Division of Research Graduate School of Business.dministration The University of Michigan IMPLEMENTING INTERNAL DIVERSIFIC.TIONS: SETTING THE STRAGETIC CONTEXT Working Paper No. 360 Robert K. Kazanjian The University of Michigan and Robert Drazin Columbia University February 1984 FOR DISCUSSION PURPOSES ONLY None of this material is to be quoted or reproduced without the expressed permission of the Division of Research.

I

IMPLEMENTING INTERNAL DIVERSIFICATIONS: SETTING THE STRATEGIC CONTEXT Robert K. Kazanjian Graduate School of Business Administration The University of Michigan Ann Arbor, Michigan 48109 (313) 764-0122 and Robert Drazin Graduate School of Business Administration Columbia University 721 Uris Hall New York, New York 10027 (212) 280-4431 Pre-publication draft. Not to be quoted without permission of the authors.

IMPLEMENTING INTERNAL DIVERSIFICATIONS: SETTING THE STRATEGIC CONTEXT ABSTRACT Implementation of internal diversification is presented as an increasingly employed strategic option, independent of the degree of relatedness of the proposed new business. Existing frameworks are reviewed and found to be too general to support the specific decision making requirements of implementation settings. An organizational learning perspective of the diversification process is presented. Further, an implementation oriented framework for assessing the degree of relatedness for new business candidates is developed. The degree of relatedness established by the choice of a targeted diversification domain defines the Strategic Context. It is proposed that Strategic Context is a critical construct for understanding the internal diversification process and for guiding specific implementation actions.

INTRODUCTION As a result of concentrated research efforts in recent years across various disciplines, significant advances in the knowledge of diversification strategies have been realized. A range of strategies have been defined and the performance of firms adopting them has been explored (Wrigley, 1970; Rumelt, 1974; Montgomery, 1979). Also, organizing and staffing patterns to support attained strategies have been documented (Galbraith and Nathanson, 1978; Pitts, 1977). While recognizing the significance of these accomplishments, it should be noted that little has been done to advance our understanding of the dynamics associated with diversification. Very little is known about the processes associated with formulating diversification strategies (Fredrickson, 1983), or how to go about implementing a strategy once it has been decided upon (Hrebiniak and Joyce, 1983). The absence of research in this area can be attributed, in part to the lack of an appropriate conceptual framework for dealing with the process issues and organization design choices associated with diversification. It is proposed that conceptualizing diversification as an organizational learning process can significantly advance our understanding by providing such a framework. The implementation of internal diversifications is seen primarily as a process of developing the knowledge necessary to enter and compete in a new domain. The implementation of diversification occurs through the design and administration of transitional organizational forms (Harrigan, 1983) that foster the development of this knowledge. The design of effective forms is contingent upon their fitting with the task requirements imposed by the diversification strategy. Specific organizational design choices must be made in accordance with the degree and type of diversification attempted, so as to balance the generation of new knowledge with the synergistic exploitation of

existing knowledge. The more unrelated the domain chosen, the more knowledge to be developed. The degree of unrelatedness established by the choice of a targeted diversification domain sets the strategic context that guides the learning process. We maintain that strategic context is a critical construct for understanding the internal diversification process and guiding specific implementation actions. The purpose of this pnper is to elaborate on the central role of strategic context in the process of internal diversification. First, a model of internal diversification is presented which describes how strategic context is developed and how it affects subsequent stages in the process. Then, our concept of unrelatedness is developed in more detail and compared to other approaches that have been used. In addition, prescriptions for future research on diversification processes are offered. INTERNAL DIVERSIFICATION AS A STRATEGIC OPTION Present competitive conditions coupled with the restructuring of several major industries has forced many firms to reassess their current strategies, resulting in a renewed emphasis on the topic of product/process/technology innovation and the pursuit of a more diversified business mix. The existence of these and other driving forces is pushing more firms toward a diversified strategy which in some circumstances must be executed internally by direct entry in contrast to entry via acquisition. It would appear that empirical findings argue for such moves only when the firm can capitalize on some relatedness or distinctive competence and therefore against such moves into unrelated domains. More specifically, Rumelt's (1974) milestone study, which offered a categorical scheme to capture the range of business mix strategies, indicates that as a group, the highest

