Tracking Strategy in an Entrepreneurial Firm
Author(s): Henry Mintzberg and James A. Waters
Source: The Academy of Management Journal, Vol. 25, No. 3 (Sep., 1982), pp. 465-499
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/256075
Accessed: 30/03/2010 11:02
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=aom.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy
of Management Journal.
http://www.jstor.org
? Academy of Management Journal
1982, Vol. 25, No. 3, 465-499.
in
an
Tracking Strategy
Firm'
Entrepreneurial
HENRY MINTZBERG
JAMES A. WATERS
McGill University
Thisstudy tracksthe strategiesof a retail chain over
60 years of its history to show how that vague concept
calledstrategycan be operationalizedand to drawconclusionsabout strategyformation in the entrepreneurial
firm that grows large and formalizes its structure. The
conclusionsfocus on patternsof strategicchangeand on
andplancontrastingcharacteristicsof entrepreneurship
ning.
Little researchhas been undertakenon how strategiesactuallyform in
organizations,for some obvious reasons. First, strategiesdo not change
on schedule; they may remain stable for years, even decades, before
changing.Second, even whenthey do change,the processcan be complex.
Both of these characteristicscall for intensive, longitudinal research,
whichmeanssmallsamplesizesand largeinvestmentsof time. Researchers
in managementgenerallyhave hesitatedto undertakesuch research.
More important,perhaps, researchhas been discouragedby the very
way in whichthe conceptof strategyhas been conceived.In the literature,
strategyalwayshas been definedin termsof intentions,guidelinesfor the
future-essentially in termsof plans. Chandler'sdefinitionis typical:"the
determinationof the basiclong-termgoals and objectivesof an enterprise,
and the adoptionof coursesof action and the allocationof resourcesnecessaryfor carryingout these goals" (1962, p. 13). But conceivingstrategy
in terms of intentionsmeans restrictingresearchto the study of perceptions of what those who, it is believed, make strategyintend to do. And
that kind of research-of intentions devoid of behavior-simply is not
very interestingor productive.
Are strategiesnot also enacted?In otherwords, is therenot a need for a
definitionof the wordthat encompassesthe "strategies"actuallypursued
'The authors wish to express their appreciation to Alexandra McHugh and William Taylor, who
took part in the brainstorming sessions, to the Social Sciences and Humanities Research Council of
Canada, which funded the research, and especially to the executives of Steinberg Inc. for their enthusiastic cooperation throughout the study.
465
Academyof ManagementJournal
466
September
Figure 1
Types of Strategies
y^-
-
.".... >
Intended
Strategy ~~~~~~~~Strategy
,1Strategy
Unrealized
Strategy
~
Deliberate
Strategy
X
Realized
Emergent
Strategy
by organizations?And, if so, is it not then conceivablethat organization
may sometimesnot succeed in pursuingthe strategiesthey intended, indeed that they may end up pursuingstrategiesthey never intended?The
authors believe that the answersare yes. Hence it is proposed that the
usual definitionbe called "intendedstrategy"and that strategyin general
and realizedstrategybe defined as "a patternin a streamof decisions"
(Mintzberg,1978). In other words, strategiesbecome consistenciesin the
behaviorof organizations,which rendersthe concept operationalfor the
researcher.The study of strategymakingbecomes the searchfor consistenciesin decisionmakingbehavior,the investigationof their appearance
and disappearance,and the analysisof the relationshipsbetweenintended
and realizedstrategies.Specifically,with referenceto Figure 1, one can
then ask when and why strategiesare deliberate(intentionsare realized),
when unrealized,and when emergent(patternsare realizedwithoutintention).
Method
Followingthis line of argument,a major researchprojectwas designed
to trackstrategiesand the processesby whichthey form in organizations.
The organizationsfor investigationwere selected one by one; and their
strategiesand strategymaking behaviors,together,withtheir structures,
environments,and performances,werestudiedintensivelyover periodsof
decades. Specifically,each study proceededin four major steps:
Step1
The first step was the collectionof basic data. Each study began by developingchronologiesof the decisionsand actionsthat shapedthe organization's history,as well as of relatedtrendsand eventsin the environment
and availablefigureson the organization'sperformance.The purposehere
is to record systematicallywhatevertraces remain of the organization's
history,in graphicalform wherepossible, otherwiseas simplechronological lists.
1982
Mintzberg and Waters
467
Step2
Step 2 was to infer the patternsand periods. Strategieswere inferred
from these graphsand lists as patternsin streamsof decisionsor actions.
(Actionsimply that decisionshave been made.) Each strategywas given a
label(e.g., "expansion"for a steadyprogressionof storeopeningsin a retail chain)and representedsymbolicallyon a time scale. (A similarprocedurewas carriedout for organizationalstructures.)These symbolicrepresentationswerethen stackedup on the commonscale, coveringthe entire
periodof the study, and scannedall together,so that majorperiodsin the
historyof the organizationcould be inferred.
Step3
An investigationof each periodwas the thirdstep. As attentionwas focusedon explainingthe transitionfrom one periodto the next, the orientation of the researchchanged, from the rathersystematicanalysisof tangible data to the more probinginvestigationof softer data. At this point,
use was made of organizationalrecordsand in-depthreportscarriedout
by, for, or about the organization,and interviewswere conductedwith
key actorsof the periodsin question.Of interestherewerethe people and
forces that shapedthe strategiesof each period-in other words, the underlyingcauses of the major changes-as well as the interrelationships
among the differentstrategiesand relatedstructures.
Step4
Step4 was the buildingof theory.The researchteam sat down with a detailed reporton the organization'shistory-the descriptionsand explanations of its patternsand periods-for a series of brainstormingsessions.
These focused on a set of major conceptualissues, with the intentionof
extractingand inducingwhatevertheoreticalconclusionscould be drawn
from these particularresults. The results served as the stimulus for the
team'sthinking,but its conclusionsoften go beyondthem, drawingon the
resultsof other studiesas well as the entire experienceof conductingthe
research.Of interest,for example,are the patternsof changein strategies
themselves;the interplayof environmentalchange, leadership,and bureaucraticmomentumin determininghow strategieschange;the relationshipsbetweenstructuralconfigurationand strategymakingmode;and the
place of deliberatenessand proactivenessin these processes.
This paperpresentsthe resultsand conclusionfrom one majorstudyusing this methodology-strategy formationin SteinbergInc., from 1917,
when Mrs. Ida Steinbergopened a tiny fruit and vegetablestore in Montreal's ethnic area, to 1974, by which time the companyher son Sam had
built up reachedsalesof over$1 billion, with 191supermarkets,32 department (discount) stores, 33 catalog stores, and 119 small restaurants,in
468
Academy of Management Journal
September
additionto the 15 pharmaciesfor whichit providedmanagementservices,
the 28 shoppingcentersit owned, as well as its own flour mill, sugar refineryand food manufacturingfirm. (Some readersmay be familiarwith
the series of films on this company, called "Corporation," by the National Film Boardof Canada,whichis widelyusedin Americanand Canadian businessschools.) Otherstudiescompletedor nearingcompletionusing the methodpresentedherehave includedan automobilecompanyand
a governmentengagedin a foreignwar(Mintzberg,1978),a magazineand
a newspaper,a film makingagency,an airline,a garmentmanufacturer,a
mining company, and a university. In addition, doctoral students in
Franceas well as MBA studentswritingthesesin Canadahave carriedout
a largenumberof smaller,but relatedstudies.
Two themesarepursuedin this paper.The firstis simplyto demonstrate
how the wide arrayof strategiesused in an organizationover the yearscan
be describedand analyzed, in both concrete and conceptual terms. In
other words, differentpicturesof that vague concept called strategyare
shown-some quantitative,representingspecific traces left behind in an
organization,and some symbolic,representingpossibleinterpretationsof
those traces. Second, from the investigationof what is believedto be a
classic case of growth, formalization,and diversificationof an entrepreneurialfirm, some conceptualconclusionsabout the process of strategy
formationunderthese conditionsare drawn.To begin, specificstrategies,
and related data, of SteinbergInc. from 1917-1975are presented, and
seven distinctperiodsin the historyof the companyare inferred.
