not to carry
How can words and numbers on paper possibly out their decisions or even to plan realistically
substitute for the intimate knowledge of a for their implementation;’ they ‘are rarely exposed
complex organization? Can the ‘critical questions’ to “real” people or to “live” cases,’ but rather
really be asked through the process of ‘selecting to ‘problems or opportunities discovered by
and ordering’ this kind of data? And what effect someone else, which they discuss, but do nothing
does this ‘drill in the formal and analytic’ have about.’ Thus, many ‘are not able to learn from
on the students when they finally do enter the their own firsthand experience . . . . Since they
executive suite? have not learned how to observe their environ-
Given the requirements of case study teaching, ment firsthand or to assess feedback from their
how else can the faculty proceed but to keep actions, they are poorly prepared to learn and
the model simple, especially to presume that grow as they gain experience’ (pp. 79, 83, 84,
organizations can be quickly and easily under- 89).
stood, and to assume the necessity for fully The fact is that the design school model
developed and explicit but nonetheless simple dominates not only the world of pedagogy, either
strategies. And even if it is accepted that in its pure form, or as the foundation of the
formulation and implementation are intertwined thinking behind the planning and positioning
in practice, what good is that in the classroom schools; it dominates beliefs in practice too. In
where formulation (thinking) is possible while other words, ‘one best way’ thinking is alive and
implementation (acting) is well in the practice of strategic management,
Of course, proponents of this school might and it dictates that formulation must precede
argue that this is a small price to pay for bringing implementation, that this formulation must be
reality into the classroom, enabling the students conscious and controlled, by the chief executive
to gain exposure to many different organizations. as the architect of strategy, and that the resulting
True enough. But need the reality-ven the strategies must be deliberate and explicit. Here
‘reality’ of the 20-page case-be dealt with in is how Robert McNamara, also formerly of
only this way? Is there not another option, which the Harvard Business School, spelled out his
is to open up the students’ perspective beyond approach to military strategy as Secretary of
the design school model, indeed even to use Defense: ‘We must first determine what our
cases themselves to do so, but written and taught foreign policy is to be, formulate a military
from a broader point of view? strategy to carry out that policy, then build
the military forces to successfully conduct this
l3 In his 1987 book, Andrews acknowledges that ‘How to strategy’ (quoted in Smalter and Ruggles, 1966:
get results is harder to teach and to learn in a classroom 70). He did just this in Vietnam, distant from
than on the scene. This difficulty may explain the neglect in
business education of the art of implementation in favor of the realities of the rice paddies and for too long
the analysis of potentially ideal strategies’ (p, ix). deaf to the calls to learn from the devastating
The Design School 189

results. O r consider the comment of one manager business’ (p. 834) might be more of a problem
about an earlier chief executive of the General than a solution. Can anyone, director or student,
Electric Company: ‘Borch had a sense that he even manager, really know an organization
wasn’t looking for lots of data on each business without being in it? The time of directors is
unit, but really wanted 15 terribly important and limited; they must be briefed through short
significant pages of data and analysis’ (quoted in documents and snappy presentations that articu-
Hamermesh, 1986: 191). As noted earlier, the late strategies clearly and simply, so that they
problem may be most acute in diversification by can be evaluated on the spot. Case study
acquisition, which often appear to have been discussions in the boardroom. But at what cost
undertaken by detached executives sitting up in strategic thinking? And strategic action?
in executive suites designing strategies quite Andrews claims that ‘graduates of a demanding
independently of any intimate understanding of Policy course feel at home in any management
the organization’s real strengths and weaknesses. situation and know at once how to begin to
This ‘one best way’ thinking applies also t o understand it’ (p. 6). But that may be the very
many of the consulting firms that specialize in essence of the problem. Mary Cunningham is a
this field-the so-called ‘strategy boutiques.’ graduate of the course Andrews had in mind.
