Project management for intensive innovation based strategies
new challenges for the 21st century
Christophe Midler, Directeur de Recherche, Centre de Recherche en Gestion de l’Ecole polytechnique,
The issue of growth by innovation is central to current dynamics in the industrialized nations. As Navarre (1992) puts it, we have moved on from “the battle to produce better to the battle to design better.” In markets which are saturated, as European markets typically are, gaining competitive edge means differentiating and focusing products with increasing precision (“niche” strategies),by means of more frequent product replacement to anticipate market trends, by reacting before others to the often unpredictable signs given out by fluctuating markets, by integrating as quickly as possible technological innovations which make cost savings possible, the better to withstand price warfare in increasingly globalized markets.
Thus, during the 90s we have witnessed a number of changes in various sectors, some radical, in the processes whereby new products are designed. The concept of project management has deeply reshaped firms’ inside organizations as the overhaul of sub-contracting systems. In our researches we have analyzed how this “projectisation” process (Midler, 1995; Jouini & Midler, 2000) developed in various industrial contexts: auto industry, construction, chemical and pharmaceutical firms.
But, on the other side, such enlargement in project management application scope, combined with greater expectations for efficiency in project results call for deeply and rapidly revisiting the traditional project models. The objective of the present paper is to characterize these dynamics during the 90s in the French context and point out the main challenges that need to be faced in the following century.
This paper is based on a research program, which has been ongoing since the beginning of the 90s, conducted by a team at the Management Research Center of the Ecole polytechnique. The approach adopted has been that of interactive research with manufacturing companies over a long period of time (usually three to five years), against the background of the problematics of contributions to experimentation and the acquisition of knowledge on the dynamics at work in French companies. This kind of in-depth enquiry has been, or is currently being conducted in the armaments, automotive, construction, chemical, electronics, power, pharmaceutical and steel industries. A common theoretical matrix for analysis has been constructed, and regular exchanges of views make it possible to compare the dynamics observed in these different cases.
The general hypothesis behind this research program is a contingency vision of project management models (Lundin, 1998), due to the specificity of the activity (with variables such as risk, complexity and size of the project) and of the social context where the project is embedded. In that perspective, the development of the project management field appears as a combination of contingent organizational and instrumental knowledge creation processes (for example, the PERT method for the Polaris program for U.S. Defense in the beginning of the 60s) with formalization of de-contextualized bodies of knowledge (typically, the PMBOK® Guide), intersectorial dissemination via institutional processes (professional associations, education systems, normalization and regulatories authorities, …) and the hybridization of these outside contributions with the traditions already in place. The research program is designed to focus on emblematic cases, where implementation of existing best practices is problematic because of the singularity and novelty of the situation.
After characterizing the projects context of intensive innovation-based strategies in a first part, we will analyze the consequences on project coordination processes.
Exhibit 1.The Three Components of Design Systems in Innovation-Based Strategies
Source: Ben Mammoud-Jouini 1998
Exhibit 2.The Deployment of the Design System Across Projects and Non-Project Areas
Source : Ben Mammoud-Jouini 1998.
After focusing on product development processes, the ongoing “design revolution” reaches a new and more radical step in the 1990s. The performance achieved by development programs, as measured by the yardsticks of quality, cost and lead-times, is not sufficient to ensure competitiveness if the resulting products are not genuinely and radically innovative. K. Eisenhardt and B. Tabrizi (1995) and Brown (1997) found, when looking at the information technology sector, a new model for the innovative enterprise, which explodes the traditional dichotomy between continuous, marginal innovation on the one hand, and rare, singular breakthroughs on the other. In this context, the boundary line between research and development becomes less clear. A similar statement can be made in the rather different contexts of mature industries like automobile (Weil, 1999) or household equipment industry (Chapel, 1996). In upstream industries like chemical and steel, the importance of innovation in French firms’ strategies grows also considerably in the 1990s (Charue-Duboc, 1998; Lenfle,2000), by the effect of two combined causes: on the customers’ side, the accelerating pace of the more or less radical (re)design of end products downstream is increasing the opportunities for the substitution of raw materials as part of the never-ending search for better compromises; on the competitors’ side, firms from emerging countries get more and more aggressive on commodities markets as they catch up the occidental technological know-how.