The studyof an entrepreneurial
firm, especiallyone like SteinbergInc.,
is a mixedblessing.On one hand, the companyis very open and cooperative, whichmakesthe researcha real pleasure.On the other hand, the informalityalso meansa relativeabsenceof formalrecords.(It is suspected
that the study of only organizationswith detailedrecordsof their history
would bias the conclusionstoward bureaucratic-typeorganizations.)Accordingly,it was difficult to extractthe data. Annual reports, which appeared from 1953, helped, and many membersof the organizationkept
their own recordsof store openings and closings (not always consistent
with each other). Beyondthat, data collectionmeantgatheringeveryscrap
available-an advertisingmanager'sown list of promotionalcampaigns,
an ad hoc recordof privatelabel products,occasionalnewspaperarticles,
and so on. These problems, however, were alleviated by the excellent
memoriesof a numberof the people we interviewed.Finally,it should be
madeclearthat the approachused producesnot so mucha full historyof a
companyas one orientedto strategyand majorturningpoints. Thismeans
less attentionto more regularoperatingissues and to the rich patternsof
humaninteractionpresentin any organization.
The Strategies of Steinberg Inc. 1917-1974
In the courseof this company's60-yearhistory,somethingon the order
of 50 distinct strategieswere inferred. Some of these were from actual
1982
Mintzberg and Waters
469
plots of specific decisionsand actions, as in the case of expansionstrategies basedon storeopenings.Otherswereinferredfrom chronologicallists
in wordsof decisionsand actions, as in the case of servicestrategiesbased
on descriptionsof specificactions.Presentedin brief form arethesestrategies as well as related materialon structure,environment,and performance, in symbolicform. (The full reportruns some 200 pages.)
QuebecFood Outlets
Figure2 depictsSteinbergInc.'s food outlet strategiesin the Quebecregion over the period of the study. (Note that in all the strategydiagrams
the vertical dimensionhas symbolic meaning only.) The strategieswere
Figure 2
Food
Outlets Strategies
Quebec
Very Rapid
I
Expansion
Steady
Postwar
Moderated
Holding Expansion
Steady
Expansion
Closings
Expansion
1930
1920
.. l
....
.!I I'
e....
1940
. I
i ...........1.... ..'..
1970
1960
1950
Figure 3
Openings and Closings of Quebec Division Food Outlets
+20
1920
1925
1930
1935
1940
1945
1950
1955
1960
1965
1970
1975
470
Academy of Management Journal
September
inferredfrom the patternof actual openingsand closingsof food outlets
in each (fiscal) year. These are plotted in Figure3.
The label "piecemealexpansion"for 1917to 1930is used becauseonly
two new stores were opened, one in 1926 and one in 1928. This pattern
changedin 1931and on to 1941a net averageof 2.2 storeswereaddedper
year, the expansionoccurringin waves. Followingthe "holding pattern"
of the war years, the steady expansionresumed,and between 1948 and
1952 the averagenet addition of stores was 1.8 per year. Following the
year 1953, which sits off by itself, therewas a very rapidexpansionfrom
1954to 1960-stores were addedat an averagerate of 7.7 per year. From
1961 to 1975, this rate fell to 3.7 stores per year, hence the label "moderatedexpansion."
GeographicExpansion
The firm began operationsin Montrealand moved into other areas of
Quebec and into Ontariolater in its history. Figure4 shows the strategy
followedwith respectto food outlet location, both in the Quebecareaand
in Ontario. These strategieswere inferredfrom the patternof openings
Figure 4
Geographic Expansion Strategies
QUEBEC
I
llExpansion
Expansion Primarily Within Montreal
ONTARIO
. 1 ... .. . ,
1920
Balanced
Insideand Outside Expansionin Waves
Montreal
Outside Montreal
Consolidation Outside Expansion Outside
Toronto/Slow, Uneven Toronto/Slower Growth
i Toronto
Growth in Toronto
Purchase
39
>
Purchase
39
Grand Union
Stores
. .I . . . ...
.
.
. i
.
I. . .. . . . . . . . . ... . . . . . ..
1940
1970
1930
1950
1960
and closings inside and outside the major metropolitanareas in the two
areas (later called divisions). In the Quebec division (Figure5), starting
with the secondstorein 1926,expansiontook placeessentiallyin Montreal
until 1954.
During the 1954-1960period, expansion was balanced inside (+24
stores)and outside (+ 29 stores)Montreal.From 1961to 1975, there was
expansionin wavesoutsideMontreal;of the total net additionof 56 stores
over this latter period, 54 were outside Montreal.
An importantevent in the historyof the firm occurredin fiscal 1959the purchaseof 39 stores in Ontario from the GrandUnion Company.
This was the start of the OntarioDivision (see Figure6). After this very
1982
471
Mintzberg and Waters
Figure 5
and
Quebec Openings
Closings Inside and OutsideMontrealArea
Figure6
OntarioOpeningsand Closings Inside and OutsideToronto Area
I
OUTSIDE TORONTO AREA
+ 10
Total
-31
zC3
,+ 5
z
.
O
Z
i
rn I
T-1'u L' .--
472
Academy of Management Journal
September
largeadditionin 1959, there followed a period from 1960to 1970of consolidationoutsidethe Toronto area (- 13 stores)and slow unevengrowth
withinthe Torontoarea(+ 16 stores).Thispatternwas somewhatreversed
from 1971to 1975when therewas expansionoutsideToronto (+ 8 stores)
and slowergrowthin Toronto (+ 4 stores).
StoreSize
Beginningin 1931,the averagesales area of both openingsand closings
increased,although erraticallyand at different rates. The net effect of
these changeswas that the averagesize of operatingstores was increased
graduallyover the study period, as depictedin Figure7.
Figure 7
Store Size Strategy
e
sstet
1920
1930
1940
1950
1960
1970
Modernization
In 1963the firm begana programof upgradingand modernizingexisting outlets (changinglighting,fixtures,displays,etc.), as shown in Figure
8.
Figure 8
Modernization
Store
Strategy
Steady
Investment
1920
1930
1940
1950
1960
1970
Food RetailingService
Figure9 shows the servicestrategiesfollowed in the food retailingbusiness over the study period. Following the shift in 1922 to takiftggoods
from behind counters and exposing them to customers' view, a major
1982
473
Mintzberg and Waters
Figure 9
Food Retailing Strategies
Shopping
Center Support
Partial Reintroduction
f Services
MODERN SUPERMARKET
CONCEPT
Traditional
Service
1r^)
-*^
x
SelfXrService
Exposed
:onfod
Goods
..
.. .
1920
,
1930
..
1940
........
i
1950
Lines
.......
1960
.
1970
event in 1933was the conversionof one store to self-service.Betweenthat
event and 1936, all stores were converted to self-service, as shown in
Figure9 by slopinglines. The next majorshift was the additionin 1939of
meat in the stores. Duringthe period 1939to 1944, stores were gradually
convertedto includeself-servicefruitsand vegetablesand self-servicemeat
as indicatedby the slopinglines;at that point storesresembledsmallerversions of the supermarketas it is knowntoday. In 1954therebegana significantgrowthin sales of nonfood productsand a gradualreintroductionof
servicesin selectedareassuch as meat, cheese, and delicatessenproducts.
The year 1954also was the startof the move to shoppingcentersas an enlargementof the one-stop shoppingconcept.
FoodPromotion
Figure10 showsthreerelateddimensionsof the generalpromotionstrategies followedover the studyperiod. In termsof generalcommunications
with the public, therewas a shift from relianceon personalcontactto item
promotion(handbillsfrom 1920, first newspaperad in 1931),with the addition of more institutionaladvertisingstartingin 1946. The logo used in
advertisingand on store fronts and privatelabel productsalso changed
four differenttimesas shown. The name"MiracleFood Market"was employed in Ontariostartingin 1969.
The conversionto self-servicein 1933also markedan emphasison discount pricing,which lasted until the price controls broughton by World
WarII. From 1957to 1967,premiumstamps,redeemablefor merchandise
at companyoperatedredemptioncenters,wereused to buildand maintain
store traffic. Also duringthis same period customerscould accumulate
cash tapes to buy china and tablewareat low pricesin the store. A major
474
Academy of Management Journal
September
Figure 10
Food Promotion Strategies
Miracle Food
Wholesalearket
Groceteria
Logo\
rEIOGTISTEINBERG
SIGNATURE
TRIPLE-S LOGO
BLOCK "S"
LOGO
LOGO
]Premium Stamps
Disc~~
ounit=
|Discount
I >|Miracle
________
| Pricing
Png
TNT
Pricing
Pinky
""^ ==^-. __ j *.
ersona.'Z ~ -- _ ___
Contact
"
,__
Corporate
Promotion
Item Promotion
'
I
I
1920
1940
1930
i
. ....I
....
.........
I
....I'
1950
.
I
.
.
1960
.
.
.
.
.'
1970
eventin 1968was the introductionof systematic,across-the-boardmaintenanceof discountpricesand eliminationof specialpricepromotions.The
programwas called "miraclepricing."
BackIntegration
Three related dimensions of back integrationstrategiesare given in
Figure11. Beginningin 1927,the firmutilizedrentedwarehousefacilities.