Called in with limited knowledge of the industry She may not be typical, but her experience does
in question, and limited time to find out, the reveal the problem in its extreme. With a great
design school provides a most convenient model. deal of publicity, Cunningham leaped from
To quote from a popular book by two consultants: the Harvard MBA program to the personal
‘Four o r five working days over a two-month assistantship of William Agee, chief executive of
period are required to set strategy. Two or three the Bendix Corporation, himself a Harvard MBA.
working days are required for the review and one- Later she wrote a book on those experiences,
year update of strategy’ (Tregoe and Zimmerman, entitled Powerplay (Cunningham and Schumer,
1980: 120). There is not a lot of money t o be 1984). Kinsley published a scathing review of it
made in saying: ‘It’s too complicated for us; go in Fortune magazine, at one point hitting precisely
back and d o your own homework; learn about on the issue under discussion here:
your industry and your own distinctive com-
petences by immersing yourself in the details and There is nothing in Powerplay to support
Cunningham’s contention that she is a business
trying things; get many people involved; maybe genius. Her chapter about learning curves and
over a few years you’ll be able to develop an other B-school buzzwords seems infantile. What
effective strategy. It’s your responsibility; no one little discussion there is of actual business consists
can do it for mainly of genuflecting in front of a deity called
As for Andrews’ proposals about directors, his The Srrategy. The Strategy is what Mary and
Bill were up to when nasty-minded people
claim about ‘the power of strategy as a simplifying thought they were up to something else. Near
concept enabling independent directors to know as I can tell, it consisted of getting Bendix out
the business (in a sense) without being in the of a lot of fuddy-duddy old-fashioned products
and into glitzy high tech. What makes this a
terribly ingenious idea, let alone a good one,
In the early 1980s. frustration with the planning school she does not say. But she became very attached
and technique in particular, seems to have driven a number
of practitioners and consultants back to the simpler design to it. ‘How’s The Strategy going?’ she asked
school model. Typical is the article by Walker Lewis (1984). Agee the first time they met after her departure
founder of Strategic Planning Associates, and entitled ‘The from Bendix. And at the book’s emotional
CEO and corporate strategy in the 1980s: back to basics.’ It climax, as Agee realizes he’s going to lose
rediscovers all the elements of that model; for example, ‘the control of Bendix to Allied Corp., he says: * “Of
CEO must be an informed generalist;‘ ‘he must foster course, you know what this means? . . . The
the huilding of comparative advantage;’ ‘good strategic Strategy that we’ve worked on so hard“-and
management requires taking the wide view . . . it means here he nodded at me--“won’t be in our hands.” ’
setting a corporate direction based on . . . a comprehensively And they cry (1984: 142).
developed strategy;’ ‘In the end. it is the CEO who must
serve as the force behind a return to basic integrated strategies
in the 1980s;’ and ‘he must prod the corporation along the If the case study method, based on the design
path to implementation’ (pp. 1, 2. 6). Ironically, Lewis school model, has encouraged leaders to over-
concludes his article with the claim that ‘coming to terms
with these changes requires more than the old answers’ (p. simplify strategy, if it has given them the
6 ) . although that is precisely what he offers. impression that ‘you give me a synopsis and I’ll
190 H. Mintzberg
give you a strategy,’ if it has denied strategy That is why the field of strategic management
formation as a long, subtle, and difficult process has need for these different schools of thought,
of learning, if it has encouraged managers to so long as each is considered carefully in its own
detach thinking from acting, remaining in their appropriate context.
offices instead of getting into factories and Accordingly, we can begin to delineate the
meeting customers where the real information conditions that should encourage an organization
may have to be dug out, then it may be a major to tilt toward the design school model end of the
cause of the problems faced by contemporary continuum. We see a set of four in particular.
organizations.
This critique may sound extreme. We do not 1. One brain can, in principle, handle all of the
believe it is; as we shall discuss below, there is information relevant for strategy formation.
much in the design school to recommend it, at The assumption of the single strategist some-
least under certain circumstances (indeed much times does hold up: a chief executive (perhaps
in using cases as pedagogical devices too). But teamed up with other top managers), albeit
not when it is applied without a depth of one who is rather clever and especially adept
understanding of what a particular organization at synthesis, can take full charge ‘of the process
is, and how it must sometimes learn. for creating strategy. Here the situation must
be relatively simple, involving a base of
knowledge that can be comprehended in one
THE DESIGN SCHOOL: CONTEXT AND brain.