-3 - performing types of firms were in his dominant constrained and related constrained categories —both of which could be described as highly related. Utilizing a different categorization scheme, Nathanson (1982) found that firms in his related-nexus category to be the high performers. In both cases, unrelated diversified firms as a class fared poorly. However, a recent study suggests that the poor performance levels of unrelated diversified firms may not be attributable to the strategy exclusively, but to how it is implemented. Based upon the observation of considerable within group performance variation among unrelated diversified firms, Dundas and Richardson (1982) suggest that management of these firms generally fail to realize the stringent administrative and organizational requirements this strategy demands. Therefore, it is the implementation of this particular strategy that determines its success, not simply its appropriateness to a given situational context. Dundas and Richardson argue in favor of unrelated diversification via acquisition. This is in part due to a range of barriers to entry. Such barriers include required economies of scale, non-scale related cost advantages, product differentiation, capital requirements, access to distribution channels, and government policy (Porter, 1980), which all can pose disadvantages relative to market incumbents for those seeking to enter via internal development leading to direct entry (Yip, 1982). We conclude, however, that a sub-set of firms are either precluded from acquisition or have direct entry as their only feasible course to diversification regardless of the degree of relatedness. Two types of firms come to mind as facing "barriers to acquisition." Based upon sheer size alone, some firms can expect extreme scrutiny from a range of governmental regulatory bodies when contemplating an acquisition of any size. The experience of Exxon

-4 - would be one example. As discussed in the press, (Beman, 1981) their bid to diversify, in part through the acquisition of Reliance Electric Company in the late 1970's, caught the attention and energies of several regulatory agencies. The impact of such restrictions slow negotiations, reduce responsiveness and can raise the price of a potential acquisition; all of which, experience demonstrates, drives such firms away from acquisition. Secondly, high-technology based firms, (for example aerospace companies conducting contract research or production tasks for the Department of Defense) or NASA, who are working at the state-of-the-art in various technologies typically generate numerous business ideas as byproducts of their primary focus. In many instances, these potential new business ideas suggest either the creation of a new market or the redefinition of an existing one with radical new products or technologies. Such firms that develop a proprietary technology for a product or process are often wont to capitalize on it directly if at all, in part to protect their proprietary holdings. Harrigan (1983b) for example, has discussed the reluctance of such firms to enter coalitional arrangements for fear of loss of their proprietary positions. When acquisition is considered as the entry mode in such cases existing firms would at best offer a "shell" within which considerable focus and development would be required by the parent company to develop that new technology. Interestingly, such practices challenge the notion of acquisition versus direct entry as discrete, dichotomous choices, when in fact, they might better be thought of as a continuum. Nonetheless, it is somewhat common practice in the field to suggest that unrelated diversification is typically effected via acquisition. Although this could safely be said of what Rumelt (1974) termed the "acquisitive conglomerates," or what Berg (1973) and Pitts (1976, 1977) term "acquisitive

-5 - diversifiers," Yip (1982) in one of the few empirical assessments of this issue found no support for the relationship of "degree of relatedness" to entry mode (acquisition versus direct) based upon a sample of 59 entrants into new markets. Further, Rumelt's categorization scheme delineated a group of "passive unrelated" diversified firms, which, it can be determined from the classification requirements, developed their business mix through internally generated offerings. His related-linked category represents firms "that had added new activities to old in such a way that they were eventually active in businesses which, considered by themselves, were virtually unrelated" (p. 17). Here again, many of the firms that achieved this category of diversity did so through internally generated, direct entry mechanisms. Therefore, it can be shown that the full range of diversifications, related and unrelated, are attempted and attained through a strategy of internal development. We propose therefore that it is the lack of understanding regarding appropriate implementation requirements —organizational and administrative (Dundas and Richardson, 1982) —which leads to problems with internally executed diversifications, particularly those characterized by modest to high degrees of unrelatedness. This is not to suggest that little is known regarding this topic. In the organization innovation literature, there exists a growing body of work that discusses the role of structure, process, individuals and other variables in facilitating idea generation and development (Zaltman, Duncan and Holbek, 1973; Tushman and Moore, 1982; Galbraith, 1982). More narrowly, Fast (1978), Burgelman (1983), and Ansoff and Brandenberg (1977) have focused on a particular vehicle for diversification — the new venture department. Van de Ven and Delbecq (1974), Duncan (1976) and others have focused equal attention on other structural approaches. However, such research is generally unrelated to any given strategic context,