After the purchasein 1945 of an old aircraft propeller factory, warehousing was done primarilyin companyowned facilities.
Manufacturingalso beganin 1945in that samebuilding.A smallbakery
was startedand was expandedseveraltimes over the study period. Facilities for roastingcoffee and nuts and producingmeat pies, and so on were
expandeduntil 1963, from which point capacityremainedconstant.With
initialinvestmentsin 1965in a sugarmanufacturerand in 1967in a flour
manufacturer,the company moved into staples manufacturing.Investment in both firms increaseduntil 1972, as indicatedin the diagram.
Privatelabel activitybegan in 1932 with a few productssuch as coffee
and tea. The rateof additionof privatelabelproductsacceleratedover the
studyperiodin steps:from approximately4 or 5 itemsin 1947,to 50 items
in 1958, to 180 in 1968, and an estimated600 productsby 1975.
Retail Diversification
Figure 12 shows diversification into retail businesses other than food
and was inferred from Figure 13, a plot of stores in operation over the
1982
475
Mintzberg and Waters
Figure 11
Back Integration Strategies
Staples
Manufacturing
Expansion of
Bakery Manufacturing
Continuation
of Nonbakery
Introduction of
Nonbakery Manufacturing
1
Bits & Pieces
Private Label
Expanded
Rate of
I[Introduction
Volume
Items
Full
Out
!
I
P~~~~~~~~~~~~~~i~
Owned Warehouses
Rented Warehouses
1920
1930
1940
,= j-,. j. . .
1950
. 1
.
1960
. j i -
1970
Figure 12
Retail Diversification Strategies
1920
1930
1940
1950
1960
1970
study period in various businessareas. Following45 years exclusivelyin
food retailing,from 1962 to 1964 the companyfollowed a steady expansion strategy focused in two new business areas-discount department
stores and restaurantsin Quebec.
Though not shown in Figure 13, in 1965 the company opened (and
closed)two gas stationsand announcedplansto open a furniturestorebut
Academy of Management Journal
476
September
Figure 13
Stores in Operation by Business Area
w
0l
0
[1.1
ID
z
1915
1920
1925
1930
1935
1940
1945
1950
1955
1960
1965
1970
1975
neverdid so. Also not shown was a joint venturein largesupermarketsin
France, begun in 1966 and terminatedin 1973. Pharmacyoutlets in Ontario grew from a startin 1969until terminationin 1974. Restaurantoutlets in Ontariobeganin 1965and increasedsteadilythrough1974.Catalog
storesgrewsteadilyform 1971through1974.Beginningin 1974,pharmacy
1982
Mintzberg and Waters
477
outletswereopenedin Quebec(Pharmaprix).Thus, basedon thesemoves,
the diversificationstrategy followed from 1965 to 1974 is describedin
Figure 12 as "expandedvarietyand pace (with consolidation)."
RealEstate
Until 1937all store sites wereleased(Figure14). Startingin 1937, an increasing proportion of stores and sites were owned and sites were
"banked" for future development.A discontinuityis shown in 1959 for
this strategy to indicate that sites in Ontario (purchasedfrom Grand
Union) were largelyleased; subsequently,land banking took place very
slowly there. The real estate activity in shopping centers showed steady
growthwith an increasedpace startingin 1970, when largeregionalshopping centersbegan to be developed.
Figure 14
Real Estate Strategies
1920
1930
1940
1950
1960
1970
Finance
Figure 15 shows the finance strategiesemployedover the study period.
At first, all capitalwas generatedinternallyfrom operations.Startingduring 1937with the first owned stores and sites. This sourcewas augmented
by mortgagefinancingin the form of sale-leasebackarrangements.
Academy of Management Journal
478
September
Figure 15
Finance Strategies
I
First Public
Financing
l
Mixed Mortgage and
Financing
Internal
1920
1930
(((I
,I
.....I........
1940
1950
Conventional
Balance Sheet/
Increased
Equity Capital
1960
Accelerated
Debt
FinanciDebt
I'.....
1970
A majorevent in 1953was the first publicfinancingin the form of a $5
milliongeneraldebenture.The firstpublicsale of preferredstock occurred
duringfiscal 1954, and until 1965capitalwas suppliedthrougha balance
of equityand debt financing(althoughall stock sold was nonvoting).Beginningin 1966and continuingthroughthe periodof the study, increasing
amountsof debt were undertakento finance expansions.
OrganizationStructure
In the evolutionof organizationstructure(Figure16), underpinningall
changeswas constantcontrol of all voting stock by Sam Steinberg.From
1917to 1930, to quote Sam Steinberg,"everybodydid everything."This
fluid approachchangedbeginningin 1930with assignmentof responsibility for functionalareas to specific family members.Startingduringthis
time and continuingthrough the period of study, nonfamily managers
began to assumekey positions.
Startingin 1947,the hierarchybecameelaboratedto the point that store
operationswere no longerunderthe directsupervisionof Sam Steinberg.
This trend acceleratedfrom 1955 to 1959 as more staff groups were
created, managementtraining programswere begun, and management
budgetingprocedureswere established.
The period 1959 to 1963 witnessedthe gradualemergenceof independent Ontarioand Quebecoperationdivisions.As a temporaryblip in this
trend, between1964and 1967many staff departmentswereestablishedin
the corporate office, and procedures and standards were formalized
throughoutthe firm. With some key personnelchanges,this trendwas reversed between 1967 and 1969 and many centralgroups were either cut
back or eliminated.
After 1970,divisionalizationwas extended;priorto this time, a number
of businessoperationsreportedloosely to Sam Steinberg.From 1970, all
operationswere run by two executivevice-presidentsand the president,
with Sam Steinbergas chairman.Between1972and 1974, reorganizations
took place in the discountstore operations.
479
Mintzberg and Waters
1982
Figure 16
Organization Structure
1920
1930
1940
1950
1960
1970
Environment
Some key elementsof the environmentof SteinbergInc. may be seen in
Figure 17. The period until 1929 saw the establishmentof food chains in
Quebec and Ontario(though their image, in terms of size of outlets and
type of service,was quite similarto the independents).Precipitatedby the
onset of the depression,the numberof chain outlets declinedduringthe
1930s.
Startingin 1941and acceleratingafterthe waruntil 1961,chainsbecame
increasinglyimportant.In Quebec duringthis period, their marketshare
rose from 15 percentin 1941 to 35 percentin 1961. The period 1961 to
1968was a plateau, and chain marketshareactuallydeclinedfractionally
duringthe period. Precipitatedby the discountpricingstrategyfollowed
by SteinbergInc., the period from 1968to 1973involveda price war betweenthe majorchainsin Quebecand Ontario,whichdecimatedthe ranks
of the independents.In 1973 and 1974, five major chains controlled 80
percentof the Quebec food market.
Performance
The actual sales and profit data for the firm are shown in Figure 18,
starting in 1931, the earliest date for which reliable data were available.
Academyof ManagementJournal
480
September
Figure 17
The Environment
1920
1930
1940
1950
1960
1970
As indicated,the only net loss experiencedby the firm occurredin 1933,
and the only declinein sales volume in the company'shistoryoccurredin
1934.
Periods in the History of Steinberg Inc.
All of the above strategydiagramswerecollectedon one largesheetwith
a common time scale from 1917-1975.All the strategieswere scanned
together to identify major turning points and infer key periods in the
history of SteinbergInc. In all, seven distinctperiodswere isolated. The
interestedreaderis invitedto scan the strategydiagramsto see why those
particularyearswerechosenas the startof new periods.(Spacelimitations
prohibitdiscussionof this.)
1917-1930
The years 1917-1930werea periodof formation-the traditionalfamily
store with seeds of change. The Steinbergstore opened in 1917 and in
many ways was a typical "mom and pop" operation, except that it was
mom and the kids. It was owned and operatedby membersof one family,
merchandisedits goods over the counter, knew its customerswell, sold to
them on credit, and often deliveredits goods to them. Sam Steinberg,the
1982
Mintzberg and Waters
481
Figure 18
Annual Sales Volume and Net Profit
10,000-
1,000.0-
Net Profit
100.0-
1,000-
.
~.
~
/\ /f
Sales Volume
1950
1960
6'?
O
o
|''V?O
<
Z
10.0-
100-
*
-
*L4I
0
'
II
10310...0
a
1.01.0-
I
1930
1940
1970
482
Academy of Management Journal
September
second eldest son, helped his motherout everyday after school from the
beginning,and he joined the storeon a full time basisin 1919,at the age of
13.