CONTRIBUTION 2 That brain has full, detailed, intimate knowl-
edge of the situation in question. The potential
Our critique has not been intended to dismiss the for centralizing knowledge must be backed up
design school model, but rather the assumption of by sufficient access to, and experience of, the
its universality, that it somehow represents the organization and its situation to enable the
‘one best way’ to make strategy. In particular, strategist to understand in a deep sense what
we reject the model where we believe strategy is going on. We might add that he or she can
formation must above all emphasize learning, only know the organization by truly being in
notably in circumstances of considerable uncer- the organization. This precludes the image of
tainty and unpredictability, or ones of complexity the case study classroom, the detached CEO
in which much power over strategy-making has with a pithy report, the ‘quick-fix’ consulting
to be granted to a variety of actors deep inside contract, the quarterly directors’ meeting,
the organization. We also reject the model even the weekend retreat of executives
where it tends to be applied with superficial (although this may culminate the process).
understanding of the issues in question. Rather it describes the strategist who has
Andrews thought it sufficient to delineate one developed a rich, intimate knowledge base
model and then add qualifications to it. The over a substantial period of time.
impression left was that this was the way to make 3. The relevant knowledge is established and set
strategy, although with nuance, sometimes more, before a new intended strategy has to be
sometimes less. But that had the effect of implemented-in other words, the situation is
associating strategy-making with deliberate, cen- relatively stable or at least predictable. Not
tralized behavior and of slighting the equally only must the strategist have access to the
important needs for emergent behavior and relevant knowledge base, but there must also
organizational learning. Another extreme-what be some sense of closure on that base: at
we have elsewhere presented under the label of some point in time, the strategist must know
the ‘grass roots model’ (Mintzberg and McHugh, what needs to be known to conceive an
1985)-makes no more sense, since it overstates intended strategy that will have relevance well
equally. But by positioning these two at ends of beyond the period of implementation. The
a continuum, we can begin to consider real-world world must, in other words, hold still, or-
needs along it. In other words it is not Andrews’ what amounts to a much more demanding
qualifications that will hold the model in check assumption-the strategist must have the
so much as an alternate depiction of the process. capability to predict the changes that will
The Design School 191

come about. What this means is that individual maximum period since the last full review’ (1982:
learning must come to an end before organi- 215-216). Such strategic reassessments may also
zational action taking can begin. And that can result from the introduction to the organization
happen effectively only when the future can, of fresh strategic thinking on the part of new
in fact, be known. leaders.
4. The organization in question is prepared to There is another context where the design
cope with a centrally articulated strategy. For school model might apply, and that is the new
one thing, others in the organization must be organization, since it must have a clear sense of
willing to defer to a central strategist. For direction in order to compete with its more
another, they must have the time, the energy, established rivals (or position itself in a niche
and the resources to implement a centrally free of their direct influence). This period of
determined strategy. And, of course, there initial conception of strategy is, of course, often
has to be the will to do that implementation. the product of an entrepreneur with a vision who
created the organization in the first place.
These conditions suggest some clear contexts in Context describes structure as well as time and
which the design school model would seem to situation. In the context described above, the
apply best-its own particular niche, so to speak, structure tends to be simple-flexible, non-
related to time as well as situation. In other elaborated, very responsive to the dictates of a
words, this is a model to be applied only in single leader (Mintzberg, 1979: chapter 17). Once
certain kinds of organizations, and even there under way, however, even simple structures with
only in certain circumstances. Above all is the entrepreneurial leaders may not follow the design
organization that needs a major reorientation, a school model, even in times of reconception,
new conception of its strategy. Newman recog- because the leader’s considerable personal discre-
nized this early, referring to the ‘quick reversal,’ tion (including personal control of ‘implementa-
the ‘sharp break’ (1967: 117). Or, as Rumelt has tion’) allows him or her to change strategy
put it, ‘a good strategy does not need constant gradually, even continuously, without any need
reformulation. It is a framework for continual to articulate it. In a way, Andrews recognized
problem solving, not the problem solving itself’ this when he sought to distance his model from
(1980: 365; see also Henderson, 1979: 38). the entrepreneurial context and its reliance on
Two conditions would seem to characterize intuition and ‘opportunism.’ But in so doing he
what we call this period of reconception. First, also distanced it from some of the most creative
there was a major change in the situation that strategy-making behavior found in organizations.