-6 - especially diversification strategy. One stream of research (Berg, 1973; Pitts, 1974, 1977, 1978) does link diversification strategy to organization, but is not implementation oriented. By investigating the strategy-structure relationships of firms which have already diversified, Berg and Pitts have described the organizational configuration associated with categories of internal diversifers and acquisitive diversifers, but provide little insight as to the structures and procedural mechanisms necessary for the development and execution of diversification activities themselves. We argue that existing studies, which attend almost exclusively to attained strategies, provide little insight to the requirements of the process of diversification as it unfolds over time. Ultimately, implementation actions will require choices regarding various configurations of organizational structure, processes, reward systems and staffing patterns which will position the firm to execute a diversification internally. Unfortunately, the existing literature provides little assistance in directing the process of strategic transition, (Harrigan, 1983a) especially as pertains to diversification via direct entry. DIVERSIFICATION AS AN ORGANIZATIONAL LEARNING PROCESS Diversification is a process by which the firm positions itself to jointly exploit perceived new market opportunities while maintaining or expanding its current domain (Thompson and Strickland, 1983). The central focus of this diversification process then is one of organizational learning, concentrating on the requirements for developing the new knowledge necessary to operate and compete in a new domain. As the existing organization consists of various sub-systems representing types of knowledge or competence (Normann, 1971), organizational structure and systems facilitate the

-7 - learning process and must be consistently matched to each other as well as to the diversification strategy itself. These learnings are mapped onto the organization in the form of either altered or additional structures and processes. More specifically, the role of organization in support of knowledge developed is elaborated in a four stage process model of diversification, depicted in Figure I. The diversification process is evolutionary, emerging from a series of explicit and implicit actions and decisions. Although this process is highly complex, and in many instances best recognized and understood by the firm retrospectively, it is nonetheless suggestive of a set of more or less discrete stages, each focusing on a series of key, problematic activities that have been noted in the innovation and corporate venturing literatures (Fast, 1978; Hlavacek and Thompson, 1977; Tushman and Moore, 1982). — PLACE FIGURE I ABOUT HERE — Prior to presenting the individual components of this model, several observations are in order. For purposes of conceptual clarity, the four major steps or activity foci are presented in a step-wise, sequential fashion. Admittedly, the nature of decision-making under such conditions of uncertainty leads to a much more incremental, iterative style (Quinn, 1980; Hrebiniak and Joyce, 1983) suggesting considerable overlap between steps. This will be addressed in terms of linkages between steps. Secondly, we concur with Burgelman (1983) whose process model study of intra-corporate ventures suggested that "managers from different generic levels in the organization" played key roles at various points (p. 224). Multi-level process characteristics will be raised within each stage as well. Step 1: Diversification Decision. The initiation of any strategic implementation effort is contingent upon the formulation and presentation of

-8 - an explicit and detailed strategy —in this case, the decision to diversify along certain degrees of relatedness. It is assumed that an initial diversification decision targets, at least tentatively, a general domain. As Normann (1977) has suggested, typically two prime driving forces influence the firm in its decision to diversify. For many firms, especially those positioned in one or more mature industries, or those which face the prospect of prolonged environmental uncertainty, dissatisfaction with the current state or anticipated future conditions can be a driving force toward diversification. This obviously must be coupled with adequate capital for investment. A review of the change in the product/ market mix of steel, oil, can, and chemical companies over the past 10 years would offer ample evidence of the strength of such forces and the propensity for action on the part of threatened firms. Numerous examples of internal diversification attempts within these firms can be cited in the drive toward diversification within these and other mature industries. Typically, it is top management, or the dominant coalition (Cyert and March, 1963; Thompson, 1967; Tushman and Romanelli, 1983) who are the prime movers in this scenario. Alternately, driving forces for diversification can emanate from the technical level of the firm as well. In the act of pursuing existing business opportunities, many technology based firms uncover potential product or process advances, which if applied in the market (in which the firm is not yet positioned), would comprise a sizable competitive advantage. Such breakthroughs or opportunities must then, however, be made known to top management either through the hierarchy ("channels") or though some more direct mechanism specially designed. In such a case, a particularly attractive business opportunity on its merits alone might induce management to elect a diversification.