But in other ways the storewas not typical. The childrenacquiredfrom
their mothera strongbelief in the qualityof the merchandisesold to the
customersand a sense of honest dealing with them, beliefs that were to
characterizethe entire history of the firm. She also taught them a single
minded dedicationto the business (for example, investingin vegetables
when "killings"werebeingmadeeveryday in the stock market),to which
they attributetheir subsequentsuccess. Three other changes in the first
yearssignalledstrategiesthat wereto come-innovation in service(the exposureof goods) in 1921, expansionof the first storein 1919(a snap decision by Sam Steinberg,at the age of 13, duringa call from the landlord
askinghim to post a "for rent" sign on the spacenext door), the opening
of new storesin 1926and 1928(the formercalled,prophetically,"Number
1"), and the move into bulk purchasingand warehousingin 1927. During
the interviewsSam Steinbergrecalledimages of "beautiful" competitor
stores he wished to emulate, and of other competitors' stores: "dark,
dingy... Goddamit,after 10 yearsin business,I'm not going to look like
that."
From a conceptualviewpoint, this is seen as a period of formation, of
the establishmentof the basic values or ideology of the company-to-be
and the thoroughtrainingof the principalactors in the ways of the business. The strategies of the period were tightly integrated and rather
stable-this was a periodof continuity.The serviceand control strategies
could be characterizedas deliberatethough implicit,the piecemealexpansion perhapsas deliberatewith some emergentcharacteristics.
1930-1941
The period 1930-1941can be calledone of evolvingglobal change-perfectingand elaboratinga new retailingformula.The adventof the depression notwithstanding,1930 seemed to usher in a new era of growth for
what had just become Steinberg'sService Stores Limited. Three stores
wereopenedin 1931, and not a yearwould pass until 1943withouta store
opening. Successin those yearsof declinefor other, largerchains, appears
to have been based on the reputationthe young chainhad built up for the
qualityof its productsand services.But until 1933, there was a parochial
characterto the expansion-new stores were opened to "take care" of
membersof the familyand to pursuethe old customersto new areasof the
city of Montreal.
One event in 1933changedall that. The company"struckit bad" with
one new store, incurringunacceptablelosses ($125 per week). And so,
over the course of one eventful weekend, its name was changed to
"WholesaleGroceteria,"prices were slashed, personalservicescut, and
full self-serviceinstituted.Fromthere,in the wordsof SamSteinberg,"we
1982
Mintzberg and Waters
483
grewlike topsy." After sufferinga smallloss in 1933,profitsrose dramatically throughto 1939, and sales, after a small dip in 1934, rose to four
timestheir 1933level by 1940.The companywas neveragainto experience
eithera loss or a declinein sales. One articledecadeslaterclaimedthat the
companyassumedin these yearsa lead in pricesthat it neverrelinquished.
The other stores were all convertedto the same format by 1936, and expansion proceededin two major waves, the first peakingin 1937with six
stores, the secondin 1941with four. In effect, 1933usheredin new service
and promotion strategies(full self-serviceand discount pricing), altered
the expansionstrategy,and eventuallyled to majorchangesin other strategies-to one of buyingand bankingstore sites and to mortgagefinancing-as well as to the continual perfectingof self-serviceand later the
"supermarket"concept and to a continualelaborationof the structure.
The only reversionwas the gradualreintroductionof the Steinbergname
(in the form of Sam Steinberg'ssignature)on store fronts.
Froma conceptualviewpoint,herethereis an interruptdue to a competitive threatthat turneda companyaround,sooner or laterleadingto major changesin almost everyimportantstrategy.Sooner, the year 1933saw
global change-sudden reversalson a numberof importantstrategiesand later, in exploratoryor piecemealfashion, a numberof other strategies were changedin consequence.Throughoutthis decade, the company
movedslowlyto a new tightlyintegratedset of strategies,a new "gestalt,"
focused on the dominantelementof self-service.These strategiesseemed
deliberate(afterthe initialtest, a brief emergentphase, as in the converted
store of 1933),yet not so much plannedas opportunistic.In other words,
they were intended, but in no particularway and at no particulartime.
The approach,above all, was entrepreneurial,centeredon the vision of
one man who, in his words,made"all the decisionsat all times." Yet, paradoxically,all of this was stimulatedby a crisis,albeit one initiatedby the
company's own expansion. Moreover, the push to serve family needs
eventuallybecamea pull to transcendthem;one elementin the company's
success-personal service-became one of the first casualtiesof that success; and the most hostile of environmentsbecamethe host for successful
innovation.But this majorchangewas not a rethinkingof "what business
are we in?" The companywell knew what businessit was in; indeed, that
was the source of its strength. Its was a search for "how are we in our
given business?"
1942-1947
This was a periodof delayand diversion,a holdingpatternfrom 1942to
1947.The 1940sbroughta prolongedinterruptionof growth-the Second
WorldWar. Buildingmaterialsbecamescarce,labor was in short supply,
food rationed,and priceswereundergovernmentcontrol. Expansionwas
halted, as was strategicinnovation. What does a company with all this
energydo when it is held back?First, a greatdeal of energyhad to go into
484
Academy of Management Journal
September
keepingthe systemgoing-"begging" for stock, imposing"rationingon
the rationing" to ensure that the scarce goods were distributedfairly
among customers,and so on. Second, the delay allowed the companyto
continue perfectingthe new retailingconcept. Third, particularlyat war
end, when buildingmaterialsremainedscarceand the economyuncertain,
energieswereredirectedto new areasthat supportedthe basic operations.
A bakerywas opened in 1946, other small forms of manufacturingwere
initiated,and the privatelabelingprogramwas expanded.But most significantly, the companyprepareditself for expansion:it vigorouslypursued
its strategyof land banking,it engagedan architectto designfuturestores,
it boughtan oversizedwarehouse,it beganinstitutionaladvertising,and it
elaboratedits servicestructurein ways that suggestedan intentionto expand.
From a conceptualperspective,neverbefore had this companyexperienceda periodof consolidation;now one was imposedon it, and it reacted
in part by coping with the constraints,in part by preparingitself for the
expansionthat it believedwould inevitablycome. In other words, this was
not contingencyplanning, planning"if." It was planning"when." Nor
was this planningon paper;this was acting,buildingthe foundationrather
than designingthe building. One can attributethe long term perspective,
and the confidencein the eventualresumptionof expansion,to the presence of an ownerwho knewthat he would be leadingthe firm in that long
term. So the reactionto the interruptwas in some basic sense, again, proactive.
1948-1952
From 1948 to 1952 was a period of resumption,a returnto steady expansion. The economyhad begunits changeby 1948, and SteinbergLtd.,
as it was then called, was all readyto react. Whereasthe previousperiod
was characterizedby divertedenergiesand preparation,this one was characterizedby a returnto one centralfocus of attention-the expansionin
the numberof retailoutlets. This was the only majorchangein strategyin
this period, but it was one that redirectedenergiessignificantly,back to
wherethey were in the 1930s. A new wave of expansionwas underway.
In conceptualterms, this was a period of resumption,a returnto the
strategythat most interestedthis company.But unlikethe 1930s,therewas
less of an exploratorytheme. The formulahad been workedout. Now it
was pursued, deliberately.The years 1948-1952representeda period of
continuity.
1952-1953
The 1952-1953period was one of shifting gears, preparingfor the big
push. Sam Steinbergemphasizedin interviewsthat waves of expansion
were always followed by "pauses," in orderto solidify the changesthat
1982
Mintzberg and Waters
485
had been made, and to bring up the logistical support. But the pause in
1952 was unlike any other, for it led to a fundamental reorientation followed by the most important wave of expansion in the company's history.
The pause at first seemed like the others. In a 1952 article, Sam Steinberg boasted that "not a cent of any money outside the family is invested
in the company" and, asked about future plans, he replied with a "Who
knows? There is so much to do right ahead that it would sound like a wild
dream to talk about 10 years from now.... We will try to go everywhere
there seems to be a need for us." Yet a few months later he announced a
$15 million 5-year expansion program, one new store every 60 days for a
total of 30, the doubling of sales to 1957, new stores to average double the
size of existing ones, with parking lots, children's playrooms, and so on.
What caused such a dramatic change in thinking? The industry was
perched for its most dramatic expansion ever, with the postwar baby
boom and the population shift to the suburbs. Everything was set for a
new form of merchandising-the shopping center-and the company
most capable of exploiting the trend was the one that had been banking
choice land sites for over a decade.
But one component of the strategy had to be changed. Conventional
forms of financing were insufficient. The company had to go to public
markets for capital. And once that decision was made, things changed permanently for Steinberg's Limited.
After months of searching, Sam Steinberg finally found a financial
house that would support a debt issue-allowing him to retain 100 percent
control-and $5 million of general debentures were issued in December of
1952. That issue required "plans"-formal statements of intent. The market would not accept the word of one man; it needed precise descriptions.