previously supported the existing strategy, so that The structural context Andrews seemed to
it has been seriously undermined. And second, favor for his model (although he would hardly
there has developed the beginnings of a new use the label we are about to apply to it), and
stability, one that will support a new conception the one that appears to be most appropriate for
of strategy. In other words, the design school the period of reconception of strategy in an
model would seem to apply best at the junction existing organization, is what we call ‘machine
of major shift for an organization, coming out of bureaucracy’ (Mintzberg, 1979: chapter 18). This
a period of changing circumstances and into one is structure characterized by a centralization of
of operating stability. authority and a relatively stable context of
We would normally expect the provoking operations, typically used in mass production
change to be one of a crisis or problem in and the mass delivery of services. Machine
the external condition of the organization, for bureaucracies commonly pursue highly articulated
example a major realignment of competition, a and stable strategies. They therefore require in
key shift in market demand, a technological periods of reconception much of what the design
breakthrough. Yavitz and Newman also suggest school has to offer: a process whereby someone
that what they refer to as ‘total reassessment’ in central command somehow pulls the new
can be proactive too, triggered, for example, by conception togetherdehes it if not actually
‘milestones in major programs,’ periods when creates it-and then articulates it fully at a point
‘large commitments of resources must be made’ in time so that everyone else can implement it
or ‘key uncertainties are resolved,’ or simply La and then pursue it.
192 H. Mintzberg

But there is an interesting anomaly here. The late. Thus, the design school model tends here
call from the design school for a personalized to get ‘caught in the middle,’ to use Porter’s
and creative form of strategic management (one phrase, tilts toward the personalized intuition of
strategist, strategies as novel conceptions) is not entrepreneurship for major reformulation and
really compatible with machine bureaucracy, toward the analysis of planning for the more
which tends to rely on standardized procedures routine pursuit of strategy. Can we conclude,
and detached forms of control. In other words, therefore, that by trying to position the design
machine bureaucracies are not mobiles to effect school model free of intuition on one side and
strategic change but stabiles for the continued planning on the other, Andrews left it little room
pursuit of given strategies. For example, our own for real application, perhaps mainly marginal
research on strategy formation (Mintzberg, 1978; strategic change in the machine bureaucratic
Mintzberg et al., 1986) suggests that chief type of organization where leaders can exercise
executives of machine bureaucracies tend to be ‘judgement’ but not rely on intuition or analy~is?’~
caretakers of existing strategies-fine-tuners of As for more complex types of organizations,
set directions rather than champions of radically which depend on expertise for their functioning,
new ones-in part because of the constraints as we have argued elsewhere, ‘professional
imposed by their own standardized procedures. bureaucracies’ and ‘adhocracies’ cannot rely
These organizations are, after all, machines on the conventional prescriptive approaches to
dedicated to the pursuit of efficiency in very strategy-making, whether design, planning, or
specific domains. Indeed, the whole array of positioning school oriented, but must instead tilt
mechanisms proposed in the design school’s own toward the learning end of the continuum,
model of implementation-performance meas- developing strategies that are more emergent in
ures? incentive systems, various other control nature through processes that have more of a
procedures, not to mention the articulation of grass roots orientation (Hardy et al., 1984;
strategy itself, as noted earlier-nce in place Mintzberg and McHugh, 1985).16
act not to promote change in strategy but To conclude, should we take the design school
to resist it. Formal implementation, ironically, model literally? In assessing the real contribution
impedes reformulation. of this school, perhaps we should not. For while
Our own evidence (Mintzberg, 1978), as well the model (even the framework) may have
as that of Miller and Friesen (1984), suggests restricted application and often be overly simpli-
that major reformulation in machine bureaucracy fied, this school’s contribution as an ‘informing
typically occurs through a form of revolution; idea’ has been profound. The design school has
power is centralized around a single leader who provided important basic vocabulary by which
acts personally and decisively to unfreeze existing we discuss grand strategy, and it has provided
practices and impose a new vision. In other the central notion that underlies all prescription
words, in such ‘turnarounds,’ the organization in this field, namely that strategy represents
tends to revert to the more flexible simple a fundamental congruence between external
structure, and to its more entrepreneurial mode opportunity and internal capability. These impor-
of strategy-making, at least until it has developed tant contributions will stand no matter how many
a new realized strategy, after which it tends to of this school’s specific premises may fall away.
settle back down to its old machine bureaucratic
way of functioning.
The implication of this is that while the machine
l 5 A study of the cases favored by the design school may be
bureaucracy may occasionally require a period
instrumental in this regard. Our own suspicion is that there
of reconception as provided for by the design is probably a predisposition toward mass production or mass
school model, its own procedures impede the service organizations, typically machine bureaucratic in
faithful use of that model. In a sense, implemen- nature, although the role of the intuitive leader in trying to
effect turnaround in them in a personalized way may be
tation fits, formulation does not. Indeed, initial more evident in the cases than in the theory (e.g. in the J.
use of the model itself discourages later use of I. Case case, in Learned er al., 1965: 82-102).