-9 - However, in such idea-rich settings, the competition for resources it typically intense. The prime contrast of these two driving forces is one of specificity. Certain interdependencies in technology development processes for example, may generate byproducts which constitute new business opportunities whose substance and general direction are reasonably clear. Alternatively, a surplus of financial resources in and of itself can constitute a kind of driving force, but neither the content nor the direction of any new business candidates will reveal themselves without very active contribution on the part of management and others (Normann, 1977). Step 2: Search and Idea Generation. New business ideas emerge as a result of search activities associated with the identification, definition and elaboration of specific candidates. These tasks are closely linked with those of Step 1. The nature of activities within this step can take several forms. For the "idea rich" firm, idea generation may be an ongoing activity, but their articulation into more clearly defined business ideas and their publicity at higher levels of the firm cannot be assumed. For such firms, the focus is clearly on a "cure looking for a disease" (Galbraith, 1982). In the case of a firm that has a- less well defined view of the intended diversification or is currently mature and bureaucratic, it must be concerned with fostering an organizational climate conducive to innovation. This would include an acceptance of failure from those willing to try something different and a reward system —formal and informal —that rewards entrepreneurism and innovation. In both instances, the firm might also engage in environmental scanning activities to assess market needs and opportunities, and their fit with internally generated proposals. These examples would be consistent with what Burgelman (1983) has described as autonomous strategic behavior.

-10 - It must be recognized that funding becomes the key instrumentality for activities within this stage. Funding is required primarily for release-time for existing personnel or hiring of additional personnel to engage in required searches —either technical or market. However, beyond its role as a key instrumentality, funding serves as a critical symbol to the organization as to the level of commitment to the diversification decision. Step 1-Step 2 Linkages: Domain Targetting. Ultimately, acceptance of the decision to diversify is inextricably linked to the product of the search and idea generation process, for without viable candidates, diversification may well be put aside, regardless of the driving forces. Further, as has already been discussed, Step 2 (search and idea generation), if conducted, independently at the technical level, may very well trigger an inquiry at the executive level into the advisability of diversification. Thus, it is of little importance whether Step 1 leads to Step 2, or the reverse, given the highly iterative nature of these activities. Overall, domain targeting —the identification of potential new, defined product/market niches —becomes the end product of the cycle or cycles through these two steps. Step 3: Selection, Review and Development. Once an idea has been generated and defined adequately enough to be understood, it must undergo a selection process since in most situations there are far more candidate new businesses than resources to fund them (Booz-Allen Study, 1982). Therefore, top management of the firm must ultimately back one or a few candidates over many others. This is not to suggest that the firm makes a one time "go-no go" decision. The review, and by extension the selection, is ongoing into the development process of the key elements of the new business —usually some product or service offering.

-11 - A number of authors have suggested that there exists a pattern to the development of a product or business (Galbraith, 1982). Focusing on the product or technology development itself, these would typically include a feasibility study, establishment of proof of principle, development of a single prototype, development of multiple prototypes, execution of a trial production run and other activities, all run by the technology sub-systems of the organization —engineering and R&D. Correspondingly, parallel efforts are required in critical functional areas such as marketing and manufacturing. Of course, these parallel activities must be closely coordinated through project management techniques extensively elaborated elsewhere. It is important to note that there are limitations to the rigidity and formality of the development process. It must be tailored to the "natural milestones" that suggest themselves according to the situation. Early on, reviews are typically informal, brief, and revolve around the expenditure of relative small dollar amounts. They are spaced according to the nature of the technology at hand and the difficulty associated with passing each successive milestone. As the development process advances, greater formalism, regularity and deliberation typically emerges. Interestingly, it can be the earliest decisions, many times made by lower level managers, that have the greatest impact on the nature of the ultimate product/business. Step 2-Step 3 Linkages: Implementation and Domain Focusing. Selection decisions regarding new business projects must be viewed as contingent upon progress towards the next developmental milestone. In fact, reviews at each new milestone comprise an opportunity for another "go-no go" decision, right up to full scale commercialization. Obviously, if the firm is committed to attaining a diversified strategy, the death of a candidate business would cycle activity back through stage 2 for a renewed search. However, in many

-12 - instances outside the wholescale abandonment of a project, the final product/ offering is somewhat removed from the original idea. Through the review and development process, the product itself may become redefined or repositioned based upon technical break-throughs or problems. This requires additional searches of market and other environmental factors to ascertain the advisability of this redefinition. Alternatively, continued environmental scanning may uncover market opportunities which would force a rethinking of the original strategy for the new business. This iterative cycling between steps 2 and 3 then provides essential focusing and detailed definition of the new business as a fit of market opportunities, product offerings, and internal administrative support systems. Step 4: Commercialization. The culmination of the diversification process is the commercialization of a new offering. Here again, commercialization is not a state to be distinguished in a dichotomous fashion from noncommercialized. The innovation literature, in its description of this process, discusses trial implementation as preceding continued, sustained implementation. Other, more marketing-related studies, advise for the early stages of commercialization to be linked to the development process. So, pilot introductions of products or services, first to employees as an internal consumer testing group, and then targeted product introduction to a limited geographic region, provide valuable information that could lead to a revision of some product design or marketing approaches. Step 3-4 Linkages: Incremental Modification. In addition to the fine tuning of the commercialization process discussed above, it is important to note that once positioned with a new domain having attained the diversification, it becomes necessary to continue to remain sensitive to market demands and technological advances with modified product variations or new product