Hatchuel (1998) formalized the principles of the “Design Oriented Organization,” which associate knowledge and product development management. On the same line, we (Ben Mammoud-Jouini, 1998) have identified the three processes which are at the heart of intensive innovation based strategies (see Exhibit 1).
• The company’s strategy: How does it integrate innovation into its strategy? Is innovation a main, or a secondary lever for action? Can signs of the dynamics defining the priority granted innovation be seen in changes in budgets allocated to it? How does the company guide the key processes for product innovation defined below? Are explicit structures in place for the management of portfolios of groups of skills and projects? Etc.
• The process whereby key skills are developed for input into the innovation process: not only research, but also learning processes within product development programs.
• The process whereby those skills are coordinated in product development management.
Project management is of course a key component of this design system. Exhibit 2 shows how advanced project management contributes to this global design system of the firm, as it was introduced in the firms in the 80s and beginning of the 90s for developing new products (Clark, 1992; Midler, 1993).
But the transition is still uncompleted with the two other levels (1 and 2). The importance of knowledge management is increasingly more evident as competitive advantage in product markets comes from more and more radically innovative products, development projects now being a shared competency among firms. At the same time, reducing lead-time and costs in development projects creates a situation where radical innovation and learning is rarely accepted by project managers, because of the resulting risks. Traditionally, depending on their activity, firms developed more on a knowledge-based pattern (typically Pharmaceutical and Chemical sector) or on the product development side (automobile). What is new now is the necessity to combine the two processes to succeed in creating a track of successive radical and valuable innovations.
How can we make project management fit into these new challenges? We will focus on four points that call for deep revisions in the classical project management paradigm:
• Managing learning and solidarity within risky projects
• From demand-pull project to Design-push projects
• From product-projects to platform or knowledge projects
• Managing the learning externalities in project portfolios.
Beyond Contract/Process Dilemma: Management of Convergent Commitment, Collective Learning, and Solidarity Between the Project Members
The first immediate consequence of the competition by innovation context is to increase the risk level of projects. Uncertain projects are of course not new in industry. But what is more original is to consider such projects as a normal and repetitive way to sustain survival, and, if possible, growth of the firm. Therefore, the level of efficiency required for risky projects has significantly increased in terms of functional performances, quality, costs and lead-time. The importance of risk-management as a key point in P.M. recent conferences is a significant reflect of such a trend.
In that context, uncertainty creates many surprises that implies the need for revisions in means and ends of the project. Therefore, the efficiency in such projects depends on:
• First, the efficient collective learning processes, that will permit the rapid detection of the problems, and generate satisfactory answers
• Second, the solidarity between the contributors, in order to reallocate the resources, constraints and performances among the contributors, for the best of the global project.