So the entrepreneur was forced to plan formally, this time on paper, and
that drew him, and more importantly his company, partly into a new
mode of behavior.
In conceptual terms, a pause led to a rethinking, which defined a need,
which produced a decision, which altered in part the mode of strategic behavior and led (at the start of the next period) to global changes-to a reoriented gestalt. Here can be seen the utility of the externally imposed
pause, a time for reflection. Here also it can be seen that a single
choice-in this case to go to public financial markets-can have ramifications far beyond its own bounds. There seems to be a kind of push-pull
phenomenon here: an organization is pushed into changing one strategy,
and that in turn pulls it into far more consequential changes. The entrepreneur gets drawn by his own successes into a planning mode of strategy
making, one less compatible with his own managerial style. Not only does
a larger and more formalized organization structure constrain his opportunistic style, but so too does his increasing need to interact with the environment as the leader of an increasingly important organization. (In this
case, it was interaction with the financial community; later it was to become interaction with the social community.) Nevertheless, although the
486
Academy of Management Journal
September
initial move may have been reactive,the thrustremainedopportunisticthe companyhad to go public but only becauseit wishedto expandin a
new way. Yet it appearsthe environmentwas beginningto close in on an
entrepreneur.
1954-1960
The period 1954-1960was one of global change, then continuity-the
big push. Accordingto one long-termSteinbergexecutive,these were the
"make it or breakit" yearsfor supermarketchains, and Steinberg'smade
it, big. The company'sexpansionresumedat a rapid pace, most of it in
shoppingcentersand a good deal of it for the first time outsidethe Montrealarea. The strategyof publicfinancingwas pursuedat an accelerating
rate, includingthe first equity issue in 1955 (but nonvoting), as were the
strategiesof private labeling and bakery manufacturing.(Voting shares
have alwaysbeen held entirelywithinthe family, with 100percentof them
formallyvoted by Sam Steinberguntil the day he passed away in 1978.)
The structurebecameincreasinglyformalizedin those years, with a large
additionof centralstaff in the areasof training,accountingand control,
and marketingresearch.
The companyin fact almost did double its sales in five years, from $70
million to $132 million but, insteadof the 30 stores it promised,it in fact
opened 35 (and closed 9), one every47 days. The environmentwas benevolent in those years, and Steinberg'sLtd., took advantageof that fact.
A new 5-yearplan announcedin the 1958 Annual Report called for a
storea month for the next 60 months, and the 1959reportuppedthe ante,
callingfor those 60 storesto be built in 36 months. But other forces were
growingin the firm's environment.In 1957, the startof a brief recession,
the companybeganto redeemits pink cash registerslips for gifts, an indication of increasingcompetition.The hecticexpansionof the 1950shad to
lead eventually to saturation. By 1959, those slips had become Pinky
Stamps, a response,managementclaimed, to competitivepressures.The
war becameovert.
More significantly,whereasthe companyhad managedto avoid headon confrontationwith the two othermajorCanadianchains-the largesta
nationalchain, but a weak second to Steinberg'sin Quebec, the other an
Ontario-basedchain with no Quebecoperations-a move it took in 1959
engaged them directly. Up to 1959 every major strategicchange SteinBut a much
berg's made had been one of test-the-water-then-plunge.
of
1959
Ltd.
a
different
largerSteinberg's
adopted
approachin entering
the Ontariomarket:it plungedinitiallywith the purchaseof the 39 store
Grand Union chain there. As one executivenoted, "You can't get into
Ontarioone store at a time. You need a groupof stores." Its new Ontario
competitorsreactedstrongly, setting off a price war. In 1960, despite an
increasein sales of one-thirdover 1959, profitsdipped-for the first time
since 1944.
1982
Mintzberg and Waters
487
In conceptualterms, in the 1950sthe organizationunderstoodits environment, and took off. Environmentand strategy formed a perfect
gestalt. The period was one of global strategicchange at the start, followed by continuity.Throughit all, despiteanothernew orientation,the
companyremainedin the samebasicbusiness.It did embarkon something
new-shopping center development.That, however, was a means to sell
food and as such representeda form of verticalintegration.Throughout
the period, the organization'sstrategies,as previously,werelargelydeliberate and proactive,well suited to a benevolentenvironment,benevolent
at least to companies that understood it and could match its rate of
change. Steinberg'ssucceededbecauseit knew its businesswell-a condition that datedback to 1917-and becausein the 1930sit had the foresight
to bank land. "It," of course, in large part meant Sam Steinberg,but
backedup by a largerand largerorganization.
1960-1974
The 1960-1974years were a period of global change, then continuity,
with a new theme evolving, that of consolidationof traditionalbusiness
and a searchfor new relatedones. (Data on the companywerecollectedup
to the end of its fiscal year 1975, as the strategydiagramsshow. But, as
they also show, a numberof the strategiesof this last period seemed to
come to a halt, or at least a pause, in the year 1974,notablyall those associatedwith retaildiversification.Hencea tentativeend of this last periodis
shown to be in 1974.) The 1960 Annual Report announced "a year of
more conservativeachievements,"with the emphasison "consolidation"
of activities, "improvement"of existing facilities, and "integration"of
the new Ontariooperation.A new economicenvironmentfaced the company: the supermarketbusiness showed tendencies toward saturation,
which were reflectedin heaviercompetition,especiallyin Ontariowhere
the companyfound its new acquisitionto be in a weak position. Whereas
growthin the 1950scould come from findingnew placesto put stores, in
the 1960sit wouldhaveto come increasinglyfrom outsmartingthe competition and more effectively servingthe public. That meant a shift in emphasisfrom expansionto consolidation,from openingnew storesto making the existingones more efficient and attractive.
Moreover,the social environmentwas changingtoo. From a period in
which they could do no wrong, in the eyes of the public, the supermarket
chains increasinglythrough the 1960s and 1970s found themselvessubjected to all kindsof new social pressures-strikes, consumerprotestsand
boycotts, governmentinvestigations.These were the times of the cyclamate scare, Californiagrape boycotts, phosphatepollution, underweight
violations,labelinglaws, attackson excessprofits, and on and on. As they
grewlarge, the chainscould not maintaina personaltouch with theircustomers. And competitiondid not help. The whole set of stampprograms,
for example, deflected the chains from their basic missions-to deliver
488
Academy of Management Journal
September
food inexpensively.In effect, environmentsform gestaltstoo, and that of
the 1960swas dramaticallydifferentfrom the one of the 1950s.
And on the eve of this era, Steinberg'shad taken its boldest step everfor the first time a plunge without a testing of the water. And the water
proved cold. In addition to the reaction of the competition, the Grand
Union sites themselvesproved difficult. Many were inadequateand, in
sharp contrastto the Quebec operations,which expandedfrom a strong
base in Montreal,the firm found itself with only 8 of its 39 stores in the
Toronto area. Consolidationwas necessary-eventually,in fact, 34 of the
39 stores it bought were closed down. Moreover,the real estate position
provedcriticalin Ontario.Land bankingwas a key to the company'ssuccess in Quebec. But GrandUnion had no real estate position in Ontario,
and Steinberg'sdid not begina strategyof land bankingtherewith its purchase of GrandUnion. A numberof currentexecutivesattributethis to
the nonresidenceof any memberof the familyin Ontariountil 1970and to
the inabilityof the managerswho were sent thereto exerciseindependent
initiative.In Quebec,Sam Steinbergwould buy land on instinct,basedon
his intimate knowledgeof the area. That knowledgewas lackingin Ontario, and it was not to be developeduntil muchlater. And then it was too
late; the choice sites were largelygone.
These two points-a changed environment,economicallyhostile and
socially militant, coupled with the difficulty of digestingthe company's
biggestbite ever-set the stage for a globalreorientationof strategybeginning in 1960. The expansion strategy was moderated (and virtually
stoppedin the Montrealareain termsof net new stores),a new strategyof
storemodernizationswas initiated,therewas a surgeof centralizationand
formalizationof the structurein the mid-1960sby a new executivevicepresident(until Sam Steinbergput a stop to it), and attentionwas increasingly divertedto other spheresof activity. Back integrationand private
labelingstrategieswere pursuedwith more vigor. But, more importantly,
therearosea strategyof diversification(followedby movesto divisionalize
the structure).It began with discount (later to be called "department")
stores; then fast food restaurants;later, in the mid-1960s,a quickening
pace with a sugarrefineryand a flour mill (closer to diversificationthan
vertical integrationbecause most of the output was sold on the open
market);then pharmaciesand catalogstores, as well as otherventuresthat
did not take root (such as gas bars and the joint supermarketventurein
France).The resultsweremixed,with the discountstoresespeciallyposing
problemsshortlyafter their inception, the catalog chain eventuallysold,
and the restaurantsachievinga good deal of success.