-13 - lines. This will require a structured capability to refine the firm's offering with additional development work conducted in Step 3. Obviously, the introduction of a new line of products within the same business would require a cycling back through Step 2 as well. Here again, ongoing research and idea generation could suggest the introduction of a new line as well. ASSESSING DEGREE OF RELATEDNESS The diversification process serves as a major, but not solitary, guiding force for implementation activities. A second is the type of diversification to be attempted. This section will expand upon the latter element, especially with respect to the degree of relatedness of diversification candidates to existing business, and the associated requirements and implications for knowledge development. Existing diversification studies primarily are focused on assessing the degree of diversification attained, towards which numerous approaches have been employed. Pitts and Hopkins (1982) have conducted an extensive literature review and summary on this topic. As they suggest, "the first tasks facing a researcher wishing to measure a firm's diversity therefore, is to identify its individual businesses." (p. 620). In this review of strategic diversity and its modes of operationalization, Pitts and Hopkins cite three primary approaches. The first, resource independence, sees a business as discrete from others of the corporation if the "resources involved are separate from those supporting the firm's other activities." (p. 621). The least employed approach, due to data collection difficulties, defines businesses in terms of market discreteness. Finally, businesses can be defined in terms of product differences, viewing each product offering as a separate business.

-14 - This latter approach is the basis of the often employed Standard Industrial Classification system. Accordingly, having identified a firm's business by one of these methods, the researcher is then in a position to assess the firm's overall level of diversity. Pitts and Hopkins here note two primary approaches to the measurement of diversity. The first is based upon the number of businesses in which the firm is positioned as measured by an absolute count, some ratio that accounts for relative size of businesses, or other related indices. The second is termed strategic and assesses diversity by either the relatedness of various businesses or the firm's historical growth pattern. Montgomery (1982) sees the two prime modes of assessing diversity as categorization schemes and continuous SIC classifications. Pitts and Hopkins seem to suggest that the definition of the business and the assessment of diversity can be done independently. Undoubtedly, operationalization of these constructs for research purposes would tend to constrain the use of some measures while reinforcing the use of others. It is proposed here that when relatedness is employed as the diversity measure, a tighter coupling should exist between the two. That iq, the primary defining elements or dimensions of a business should logically serve as the prime determinant of diversity. The most commonly utilized operationalization of firm diversity, employed by Rumelt (1974) and others (see Montgomery, 1982), is the relatedness measure. Building upon a resource independence definition of discrete businesses, Rumelt developed, as a variation of Wrigley's (1970) scheme, a typology that placed firms into four primary categories —single business, dominant business, related business, and unrelated business (all but single business having subcategories within) —according to the degree of strategic interdependence

-15 -across businesses as well as "the proportion of a firm's revenues that can be attributed to its largest single business in a given year." (p. 14). Although such a categorization scheme can effectively depict the firm's achieved business strategy, assuming the availability of adequate and precise information allowing for the accurate categorization of each firm under study, it does so based upon aggregate business level assessments which provide little detail at a more operational or functional level. In contrast, it is the coupling of business level strategic decisions with operational level instrumentalities that must be understood for purposes of effective implementation actions directed at a proposed strategy. It is our contention that within the diversification process, domain targeting initiates a shared vision among key actors of some new business idea (Normann, 1977; Galbraith, 1982) inherent in which are some prime characteristics —typically the product or service to be offered, its potential market, and the necessary administrative systems and support mechanisms required. As the firm is currently oriented to other domains or product/ market niches, each of these elements of the new business idea potentially represents requirements for new knowledge, skills and expertise to the firm. Therefore, the firm will need to develop other competencies required to compete in the new business. These requirements for new knowledge begin as cognitive, perceptual assessments of what the firm must learn in order to implement the new business idea. The requirements for new knowledge become the task focus of the search and idea generating (Step 2), and selection, review and development (Step 3) phases of the diversification process. We believe, then, following Cooper (1977), Fusfeld (1978), and others that relatedness cannot be understood at the total organizational level, but