The classical contractual and standardized coordination model is not adapted to such priorities (Lundin, 1995; Söderlund, 1998). In that model, the targets are precisely specified and contractualized in advance with each contributor. The coordination logic is the fulfillment of these contracts. Revising the initial contracts in order to face unpredicted events leads to difficult and often little productive debates, because the solution-finding target is put behind the responsibility attribution question. Moreover, the contractual system is a significant obstacle to the creation of collective solidarity on the global objectives of the project: for contributors not directly confronted with, a problem appearing to another contributor (a delay for example) is often taken as a good opportunity to elude their own difficulties …
Modern concurrent-engineering literature insists on the importance of process coordination to ensure project efficiency. Characteristics as co-location, mutual understanding, trust and leadership styles in project teams, problem-solving methodologies and reactive decision processes,…appear as variables of great importance in risky projects, compared to formal initial detailed contracting. In the case of important projects (such as automobile development), far from neglecting the contractual side, this emphasis on collective design processes has experienced a growing interest in renewed contracting philosophy, to coordinate the contributors inside the firm (internal contracts as studied by Naklha and Soler, (1998)) as well as to regulate inter-firm relationships within a partnership relation (Garel Midler, 2000). The role and the form of the contracts are different than in the market-oriented regulation. In the traditional project regulations, contracts are supposed to ensure, by their fulfillment, the coordination of the project. They are supposed to be the “invisible hand” of the project. Adjusted penalties in case of failure are supposed to be the effective defense against the risk of opportunistic behavior and incompetence. In the renewed philosophy, contracts have no such ambition. They appear not as a substitute to the coordination process, but more modestly, as a component of it. In particular, to solve the classical problem of trusting the contributors, other certification processes are called upon. Their specific functions are:
• In the beginning, to force the actors of the project to make their expectations and priorities for the project explicit
• To formalize, confront and share the strategies and hypothesis that underlie the commitment of the contributors
• To negotiate incentives that are coherent with these priorities and strategies
• To define collective and accepted “warning processes” that will serve as the basis of the collective learning process when surprises occur.
Such an approach of contracting within projects is similar to modern conceptions in strategic planning of firms (Ponssard, 1992). On the contrary, it contradicts in many ways the juridical approach, where the contract is conceived as an instrument to solve contentious business, whereas it is here mainly an instrument to make the project converge and prevent legal dispute.
From "Demand-Pull" Projects to "Design-Push" Projects
The classical project paradigm is a demand-pull design model. It organizes the mobilization of professional to answer explicit demands from project owners. But in our specific economic context it is more a design push process than a market pull one. Innovations like the Walk-man by Sony, Post-it by 3M, or Navigator by Netscape were not designed as an answer to an explicit question of existing customers. Generally, innovation strategies are more proactive to market signals than reactive. Thus, implementation of the traditional PM model is problematic in cases of radical break through and emerging phases of innovation. In that situation, the innovation process includes the elaboration of the demand or question, at the same time as the definition of the answer.
Taking the concepts from Simon (1969) and Schön (1983), prominent theorists in the field of design activity, we could say that the “problem setting” side of the design activity is as much (and sometimes even more) important than the “problem solving” side. (The Post-it case is typical of an extreme situation. The key question was to find a valuable use for a paste that already existed. In how words, the answer existed, what was to be invented was the question that fitted with it …) Therefore, the asymmetry of the economic relation between the owner-customer and the contractor-supplier creates a bias to the need of a symmetric and dialectic exploration of setting and solving of the problem.
What are the consequences of such theoretical arguments on practical structuring of R&D projects in firms? The evolution of the French Chemical group Rhône-Poulenc can illustrate this point (Charue-Duboc, 1997). Until the 80s, R&D was clearly in a science and technology push logic, with central R&D departments. In the beginning of the 90s, the reorientation of the core chemical industry group towards production of innovative high utilization value chemicals lead to a change in the type of projects conducted: from projects involving heavy industrial production toward product-focused research.
In order to implement this new strategy, a matrix structure with Strategic Business Units and a decentralization move of R&D was implemented. The central share of R&D budgets is gradually decline, as budgets allocated by sector or SBU rise. The philosophy was to “focus research on market” by adopting the demand-pull traditional project model, with marketing people as project owners, and research people as contractors. The limits of such a model for breakthrough projects appeared rapidly. Marketing people from SBUs were more focussed on incremental than radical innovations. Many interesting innovative ideas from the researchers could hardly be financed in this situation because the innovation did not fit in with the existing market segmentation and responsibilities (generally, the market gets restructured as a result of breakthrough innovation). Marketing people did not have the skills to help the exploration of new valuable functionalities.