The supermarketbusiness-which continuedto dominate,with over 85
percentof the sales by the end of the study period-remained in the doldrums until 1968. Competitivepressureshad led to the use of stamps,
games, contests, heavy advertisingpressures;the Quebec chains actually
lost a slight market share to the independentsbetween 1961 and 1969;
Steinberg'sprofit ratelevelledout in 1966and 1967, and droppedin 1968.
1982
Mintzberg and Waters
489
And so, in 1968, in contrast to the lingeringproblemsin the new businesses, the company acted dramaticallyand decisivelyin the old one,
adoptinga strategyremarkablysimilarto the one Sam Steinberghad used
on that eventful weekend in 1933, and with the same result in performance. WholesaleGroceteriaof 1933was MiraclePricingof 1968:significant, permanent,across-the-boardprice reductionscoupled with a complete shift in merchandisingphilosophy-the elimination of specials,
games and gimmicks(the Pinky stampshad been droppeda year earlier),
reducedservicelevels, advertisingbudgetcut in half. The companyin effect returnedto whatit knewbest-basic, no-nonsenseretailing,what one
executivecalled "a pure form of retailing."And it did it in the old testway, with a test in one storein a smallQuebeccity,
the-water-then-plunge
then implementationin Ottawa, followed quicklyby the rest of Ontario
and then Quebec. But one importantdetail differed: Miracle Pricing's
championwas not Sam Steinberg,but the head of the Quebec food division, who had to convincenot only the chief executivebut otherofficersas
well. And in the larger, more formalizedSteinberg'sLtd., that took a
numberof years.
Sales rose sharply in 1969, as the company attracted a significantly
greatermarketshare,but profitsdroppedand then reboundeddramatically in 1970as the new programtook hold. The effect on the Quebecindependents,traditionallya strongsegmentof that market,was "catastrophic" accordingto one studyof the industry.But, most importantly,Miracle
Pricingturnedaroundthe Ontariooperationand set it on a healthycourse
for the first time. Ironically,however,accordingto the informationavailable, althoughit probablygave added impetusto the existingstrategyof
expandingprivatelabeling and to the short term tendencyto streamline
the structure,MiraclePricingseemsto have led to no majorshifts in strategy outside the sphereof merchandising.
In conceptualterms,the periodbeganwith a rathersudden,global shift
in the environment,to which the company, true to form, reactedearly.
But that reactionviolatedwhat the chief executivehimself stressedas the
company's key success factor-knowledge of its business. The company
was drawninto new businesses,in some sense opportunistically,but without the clear theme or vision that characterizedearlierchanges. It was a
searchfor "what businesswe shouldbe in," but one that could not be undertakenon paper. To discoverits strengthsand weaknesses,its critical
success factors, the firm had to undertakean empiricalexplorationthat
spanned decades. A strategic theme of sorts did eventually get established-the 1967Annual Reportcalled it the "total marketingpackage,"
in effect the offeringof a varietyof retailservicesto take advantageof the
shoppingcentersthe companyowned and/or managed.But, in sharpcontrast to earlierchanges, this time the strategieswere less deliberate,more
emergent.
The mannerin which the diversificationprogramunfolded is shown
in Figure 19, startingfrom the traditionalfood retailingbusinessat the
Figure 19
From Back Integrationto Diversificationa
/
P
_eti-n
|Warehousing
t
___\
Real Estate
Shopping
\
1920
1925
1930
1935
Redemption Cente
- Manufacturing
1940
1945
^s^*^^^^^^^^^
?^
\\
!Pharmacy
\
1950 \\1955
\
\
196
x \
\
\
pStaples
\
\
\
^
XSClc\~
^^"-^.~
>\\
^1^
\\
in tw
aThis diagram was inspired by Rumelt (1974), who portrayed the diversification of Carborundum, Inc.,
tinuous time scale.
1982
Mintzberg and Waters
491
left, with verticalintegrationmoves above the line and horizontaldiversification ones below it, laid out on a time scale. Figure 19 shows how the
early back integrationmoves reinforcedeach other and later gave rise to
ones of diversification.Shopping centers provided the bridge, because
whatwas a-meansof back integration-to controlthe sites for stores-became a meansfor diversification-the mechanismwith whichto buildnew
kinds of stores. The figure essentiallyillustrateshow verticalintegration
can lead to diversification,indeedhow the sameactioninitiatedas one can
become the other. Whateverthe case, the companywas drawninto businesses it knew less about. It was building not on the strengthit always
had-long termknowledgeof the business-but on its marketstrength,its
control of sites.
Overall,the periodwas characterizedby dichotomies,by the old versus
the new, by a businessbased on knowledgeversusthose based on market
strength;by havingto plungeversusbeing able to test the water;by loose,
organic versus tight, formalized structure;by intuitive feel versus systematicanalyses;by the personaltouch versusthe formalsystem.Overall,
strategieswere less deliberatein this period, less explicit, less tightly integrated, less guidedby the vision of one man.
In contrastto the Steinberg'sof the 1930s,as the little guy fightingthe
giants, vulnerableon the economicfront but able to move quickly,that of
the 1970swas beingone of those giants, possessorof a powerfuleconomic
positionbut morevulnerableon the social front. Did SteinbergInc. of the
1970sin fact have greatercontrol over its own future than did the Steinberg ServiceStores of the 1930s?
Strategy Formation in the Entrepreneurial Firm
One theme that emergesfrom the historicalreviewof strategyin Steinberg Inc. is the presenceof waves or cycles. In the largest sense, in this
study are seen the classic stages of developmentcycle as describedby a
numberof managementtheorists(Chandler,1962;Filley, House, & Kerr,
1976;Scott, 1973). From the simplestructureconfigurationof the 1920s,
the companygraduallyunderwentincreasingelaborationand standardization of structureto arriveat a more formalizedconfigurationof the 1970s.
(For a more detaileddescriptionof these configurations,see Mintzberg,
1979. For a discussionof the concept of the configuration,see Millerand
Mintzberg,1980, and Mintzberg,1979.) In essence, as the seeminglyinevitableresultof growth,the small, personalized,highlyflexible,(but economicallyvulnerable),knowledge-basedfirm transformeditself into the
larger,more systematic,more economicallypowerful(but sociallyvulnerable) corporation.This transformationwill be examinedfrom two permode and the planningmode. But first, two
spectives,the entrepreneurial
other themes in the study can be considered:the uneven progress of
growth and the infrequencyof strategicchange.
Academy of Management Journal
492
September
Figure 20
Wave Pattern of Store Openings
+ 20
*A
+15
:
10
_+
/
O\
_z
nn
?l.n
I n .
l
.
.....
)
5
---
u
....
......
1920
1925
.. j ..
1930
.. ....
1935
I .......
1940
1945
j .... .. i
1950
1955
.
1960
.1965
-1010
1970
1975
Hectic, Uneven Growth
Withinthis large, evolutionarycycle, shorterrepeatingcyclesof expansion and consolidation can be detected and referredto as sprints and
pauses.As seen in Figure20, an overlayon the plot of QuebecFood Store
openingsand closings(Figure3), the notion of sprintsand pausesis releevant to most of the historyof the firm.
The imageof sprintingis congruentwith conventionalnotions of entrepreneurialactivityas the takingof bold, riskyleaps into the future.What
is less obvious is that, in the case of Steinberg,these bold leaps or sprints
always were accompaniedby subsequent periods of pause-times for
catchingup, consolidating.Sam Steinbergwas quite awareof this. In fact
it was he who used the word "pauses":
I did another thing, and I always did that. After I expanded. . then I'd always have a
period of pause, pause meaning a year or two to make sure that everything carried and
was working out. You are able to cope with the salary, you're able to cope with what
you've taken on. I'd always do that. So if you'll study the growth of our company,
you'll see that we have a period of expansion and a pause, an expansion and a pause.
This patternof sprintsand pausessuggestsan inchwormanalogy:an organizationleads with some primarystrategy,usuallyrelatedto expansion,
then pauses to bring up laggingstrategies,for example,logistic support,
then leadsagain, and so on. A numberof new storesare opened-perhaps
too many, overextendingthe resources-and then refinancingmust take
place, warehousesexpanded,staff found to man the new operations,and
so on. The approach is fundamentallyopportunistic, as opposed to
planned, a probe into the futurewithout full considerationof the consequences.
1982
Mintzberg and Waters
493
The resultis an unsteadypatternof growth,but one that can generatea
great deal of excitementand energywithin an organization.Growthbecomes the all-out sprint, pause, the time to catch breath, "to make sure
that everythingcarriedand was workingout." The inchwormprobes its
head into an uncertainfuture, then bringsup its rear-the baggage-to
keep pace.