-16 - rather requires analysis at the functional level. Ultimately, new knowledge generators can be seen to translate into the primary functions of a firm, broadly defined as: 1) marketing (market research, sales, promotion, customer service); 2) product technology (engineering and research and development); and 3) process technology (manufacturing, materials, quality). These three primary functions also capture two approaches to business definition noted by Pitts and Hopkins (1982) —market discreteness and product discreteness. Any specific internal diversification strategy then can be categorized in terms of the requirements for knowledge generation and the degree of relatedness along each of those three dimensions. Therefore, from an implementation perspective, relatedness is defined here as the relative distance between the knowledge needed to implement the new domain and the degree of knowledge available in the current domain. Unlike the categorizational schemes discussed earlier, relatedness here refers to the vision implied in some desired future state, rather than a realized or achieved degree of diversification. Thus, relatedness is directly proportional to the amount of knowledge to be generated during the process of diversification (Normann, 1977, p. 86). As Figure II indicates, the intended diversification can be portrayed in terms of relatedness vectors of product technology, market, and process technology knowledge dimensions. High unrelatedness comprises a major reorientation by the firm, while high relatedness comprises the introduction of modifications and improvements close to the firm's current line of business (Normann, 1977). The configuration of relatedness implied by the diversification decision will vary from situation to situation. For example, diversification opportunity A, presented in Figure II, represents a highly unrelated candidate, which would position the firm into a business with highly unrelated customers, products and production processes.

-17 - Exxon's diversification in office products systems would be an example of such a situation. Diversification oportunity B can be seen as being highly related on one dimension but highly unrelated on the remaining two. The diversification of an ethical drug firm into computer based patient diagnostic equipment or drug dispensing body implants —representing a position in the same market but with new products and production technologies —would be another example. Obviously, any of the 25 other combinations of dimensions could be postulated with supporting examples of firms who have pursued such diversifications. — PLACE FIGURE II ABOUT HERE — Theoretical and empirical support can be cited to reinforce this view of relatedness. In what might be termed a market discreteness approach to business definition, Abell (1980) proposed defining a business according to the key underlying dimensions, in this case of the market. His dimensions were customer uses/functions, customer groups, and alternative technologies (Thompson and Strickland, 1983). This approach, while undoubtedly well suited to a planning perspective, still suffers from a narrow operational orientation that does not adequately aid implementation analyses and developmental decision making. However, direct support for the functional level assessment of relatedness can be found in the work of Yip (1982) who operationalized relatedness in terms of specific activities shared between the new business and the existing parent firm. His seven types of activities, which are manufacturing/production, research and development, distribution, sales, advertising and promotion, immediate customers, and end users, clearly fall within the three prime dimensions of this model.

-18 - THE STRATEGIC CONTEXT AS A CONTINGENCY FACTOR FOR IMPLEMENTATION ACTIONS Two prime components define the firm's Strategic Context for diversification activities. The first emerges from the diversification process itself which is seen as one of organizational learning, whereby the firm must first target a domain for new business activity as well as institutionalize new knowledge required to compete effectively in that targeted domain. Secondly and of equal importance is the type and nature of diversification attempted, as indicated by the relatedness dimensions discussed above. The Strategic Context of internal diversifications is therefore the product of the tasks associated with the various steps of diversification and the degree of relatedness of the new business to the existing business activity of the firm. It is a broad boundary within which specific new businesses are defined and developed. Further, it defines in general, the type and extent of new knowledge to be developed. Utilizing the foundation of this integrative view, the Strategic Context is proposed as the primary contingency factor for diversification actions of the firm. First, both idea generation, and the selection, review and development process require suitable organizational configurations designed to support their primary task orientations. Further, these organizational support mechanisms must assume forms not necessarily common to those of the operating parent firm and will in fact require appropriate oversight, as they should be altered as the diversification process moves forward. More specifically, it can be argued that choices regarding organization structure must be matched to the Strategic Context. For example, increasing levels of structural differentiation provide for greater capacity to generate non routine knowledge (Duncan, 1976). The type of knowledge generating

-19 - structure adopted should reflect the differences in capacity demanded by the targeted domain and its implied degree of relatedness. Organizations that do not sufficiently differentiate are likely to develop insufficient numbers of new business proposals or proposals that do not adequately meet the requirements of a specific new business opportunity. Organizations that over differentiate are likely to develop proposals or business candidates that are seen as too radical and which have a low probability of acceptance (Ansoff and Brandenberg, 1978; Fast, 1978). In addition to structuring for search and idea generation, management must choose among multiple new business candidates and review their progress over time. This requires the installation of appropriate coordinating and integrating mechanisms. Under conditions of related diversification, the major coordinating problems are likely to be the smooth transfer of information between the functional areas, and the priority management of shared, scarce resources between the old and new domains. Under conditions of unrelatedness, integration issues between the current organization and the differentiated subunit are likely to shift away from the coordination of scarce resources and toward issues of legitimacy, rights of review, control and power. Research regarding these choices are obviously problematic and complex, requiring considerable specificity and rigor. Unfortunately, much of the literature to date on this topic has been largely general and not implementation oriented. Subsequent research by the authors encompasses a literature review and a detailed investigation of choices regarding organization structure and systems for diversification (see Drazin and Kazanjian, 1984). Examination of appropriate human resource and reward system configurations consistent with Strategic Context are intended to follow.