I h Note that, in describing the strategy-making process
it: by articulating strategy and implementing it,
favored in different types of organizations, we are further
as prescribed, the machine bureaucracy finds it making the case for the impact of structure on strategy (see
difficult to change its strategy later, to reformu- also Normann, 1977: 9, 19 and Bower, 1970: 286-287).
The Design School 193

ACKNOWLEDGEMENTS Cunningham, M. and F. Schumer. Powerplay: What


Really Happened at Bendix, Simon & Schuster.
My thanks to one very thorough and considerate New York, 1984.
reviewer, also to Bill Newman for his comments D i m , W. A. ‘Competitive strategic planning’,
Business Quarterly, Spring 1985, pp. 22-28.
on some of the early history of the use of the Feld, M. D. ‘Information and authority: The structure
strategy concept in business. of military organization’, American Sociological
Review, XXIV, 1959, pp. 15-22.
Gilmore, F. F. and R. G. Brandenburg. ‘Anatomy of
corporate planning’, Haward Business Review,
REFERENCES November-December 1962, pp. 61-69.
Hamermesh, R. G. Making Strategy Work, Wiley,
Andrews, K. R. Memorandum: ‘The fonnulation- New York, 1986.
implementation dichotomy’ in the Concept of Hardwick, C. T. and B. F. Landuyt. Administrative
Corporate Strategy. Harvard Business School, 21 Strategy, SimmonsBoardman, New York, 1961.
October 1976. Hardy, C., A. Langley, H. Mintzberg and J. Rose.
Andrews, K. R. The Concept of Corporate Strategy, ‘Strategy formation in the university setting.’ In J.
Irwin, Homewood, IL, 1971; second edition, 1980a; Bess (ed.), College and University Organization:
Third edition, 1987. Insights for the Behavioral Sciences, New York
Andrews, K. R. ‘Directors’ responsibility for corporate University Press, New York, 1984.
strategy’ Harvard Business Review, November- Hayes, R. H. ‘Strategicplannineforward in reverse?’,
December 1980b, pp. 3044. Harvard Business Review, November-December
Andrews, K. R. ‘Replaying the board’s role in 1985, pp. 111-119.
formulating strategy’, Harvard Business Review, Henderson, B. D. On Corporate Strategy, Abt Books,
May-June 1981, pp. 18-28). Cambridge, MA, 1979.
Ansoff, H. I. ‘A quasi-analytical approach to the Hofer, C. W. and D. Schendel. Strategy Formulation:
business strategy problem’, Management Tech- Analytical Concepts, West, St. Paul, MN, 1978.
nology, June 1964, pp. 67-77. Hrebiniak, L. G. and W. F. Joyce. Implementing
Ansoff, H. I. Corporate Strategy, McGraw-Hill, New Strategy, Mamillan, New York, 1984.
York, 1965. Katz, R. L. Cases and Concepts in Corporate Strategy,
Barnard, C. I. Organization and Management: Selected Prentice Hall, Englewood Cliffs, NJ, 1970.
Papers, Harvard University Press, Cambridge, MA, Kiechel, W. ITI ‘Sniping at strategic planning’ (Inter-
1948. view with himself), Planning Review. May 1984,
Berg, N. General Management: An Analytical pp. 8-11.
Approach, Irwin, Homewood, IL.,1984. Kiesler, C. A. The Psychobgy of Commitment:
Boston Consulting Group Inc. Strategy Alternatives for Experiments Linking Behavior to Belief, Academic
the British Motorcycle Industry (A report prepared Press, New York, 1971.
for the Secretary of State for Industry, Government Kinsley, M. ‘A business soap opera’, Fortune, June
of Great Britain, 1975). 1984, pp. 141-144.
Bourgeois, L. J. 111 and D. R. Brodwin. ‘Strategic Kress, G., G. Koehler and J. F. Springer. ‘Policy
implementation: Five approaches to an elusive drift: An evaluation of the California Business
phenomenon’, Strategic Management Journal, 5 , Enterprise Program’, Policy Studies Journal, Special
1984, pp. 241-264. Issue no. 3, 1980, pp. 1101-1108.