Exhibit 3. Evolution in Customers System From Linear to Concurrent Approach
A new step was implemented to overcome these problems. New research units were created, not in order to synthesize products, but to analyze functional and utilization characteristics (application, and “applicability” laboratories). The project patterns moved from a demand-pull contractual model to a more “design-push” model, which emphasized the role of applicability researchers, also adopted a more symmetrical and integrated vision of the marketing and research part of the project. Let us insist on the fact that this move is not a come back to the science push logic of the 60s, but a new model fitted to meet the modern proactive strategies of innovative features and technologies.
This “design push” project model spread beyond the firm’s frontier by the setting of exploration partnerships with customers. The goal was to succeed in mobilizing customers’ expertise for the exploration of potential utilization values and the final development of new products. We are moving here from a linear view of the downstream value chain towards “customer system concurrent engineering” as represented in Exhibit 3.
Setting up design-push partnerships is more difficult than demand-pull ones. In the manufacturing sector, the initiative for new design approaches comes from the end firms in the chain. They are able to compel their suppliers to use a concurrent engineering approach by wielding their power as buyers. The situation is obviously different for upstream firms, as projects usually require coordinating downstream customers or specifiers rather than upstream suppliers. How can customers be convinced to take part, when, as is usually the case, it is not possible to exercise financial leverage over the downstream part of the chain?
From Product-Projects to Platform and Knowledge Projects
The third consequence of the ongoing “design revolution” we will discuss is the re-focussing on research and preliminary project design after the focus in the 90s on product development per se (Ciavaldini, 1996; Cusumano, 1998; Midler, 1995). In line with the argument developed in the precedent paragraph, the idea is to develop an innovative “generic solution,”“proven concepts,”“platform investments” (Kesseler, 1998) or a “semi-product” (Weil, 1999) that will be finalized and incorporated in a family of future product developments.
Managing such projects raises difficult and new problems because of the characteristics of the nature of the target (Lenfle,2000):
• One of the major levels for mobilization in project management is the clear identification of the goal that must be reached at the end of the project. In platform projects, the definition of the result is abstract and fuzzy, compared to the “clear” reality of the launching of a new product. In a way, such a project is by definition not finished. An excessive finalization raises the risk of a very contingent solution, which would be contradictory to the objective of extrapolating the concept to a large range of different products and markets.
• Another difficult problem is the long term nature of such predevelopment projects. This type of project often faces what we call a “hidden urgency” (Lenfle, 2000): the effective commercial launching of derived products is far ahead in time: often more than 10 years for an innovative component or technology in the car business; more than 12 years for a breakthrough concept in pharmaceutics. But irreversibility is created long before, at the beginning of the product development where matured solutions are required and patent taken.
Exhibit 4.The R&D Process: From Conventional Funnel Metaphor to “Porous” Funnel
• The evaluation of importance of such platform projects also suffers a decision-modeling conceptual problem. Their value comes from an option-value reasoning (Kogut, 1994), which is less developed and formalized in firms than the classical return on investment for product projects. Many economic works on real option themes insist on the systematic bias favoring short-term investment linked to oversimplified financial calculations.
• Last but not least, such projects support a level of risk, which is significantly higher than a product development product. “Properly” closing or radically reorienting an exploration track is important know-how for managing such projects. Never-ending up-front exploration is one of the classical burdens of research departments, resulting in a detrimental resource scattering. But the classical myths about project management tend to assimilate a project stop with a failure of the project manager, and then do not encourage such behaviors.
The Emphasis on Portfolio Management: Cross-Learning and Real-Option Market
Finally, intense innovation strategies call for a renewed interest in portfolio management logic as a specific level of analysis to take into account. In the 90s, modern development project management as “heavy weight project structures” (Clark, 1991) emphasizes integration and valuation of existing knowledge and energies into projects. The major challenge for intensive innovation strategies is to move from mobilization on single projects, to global management of bunches and series of projects along innovation trajectories (Ben Mahmoud-Jouini, 2000) or lineage (Hatchuel, 1999) of products. The increase of risk level for each isolated project is another argument for focussing on portfolio management, as observed in the pharmaceutics (Cooper, 1998; Sharpe, 1998). Last but not least, the increased allocation of resources to R&D and the strategic importance of innovative performance create new constraints in terms of communication on R&D projects portfolios.