-20 - CONCLUSIONS AND IMPLICATIONS FOR FUTURE RESEARCH It is postulated that successful diversification will result for firms which match transitional organizational forms to their Strategic Context. However, continued research on this topic must address a number of issues, both theoretical and operational. Regarding the determinants of Strategic Context, both the diversification process and the dimensions of relatedness frameworks should be field tested with firms which have attained internal diversification. The diversification process as proposed is punctuated into four activity steps with iterations between steps. This may need to be refined with more but narrower activity steps to enhance its applicability to implementation settings. Similarly, it must be assessed if functional level of relatedness provides sufficient detail, or if additional dimensions should be included. Given a valid and supportable base for the Strategic Context, a number of questions can be raised about its application. As discussed above, research is ongoing by the authors regarding its relation to traditional organizational structure and systems. However, as Strategic Context has been presented as a perceptual construct, must there be mutual agreement among managers as to what the specific Strategic Context is for a specific diversification activity? If not, then how are implementation actions and choices guided, coordinated, and integrated across the organization? Resolution of this issue would effect research design choices, specifically regarding the number of respondents in data collection from each firm. Admittedly, this approach to relatedness, knowledge-based and perceptual in character, will require different data collection and analysis methods from those used to date. Assessments from key players inside the firm will be

-21 - required as the prime source of data, in contrast to the use of second source objective data. Categories or groupings of firms according to relatedness might then be extracted from the data using multivariate analytic techniques such as mutli-dimensional scaling. Finally, what utility does the Strategic Context have for the diversified firm trying to diversify further? What modifications should be made to the existing framework in such situations? Although continued research in this area is problematic and complex, it is nonetheless timely and important, as major sectors of the economy look to restructure and reposition, at least in part, through internal diversification. The success of such efforts will depend upon a number of choices. In addition to attending to financial, technical, and market characteristics of a new business opportunity, top management must be equally attuned to organization design configurations and administrative issues. Viewing internal diversification as an organizational learning process conducted within a Strategic Context bounded by a newly targeted domain and the degree of relatedness of the specific candidate, can advance our understanding in this area.

-22 -FIGURE I DIVERSIFICATION PROCESS Current Strategy Strategic Driving Forces ~ Existing Product/Market Mix ~ Existing Knowledge Systems * Industry Maturity. Resource Availability ~ Technology Push

-23 - FIGURE II DEGREE OF UNRELATEDNESS ON THREE PRIME FUNCTIONAL DIMENSIONS "A" H MARKET I i M L ________ ____ __ ________ _ _____ _______ "B" H M PRODUCT TECHNOLOGY L L M PROCESS TECHNOLOGY H

-24 - BIBLIOGRAPHY Abell, D. F., Defining The Business: The Starting Point of Strategic Planning, Prentice-Hall, Inc., Englewood Cliffs, N.J., 1980 pp. 169. Argyris, C. and D. A. Schon. Organizational Learning, Addison-Wesley, Reading, Mass., 1978. Beman, Lewis, "Exxon's $600-Million Mistake," Fortune, October 19, 1981, pp. 68-93. Burgelman, R. A., "A Process Model of Internal Corporate Venturing in the Diversified Major Firm," Administrative Science Quarterly, 28, 1983 pp. 223-244. Cooper, R. C., "Strategic Planning for Successful Technological Innovation," Business Quarterly, Spring, 1978, pp. 46-54. Drazin, Robert and Robert K. Kazanjian, "Organizing for Internal Diversification," Working Paper, Columbia University, 1984. Duncan, R. D., "The Ambidextrous Organization: Designing Dual Structures for Innovation," Kilman, R. H., Pondy, L. R. and D. P. Slevin (editors) in Vol. I: The Management of Organization Design, New York: North Holland, 1976. Cyert, R. M., and J. G. March. A Behavioral Theory of the Firm, PrenticeHall, Inc., Englewood Cliffs, N.J., 1963. Dundas, K. N. M., and P. R. Richardson. "Implementing the Unrelated Product Strategy," Strategic Management Journal, 3, 1982, pp. 287-301. Fast, N. D., The Rise and Fall of Corporate New Venture Divisions, UMI Research Press, Ann Arbor, Mich., 1978. Fredrickson, J. W., "Strategic Process Research: Questions and Recommendations," Academy of Management Review, Vol. 8, 1983, pp. 565-575. Fusfeld, A. R., "Critical Functions: The Key to Managing Teamwork in the Innovation Process" (paper presented at Innovation Canada, 1976). Galbraith, Jay R., "Designing the Innovating Organization," Organizational Dynamics, 10, 1982, pp. 5-25. Galbraith, Jay R., and Daniel A. Nathanson. Strategy Implementation: The Role of Structure and Process, West Publishing Company, St. Paul, Minn., 1978. Harrigan, K. R., "Preparing the Firm for Transformational Strategies," in Advances in Strategic Management, Vol. I, JAI Press, Inc., 1983a, pp. 133-142.