Some definiteadvantagescan be seen in the inchwormform of growth,
at least where the organizationcan absorb the swings. The organization
finds its openings-its shortlived strategicwindows-and exploitsthemto
the hilt, damningthe consequences,with a faith that otherneeds-people,
money, warehousecapacity-will get straightenedout eventually.Sprinting is a way to focus the resourcesand energyavailableso that greatpressure is broughtto bear on opportunitiesas they presentthemselves.The
organizationutilizesits resourcesfully;werethe expansionforcedto await
a more abstractanalysisof the resourcesavailable,it mightneverhappen.
As one executivenoted, "It took nine monthsto builda store. Whenwork
began, we would ask ourselveswho we had to run it. Only bums. But as
we got closerto opening, the bums startedto look better."
However,just as the inchwormcannotstretchso far forwardthat it falls
over or is immobilized,so the entrepreneurmust know when and how to
stop sprintingand startto pause. Some pausesare forced by the environment, as in the WorldWar II periodof this study. Othersresultfrom a depletion of resources,as in the pause in the early 1950s. But leadersmust
also sense when "enough is enough" of sprinting.They may have a personal senseof overextension,may be personallyexhausted,may be unable
to keep up with all the changes. Or they may realizethe effects on key
managersin termsof long workinghours, fatigue,frayedtempers,sagging
morale. In any event, the timing of pauses would seem to be criticalin
orderto sustain entrepreneurialsuccess.
Finally,sprintingand pausingmay be seen as a way for an organization
to keep itself energetic. Sprinting provides an inspirationalperiod of
change;pausingprovidesfor the maintenanceand stabilityrequiredto renew energiesso as to be able to acceptonce againthe challengeof change.
Organizations,of course, can plan theirexpansionssystematicallyand so
maintainthem at continuousrates. But the cost of this may be reduced
emotional involvementand commitmenton the part of the employees.
Steinberg'sdiscountstore operationin fact grew steadily(Figure13), but
that growth seems to have been less inspiredand certainlyless ambitious
and successful.
Major Strategy Reorientation
Between the very long cycle of transformation from simple structure to
more formalized structure and the relatively short cycles of sprints and
pauses, major reorientationsof strategycan be identified. These do not
follow any predictablepattern,and in fact seemto occurquiteinfrequently.
494
Academy of Management Journal
September
In Steinberg's,a highly adaptivefirm, there were shifts in strategicbehavior (i.e., a new period in the history),6 times in 57 years, or roughly
once every 10 years. Furthermore,truly major reorientationsof strategy
seem to have taken place only three times: in the early 1930s, the early
1950s,and the early 1960s.The first two werekey turningpoints, the first
a complete rethinkingof strategicorientation-a changed vision by the
leaderand a gestalt shift in strategybecauseof a crisis-and the second,
the removalof a key constraint,which opened the door to a global reorientationalreadyenvisionedby the leader.ThesewereSteinberg'sstrategic
windowsto future successes,and the companywent throughboth at the
right moment. The third reorientation, in the 1960s-the search for
growth in new marketsas the old ones became saturated-was less decisive, less focused, and more difficult.
If strategicreorientationreallydoes take place only once every 10 or 20
yearsin the typicalfirm, then it can hardlybe a continuousconcernof top
management,one perpetuallyon their minds. Yet businessschools train
MBA students by having them analyze two such cases every week, and
businessmanagementengagesin formalstrategicplanningon the assumption of an annualreassessmentof strategywith a rolling5-yeartime horizon. With these long gaps between necessaryreorientations,this annual
reassessmentcan easily become a mechanicalextrapolationof information. That kind of exercise,like "cryingwolf too often," may actuallydesensitizetop managersto strategicissues, so that the need for substantive
changemay not be recognizedwhen it does arise. Conversely,it may encouragechangewhen it is unnecessary-a kind of oversensitivityto strategic issues. Millerand Friesen(1978), in fact, find evidenceof this phenomenon in one of theirarchetypesof strategicbehavior,whichthey call "the
impulsivefirm."
The essentialissueremainsunaddressed:how to be readyfor a majorreorientationwhenit may be reallynecessaryonly once every10 or 20 years;
how to avoid atrophyof the capacityto think strategically,while avoiding
needless"tinkering."Wereperiodsof globalreorientation-even every20
years-surrounded by ones of manifest stability, the problem would be
simplified. But they are not. Organizationsare always changing, but on
different levels of abstractionor inclusiveness.Sam Steinberg'sgenius
seemsto have beenhis abilityto shift mentalgearsfrom one of these levels
to another. After spendingyears in the 1930s and 1940s worryingabout
fluorescentlightingand new ways to packagemeat for self-service,he was
able to shift his thoughtsin the 1950sto the impactof shoppingcenterson
overallretailinghabits.
A strikingaspect of Sam Steinbergand many key managersin the firm
was their apparentability to invest themselvesin a question about the
qualityof a shipmentof strawberrieswith the same passion and commitment as in a questionabout openinga chain of restaurants.The strategy
analyst explicitlydowngradesthe importanceof the former questionsto
focus on the latter,the "big" questions.Somehow,that distinctionseems
1982
Mintzberg and Waters
495
less clear-cut for the managers of this study. Indeed, their thorough involvement in the day-to-day issues (such as the quality of strawberries)
provided the very intimate knowledge that informed their more global vision. That is why analysts may develop plans, but they are unlikely to
come up with visions.
The Entrepreneurial Mode
This study highlights the characteristics of the entrepreneurial mode of
strategy making in the simple structure. The literature characterizes the entrepreneur as the bold decision maker, fully in control, who walks confidently into an uncertain future (Mintzberg, 1973). If anyone fits that description, Sam Steinberg certainly does. That is what gave this organization its spirit, its drive. Even when things looked bleak, the company
"knew"-he knew-that it would bounce back, and that prophecy became self-fulfilling. One is reminded of Starbuck and Hedberg's (1977) description of how Facet turned itself around just because it knew an enthusiastic leadership was taking it over. Mood cannot be discounted as a factor in strategic behavior.
Yet the entrepreneur protects himself in his bold action, controls it, for
successful entrepreneurship is not equivalent to foolhardiness. As noted
earlier, periods of pause, following periods of sprinting, were used to ensure that the organization remained viable. In addition, with a few exceptions that were to prove significant, Sam Steinberg pursued what can be
called a "test-the-water" approach, always sensing an environment with
minor probes before plunging in. In the earlier years at least, Steinberg
never undertook a bold move until it had a pretty good idea what the consequences would be. Of course, such an approach was possible in the
supermarket business. Such stores are built one by one, as independent
units. One does not go to the moon or open an asbestos mine that way.
But then the entrepreneur chooses his business too.
In addition to the notion of controlled boldness, a major characteristic
of the entrepreneurial mode-one repeatedly stressed by Sam Steinberg-is the leader's intimate knowledge of the business. It is intuition
that directs the entrepreneur, intuition based on wisdom-detailed, ingrained, personalized knowledge of the world. In discussing the firm's
competitive advantage, Sam Steinberg remarked: "Nobody knew the grocery business like we did. Everything has to do with your knowledge." He
added:
I knew merchandise, I knew cost, I knew selling, I knew customers, I knew everything
... and I passed on all my knowledge; I kept teaching my people. That's the advantage
we had. They couldn't touch us.
This study shows how effective such knowledge can be when it is concentrated in one individual who (a) is fully in charge (having no need to
convince others with different views and different levels of knowledge,
neither subordinates below nor superiors at some distant headquarters);
496
Academy of Management Journal
September
(b) retainsa strong, long term commitmentto his organization(knowing
that, barringa naturaldisaster,it is he who will be therein the long run);
and (c) possesses the vision and ability to switch from narrow focus to
broad perspective.Underall of these conditions-so long as the business
is simpleand concentratedenoughto be comprehendedin one brain, and
this one was before it diversified-the entrepreneurialmode is powerful,
indeedunexcelled.No other mode of strategymakingcan providethe degree of deliberatenessand of integrationof strategieswith each other and
with the environment.None can provideso clearand completea vision of
direction,yet also allow the flexibilityto elaborateand reworkthat vision.
The conceptionof a novel strategyis an exercisein synthesis,which typically is best carriedout in a single, informedbrain. That is why the entrepreneurialmode is at the centerof the most gloriouscorporatesuccesses.
Embeddedin conventionalthinkingabout strategicplanningis an implicitimageof the strategymakersittingon a pedestal,beingfed aggregate
data that he uses to "formulate" strategiesto be "implemented"by
others. But the history of Steinberg'sbelies that image. It suggeststhat
clear, imaginative,integratedvisions depend on an involvementwith detail, on intimateknowledgeof specifics.As noted earlier,the abilityto be
passionatelyinvolvedat all levels of activityin the businesswas a striking
characteristicof Sam Steinberg.