This perspective emphasizes the multiple interdependencies and potential externalities of project processes, whereas the single management project view considers only the cash results obtained at the end of the projects. Such an approach focuses on:
• Reconsidering the evaluation of individual projects with respect to the global portfolio, in term of global risk and profit profile, continuity and smoothness of expected revenues as consumption of spare resources,…
• Organizing the cross-project learning of technological aspects and market opportunities (Ayas, 1997; Charue-Duboc, 1999).
• Considering the possibility of “trading,” buying, or selling projects (wholly or partly) at different stages in their development. The relevant business model is modified to take into account the management of the R&D portfolio.
That last perspective forces us to reconsider the traditional vision of the R&D portfolio as a “funnel” to an image that we might describe as a “porous funnel” (Bayart, 1999), leaving room for trading with other firms as shown in Exhibit 4.
Exhibit 5.Comparison of the Decisional Frameworks of "Conventional Funnel" R&D and "Porous Funnel" R&D
The decisional framework for R&D management in porous funnel mode is significantly different from that associated with the conventional funnel, as Exhibit 5 shows.
In that perspective, conventional analysis of project portfolios in a risk/attractiveness “bubble-diagrams” graphic (Cooper, 1998) has to be reconsidered because the same project may be positioned at two very different locations, as is shown in Exhibit 6.
In Exhibit 6, at bottom right, the representation picturing internal development implies high risk and investment, while promising large profits in the event of success. At top left is the assessment of the same project if value is gained from it during development by means of a trade with another pharmaceutical firm: the risk and the investment are smaller (because return on the project is immediate and simultaneous with the trade, and the other firm will complete the development), but the profit in the event of success is obviously lower (sale price plus royalties on any future sales for example).
Such an emphasis on a global portfolio management issue is not to be interpreted as a coming back to functional structures that pre-existed to the development of project management structures. It has to be considered on the contrary as a new step in order to take into account, and if possible conciliate, the temporary and durable identities of modern business society.
To conclude, the four above points raise important problems when the traditional project management approach is been implemented to support intensive innovation strategies. In this communication, we have indicated tracks to address these problems by referring to organizational and business theories. Analysis of such issues appears a fruitful Expected opportunity for cross-disciplinary theoretical exchanges as for cooperation between scholars and practitioners. The creation of research network as IRNOP (International Research Network on Organizing by Projects) and PRAMECO (French research network on analyzing the ongoing transitions in R&D practices and professions. It gathers researchers who share an interactive methodology in various scientific field: organization and management, engineering, sociology, ergonomics and cognitive psychology) as the researches currently under way with major European firms try to materialize this opportunity.
Exhibit 6.The Representation of Various Scenarios Associated With the Same Project in a “Bubble Diagram” Representation of Conventional Type
Ayas, K. (1997). Design for learning for innovation. Delft, Netherlands: Eburon Publishers.
Bayart, D., Bonhomme,Y., & Midler, C. (1999). Management tools for R&D project portfolios in complex organizations: The case of an international pharmaceutical firm. In 6th International Product Development Management Conference. Cambridge (U.K.).
Ben Mahmoud-Jouini, Sihem, & Midler, Christophe. (2000). Crise de la demande et stratégies d’offre innovante dans le secteur du bätiment. Paris: Plan Construction et Architecture.
Ben Mammoud-Jouini, S. (1998). Stratégies d’offre innovantes et dynamiques des processus de conception. Le cas des grandes entreprises françaises de bätiment. Université Paris IX Dauphine, Thèse de doctorat de gestion, Paris.