Bower, J. L. Managing the Resource Allocation Learned, E. P., C. R. Christensen, K. R. Andrews
Process: A Study of Corporate Planning and and W. D. Guth. Business Policy: Text and Cases,
Investment. Harvard Business School, 1970; revised Irwin, Homewood, IL, 1965.
edition, 1986. Learned, E. P. and A. T. Sproat. Organization Theory
Burgelman, R. A. ‘A process model of internal and Policy: Notes for Analysis, Irwin, Homewood,
corporate venturing in the diversified major firm’, n,1966.
Administrative Science Quarterly, XXVIII, 1983, Lenz, R. T. ‘Strategic capability: A concept and
pp. 223-244. framework for analysis’, Academy of Management
Chandler, A. D. Jr. Strategy and Structure: Chapter Review, 5, 1980, pp. 225-234. .
in the History of the Industrial Enterprise, Massachu- Levitt, T. ‘Marketing myopia’, Haward Business
setts Institute of Technology Press, Cambridge, Review, July-August 1960, pp. 45-56.
MA, 1962. Lewis, W.W.‘The CEO and corporate strategy: Back
Christensen, C. R., K. R. Andrews and J. L. Bower. to basics’. In A. C. Hax (ed.), Readings on Strategic
Business Policy: Text and Cases, fourth edition, Management, Ballinger, New York, 1984.
Irwin, Homewood, IL, 1978. Lindgren, U. and K. Spangberg. ‘Corporate acqui-
Christensen, C. R., K. R. Andrews, J. L. Bower, R. sitions and divestments: The strategic decision-
G. Hamermesh and M. E. Porter. Business Policy: making process’. International Studies of Manage-
Text and Cases, fifth and sixth editions, Irwin, ment and Organization, XI, 1981, pp. 24-47.
Homewood, IL, 1982 and 1987. Lipsky, M. ‘Standing the study of public policy
194 H . Mintzberg
implementation on its head’. In W. D. Burnham York, 1977.
and M. W. Weinberg (eds), American Politics and Ohmae, K. The Mind of the Strategist, McGraw-Hill,
Public Policy. MIT Press, Cambridge, MA, 1978. New York, 1982.
Livingston, J. S. ‘The myth of the well-educated Pascale, R. T. ‘Perspectives on strategy: The real story
manager’, Harvard Business Review, January- behind Honda’s success’, California Management
February 1971, pp. 79-89. Review, Spring 1984, pp. 47-72.
Majone, G. and A. Wildavsky. ‘Implementation as Porter, M. E. Competitive Strategy: Techniques for
evolution’, Policy Studies Review Annual, II, 1978, Analyzing Industries and Competitors, Free Press,
pp. 103-117. New York, 1980.
McKinsey, J. 0. Adjusting Policies to Meet Changing Porter, M. E. ‘The contributions of industrial organi-
Conditions. American Management Association zation to strategic management’, Academy of Man-
General Management Series, AM116, New York, agement Review, VI, 1981, pp. 609-620.
1932. Porter, M. E. ‘Industrial organization and the evolution
Miles, R . H . CofJin Nails and Coporate Strategies, of concept for strategic planning: The new learning’.
Prentice-Hall, Englewood Cliffs, NJ, 1982. Managerial and Decision Economics, IV, 3, 1983,
Miller, D. and P. H. Friesen. Organizations: A pp. 172-180.
Quantum View, Prentice-Hall, Englewood Cliffs, Porter, M. E. Competitive Advantage: Creating and
NJ, 1984. Sustaining Superior Performance, Free Press, New
Mintzberg, H. The Nature of Managerial Work, Harper York, 1985.
& Row, New York, 1973. Quinn, J. B. ‘Strategic change: “Logical incremen-
Mintzberg, H. ‘Patterns in strategy formation’, Man- talism” ’, Sloan Management Review, Fall 1978, pp.
agement Science, XXIV, 1978, pp. 934-948. 7-21.
Mintzberg, H. The Structuring of Organizations, Quinn, J. B. Strategies for Change: Logical Incremental-
Prentice-Hall, Englewood Cliffs, NJ, 1979. ism, Irwin, Englewood Cliffs, NJ, 1980.
Mintzberg, H. ‘Crafting strategy’, Harvard Business Radosevich, R. ‘A critique of “comprehensive mana-
Review, July-August 1987, pp. 66-75. gerial planning” ’. In J. W. McLuin (ed.), Contem-
Mintzberg, H. ‘Strategy formation: Ten schools of porary Management, Prentice-Hall, Englewood
thought’, in J. Fredrickson (ed.), Perspectus on Cliffs, NJ, 1974.