-25 -Harrigan, K. R., "Strategies for Joint Ventures," Working Paper, Strategy Center, Columbia University, 1983b. Hlavacek, J. D. and V. A. Thompson, "Bureaucracy and New Product Innovation," Academy of Management Journal, Vol. 18, 1975, pp. 784-810. Hofer, C. W., "Toward a Contingency Theory of Business Strategy," Academy of Management Journal, 18, 1975, pp. 784-810. Hofer, C. W., and D. Schendel. Strategy Formulation: Analytical Concepts, West, St. Paul, 1978. Leontiades, M. Strategies for Diversification and Change, Little, Brown, and Company, Boston, 1980. March, J. G., and H. A. Simon, Organizations. John Wiley and Sons, New York, 1958. Miles, R. E., and C. C. Snow, Organizational Strategy, Structure, and Process, McGraw-Hill, New York, 1978. Mintzberg, H. The Structuring of Organizations, Prentice-Hall, Englewood Cliffs, N.J., 1979. Montgomery, C. "Diversification, market structure and firm performance: an extension of Rumelt's model," Ph.D. Dissertation, Purdue University, 1979. Montgomery, C. A., "The Measurement of Firm Diversification: Some New Empirical Evidence," Academy of Management Journal, 25, 1982, pp. 299-307. Nathanson, D. A., and J. S. Cassano. "Organization, Diversity, and Performance," Wharton, 6, 1982, pp. 18-26. Normann, R. Management for Growth, Wiley, New York, 1977. Normann, R., "Organizational Innovativeness: Product Variation and Reorientation." Administrative Science Quarterly, June, 1971, pp. 203-15. Pitts, R. A., "Diversification Strategies and Organizational Policies of Large Diversified Firms," Journal of Economics and Business, 1976, pp. 181-88. Pitts, R. A. "Strategies and structures for diversification," Academy of Management Journal, 20, 1977, pp. 197-208. Pitts, R. A., and H. D. Hopkins, "Firm Diversity: Conceptualization and Measurement," Academy of Management Journal, 7, 1982, pp. 620-629. Porter, M. E. Competitive Strategy. The Free Press, New York, 1980.

-26 - Quinn, J. B. Strategies for Change: Logical Incrementalism, Richard D. Irwin, Inc., 1980. Rumelt, R. P. Strategy, Structure, and Economic Performance, Harvard University Press, Cambridge, Mass., 1974. Thompson, A. A., and A. J. Strickland. Strategy Formulation and Implementation, Business Publications, Inc., Piano, TX., 1983. Thompson, J. Organizations in Action, McGraw-Hill, New York, 1967. Tushman, M. L. and W. L. Moore, Readings in the Management of Innovation, Pitman Publishing, Marshfield, Mass., 1982. Van de Ven, A. H. and A. L. Delbecq "A Task Contingent Model of Work-Unit Structure" Administrative Science Quarterly, 19, 1974, pp. 183-97. Williamson, 0. E. Markets and Hierarchies: Analysis and Antitrust Implications, The Free Press, New York, 1975. Wrigley, Leonard. "Divisional autonomy and diversification," Unpublished Ph.D. Dissertation, Harvard Graduate School of Business, 1970. Yip, G. S., "Diversification Entry: Internal Development Versus Acquisition," Strategic Management Journal, 3, 1982, pp. 331-345. Zaltman, G., R. Duncan, and J. Holbek, Innovations and Organizations, Wiley, New York, 1973.