That this remainedpossible for such a long period of time, even as the
companygrewverylarge, likelyis a reflectionof the simpleand repetitive
natureof this business. The same simple transactionrepeateditself customeraftercustomer,storeafterstore, thousandsof timeseach day. Once
the firm shifted from personalizedto self-service(i.e., impersonalizedservice), then 200 storeswerenot unlike20 so long as they wereconcentrated
in a geographicalarea the leaderknew well.
The personaltouch of the entrepreneurwas criticalto Steinberg'ssuccess. The irony was that it was the very success of the entrepreneurial
mode that renderedit unsuitablein the longerrun. As long as the strategy
maker knew the firm's operationsintimately,the entrepreneurialmode
was effective. It was when the operationsspreadbeyond the comprehension of one man-first to diversifygeographicallyto regionsoutside of its
leader'spersonalknowledge,and then horizontallyto new kindsof retailing-that a shift in the mode of strategymaking became inevitable.No
longercould decisionsbe based on the personalizedvision of one individual, becauseno longercould all the necessaryknowledgebe focusedthere.
Growth and diversification(due to saturationof traditionalmarkets)
necessitatedthe buildingup of a more formalizedstructurewith divided
responsibilitiesand increaseddistancebetweenSam Steinbergand the operations. And so the new mode of strategy making was more decentralized,more analytic, in some ways more careful, but less flexible, less
integrated,less visionary,and, ironically,less deliberate.The controlled
boldnessof the test-the-waterapproachin some cases had to give way to
straightplunges (as in the Grand Union purchase),in others to gradual
1982
Mintzberg and Waters
497
immersion(as in the growthof the discountstore chain), althoughin still
others it remained(as in the MiraclePricing program,so similarto the
WholesaleGroceteriachangesof the 1930s).
Before the shift, strategymakingat Steinberg'scould be characterized
as the interplayof a leaderand an environment,with structurebringingup
the rear(to evoke the inchwormanalogy once again). Here was a leader
very much attuned to the environment,"reacting" to it "proactively"
with the assumptionthat the structurewas lean and flexible enough to
adaptto any changehe made(at least givena periodof pause). In this entrepreneurialmode, structureclearly followed strategy. But over time,
both the environmentand the structurebecamemoredemanding,untilthe
interplayseemedto be increasinglybetweena formalizedstructureand a
constrainingenvironment,with leadershipcaughtin between. Eventually
strategy,to some extent at least, had to follow structure,as well as environment.
ThePlanningMode
Planningseemedto be one elementof that new mode of the 1960sand
1970s.The authorsfeel that the literatureon strategyformationis in great
need of an operationaldefinitionof planning.In other words, a description is neededof what the word actuallymeansin use, not in the abstractions of prescriptivetheory. If planningsimplymeans "futurethinking,"
as impliedby some of its most ardentproponents,then all decisionmaking
is planning,becausea decisionis a commitmentto action, that is, a commitmentto do somethingin the future. By that token, Sam Steinbergwas
alwaysa planner.But, also by that token, to quote Wildavsky,"If planning is everything,maybeit's nothing" (1973). Alternately,if planningis
an exercisecarriedout by people called planners(as opposed to managers), then, in this studyat least, it has littleto do with strategyformation.
Somewherein betweenis the view of planningas the attemptto make and
integratea whole set of decisionsand to articulatethem formallybefore
executingthem.
Overtime, Steinberg'swas drawninto this kind of behavior,not out of
choice but out of necessity,becauseof the demandsof the environment.
The realturningpoint was its initialpublicfinancingin fiscal 1953.The financial community demands plans for its money; the entrepreneurial
mode is unacceptable,at leastuntempered(on paperat a minimum)by the
planningmode. Thereafter,an annualreportforces a companyat least to
go throughthe motions of planningyear after year, and that cannot help
but have an influenceon strategymakingbehavior.
At Steinberg's,planningreally was never strategyformulation;it was
programming.When Steinberg'sdevelopedits first plan in the 1950s, it
was not inventinga strategy. Ratherit was justifying, elaborating,and
makingpublic a strategyit alreadyhad, the one based on its leader'svision. That particularstrategywas conceivedin the entrepreneurialmode,
498
Academy of Management Journal
September
the creativity and synthesis taking place informally and personally. Planning involved the articulation, quantification, and eventual elaboration of
the given vision. The first plan took the shift into shopping centers as
given, and it figured out how many stores would be built, what logistic
support would be necessary, and so on. And, as time went on, the company would be called on increasingly to engage in such planning. For example, the Miracle Pricing program-another vision-required extensive
planning in terms of what prices to cut on what products, what shifts to
make on the advertising budget, and so on. Such planning-as-programming became increasingly necessary as the company grew. Growth made
the company more reliant on public financial offerings, increased the consequences of its strategic moves, and forced it to coordinate more tightly
the efforts of more units in its structure.
A tentative conclusion is that companies plan when they have intended
strategies, not in order to get them. In other words, one plans not a strategy but the consequences of it. Planning gives order to vision, and puts
form on it for the sake of formalized structure and environmental expectation. One can say that planning operationalizes strategy. Although such
planning as programming is not necessary under all conditions, under
some it is mandatory. It may be the only way to pull together the diverse
decisions of large organizations in stable environments and to handle large
and complicated commitments of resources. To draw on another of the
authors' studies, one does not invest almost $100 million in a mine in the
remotest part of Quebec without a great deal of this kind of programming.
But, as noted, there is an effect of planning on vision, for the inevitable
result of programming the entrepreneur's vision is to constrain it. The entrepreneur, by keeping his vision personal, is able to adapt it at will to a
changing environment. By being forced to articulate and program it, he
loses that flexibility. The danger, ultimately, is that the planning mode
forces out the entrepreneurial one; procedure tends to replace vision, so
that strategy making becomes more extrapolation than invention. The
very fact of programming impedes true formulation, changes in degree
driving out changes in kind. In the absence of a vision, planning comes to
extrapolate the status quo, leading at best to marginal changes in current
practice. It is suspected that these two conclusions-planning as the programming of a given strategy rather than the formulation of a new one,
and planning replacing entrepreneurial initiative as an inevitable result of
larger organization and more formalized structure-speak for a good deal
of the behavior known as strategic planning.
In conclusion, this study shows how the success of the entrepreneurial
mode evokes the forces-both in structure and in environment-that
weaken it. Steinberg Inc. at the end of the study period remained in some
ways in the entrepreneurial mode. But the forces to weaken that orientation were growing stronger. In some ways, society benefits from such a result. It gains a surer, more stable and systematic service from its enterprises. But it pays a large price too-less color, less innovation, less
1982
Mintzberg and Waters
499
excitement,less belief in a uniquesenseof identity.Onlyby allowing-and
these days by actuallyencouraging-both modesto exist in differentorganizationscan it reapthe benefits of both worlds.
References
Chandler, A. D. Strategy and structure. Cambridge, Mass.: MIT Press, 1962.
Filley, A. C., House, R. J., & Kerr, S. Managerial process and organizational behavior. Glenview,
Ill.: Scott, Foresman, 1976.
Miller, D., & Friesen, P. Archetypes of strategy formation. Management Science, 1978, 24(9),
921-933.
Miller, D., & Mintzberg, H. The case for configuration. Working paper, Faculty of Management,
McGill University, 1980, submitted for publication.
Mintzberg, H. Strategy making in three modes. California Management Review, 1973, 16(2), 44-53.
Mintzberg, H. Patterns in strategy formation. Management Science, 1978, 24(9), 934-948.
Mintzberg, H. The structuring of organizations. Englewood Cliffs, N.J.: Prentice-Hall, 1979.
Rumelt, R. P. Strategy, structure, and economic performance. Cambridge, Mass.: Division of Research, Graduate School of Business Administration, Harvard University, 1974.
Scott, B. R. The industrial state: Old myths and new realities. Harvard Business Review, 1973, 51(2),
133-148.
Starbuck, W. H., & Hedberg, B. L. T. Saving an organization from a stagnating environment. In H.
B. Thorelli (Ed.), Strategy and structure performance. Bloomington, Ind.: Indiana University
Press, 1977, 249-258.
Wildavsky, A. If planning is everything, maybe it's nothing. Policy Sciences, 1973, 4, 127-153.
Henry Mintzberg is Bronfman Professor of Management Policy, Faculty of Management, McGill University.
James A. Waters is Associate Professor of Organizational Behavior and Policy, Faculty of Management, McGill University.