Strategic Management, Ballinger,-New York, forth- Reilley, E. W. ‘Planning the strategy of the business’,
coming. Advanced Management, December 1955, pp. 8-12.
Mintzberg, H. and A. McHugh. ‘Strategy formation Rein, M. and F. F. Rabinovitz. ‘Implementation: A
in an adhocracy’, Administrative Science Quarterly, theoretical perspective’. In W. D. Burnham and
XXX, 1985, pp. 160-197. M. W. Weinberg (eds), American Politics and
Mintzberg, H. and J. A. Waters. ‘Tracking strategy in Public Policy, MIT Press, Cambridge, MA, 1979.
an entrepreneurial firm’, Academy of Management Rumelt, R. P. ‘Evaluation of strategies: Theory and
Journal, XXV, 1982, pp. 465-499. models’. In D. E. Schendel and C. W. Hofer (eds),
Mintzberg, H. and J. A. Waters. ‘Of strategies, Strategic Management, Little, Brown, Boston, MA,
deliberate and emergent’, Strategic Management 1979.
Journal, 6 , 1985, pp. 257-272. Rumelt, R. P. ‘The evaluation of business strategy’.
Mintzberg, H., J. P. Brunet and J. A. Waters. ‘Does In W. F. Glueck (ed.), Business Policy and Strategic
planning impede strategic thinking?’, Advances in Management, McGraw-Hill, New York, third edi-
Strategic Management, vol. 4, JAI Press, Greenwich, tion, 1980.
CT, 1986, pp. 3-41, Rumelt, R. P. ‘Towards a strategic theory of the
Mintzberg, H., D. Raisinghani and A. Theoret. ‘The firm’. In R. B. Lamb (ed.), Competitive Strategic
structure of “unstructured” decision processes’, Management, Prentice-Hall, Englewood Cliffs, NJ,
Administrative Science Quarterly, XXI, 1976, pp. 1984.
246-275. Scheff, T. J. ‘Control over policy by attendants in a
Moore, D. G. ‘Managerial strategies’. In W. L. Warner mental hospital’, Journal of Health and Human
and N. H. Martin (eds), Industrial Man, Harper & Behavior, 1961, pp. 93-105.
Row, New York, 1959, pp. 219-226. Selznick, P. Leadership in Administration: A Sociologi-
Newman, W. H. Administrative Action: The Techniques cal Interpretation, Harper & Row, New York, 1957.
of Organization and Management, Prentice-Hall, Smalter, D. J. and R. L. Ruggles. Jr ‘Six business
Englewood Cliffs, NJ, 1951; second edition, 1963. lessons from the Pentagon’, Harvard Business
Newman, W. H . , C. E. Summer and E. K. Warren. Review, March-April 1966, pp. 64-75.
The Process of Management Concepts, Behavior, Stevenson, H. H. ‘Defining corporate strengths and
and Practice, Prentice-Hall, Englewood Cliffs, NJ, weaknesses’, Sloan Management Review, Spring
1951; second edition, 1967. 1976, pp. 51-68.
Newman, W. H . , E. K. Warren and A. R. McGill. Sun Tzu. The Art of War ( S . B. Confitt, translator),
The Process of Management: Strategy, Action, Oxford University Press, New York, 1971.
Results, Prentice-Hall, Englewood Cliffs, NJ, sixth Thoenig, J. C. and E. Friedberg. ‘The power of the
edition, 1987. field staff The case of the Ministry of Public Works
Normann, R. Management for Growth, Wiley, New and Urban Affairs and Housing in France’. In A.
The Design School 195

F. Leemans (ed.). The Management of Change in decisions’, Harvard Business Review, September-
Government, Martinus Nijoff, The Hague, 1976. October 1967, pp. 91-99.
Tilles, S. ‘How to evaluate corporate strategy’, Harvard Yavitz, B. and W. H. Newman. Strategy in Action:
Business Review, July-August 1963, pp. 111-120. The Execution, Politics, and Payoff of Business
Tregoe, B. B . and J. W. Zimmerman. Top Management Planning, Free Press, New York, 1982.
Strategy, Simon & Schuster, New York, 1980. Zand, D. E. Information, Organization, and Power,
Wrapp, H. E. ‘Good managers don’t make policy McGraw-Hill, New York, 1981.