Brown, S.L., & Eisenhardt, K.M. (1997, March). The art of continuous change: Linking complexity theory and time-paced evolution in relentlessly shifting organizations. Administrative Science Quarterly, 42, (1).
Chapel, V. (1996). La croissance par l’innovation: de la dynamique d’apprentissage a la révélation d’un modele industriel. Le cas Tefal. Thése de doctorat spécialité Ingénierie et Gestion, École Nationale Supérieure des Mines de Paris, Paris.
Charue-Duboc, F. (1997). Maítrise d’oeuvre, maítrise d’ouvrage et direction de projet: des catégories pour comprendre l’évolution des fonctionnements en projet dans le secteur chimie de Rhône Poulenc. Gérer et Comprendre, septembre: pp. 54–64.
Charue-Duboc, F. (1998). The role of research departments in focusing innovative projects and understanding customer usage and needs. In International Research Network on Project Management and Temporary Organization (IRNOP Conference). Calgary (Canada).
Charue-Duboc, F., & Midler, C. (1999). Impact of the development of integrated project structures on the management of research and development. In 6th International Product Development Management Conference. Cambridge (U.K.).
Ciavaldini, B. (1996). Des projets a l’avant projet: l’incessante quete de la réactivité. These de doctorat spécialité Ingénierie et Gestion, Ecole des Mines de Paris, Paris.
Clark, K.B., & Fujimoto, T. (1991). Product development performance. Strategy, organization and management in the world auto industry. Cambridge, MA: Harvard Business School Press.
Clark, K., & Wheelwright, S.C. (1992). Revolutionizing product development. New York: The Free Press.
Cooper, R.G., Edgett,S.J., & Kleinschmidt, E.J. (1998, July–August). Best practices for managing R&D portfolios. Research Technology Management.
Cusumano, M., & Nebeoka, K. (1998). Thinking beyond lean. New York: The Free Press.
Hatchuel, A., Chapel, V., Deroy, X., & Le Masson, P. (1998). Innovation répétée et croissance de la firme. Paris: Rapport de recherche CNRS.
Hatchuel, A., & Weil, B. (1999). Design oriented organisations. Towards a unified theory of design activities. In 6th New Product Development Conference. Cambridge U.K.
Kesseler,A. (1998). The creative supplier: A new model for strategy, innovation, and customer relationships in concurrent design and engineering processes: The case of the automotive industry. thèse de Doctorat, spéc. Gestion, Ecole Polytechnique, Paris.
Kogut, B., & Kulatilaka, N. (1994). Options thinking and platform investments: Investing in opportunity. California Management Review Winter, pp. 52–71.
Lenfle, S., & Midler, C. (2000). Innovation-based competition and the dynamics of design in upstream suppliers. International Vehicule Design Journal.
Lundin, R.A., & Söderholm, A. (1995). A theory of the temporary organization. Scandinavian Journal of Management 11, (4), pp 437–455.
Lundin, R.A., & Midler, C. (1998). Projects as Arenas for Renewal and Learning Processes. Norwell, MA: Kluwer Academic Publishers.
Midler, C. (1993). L’auto qui n’existaitpas; management des projets et transformation de l’entreprise. Paris: Dunod.
Midler, C. (1995). “Projectification” of the firm: The Renault case. Scandinavian Journal of Management 11, (4), pp. 363–375.
Navarre C. (1992). De la bataille pour mieux produire a la bataille pour mieux concevoir. Gestion 2000, (6), pp. 13–30.
Ponssard, J.P., & Tanguy, H. (1992). Planning in firms: An interactive approach. Theory and Decision (34), pp. 139–159.
Schön, D.A. (1983). The reflective practitioner. How professionals think in action. New York: Basic Books.
Sharpe, P., & Keelin, T. (1998, March–April). How Smithkline Beecham makes better Resource-Allocation Decisions. Harvard Business Review.
Simon, H.A. (1969). The Sciences of the Artificial. Cambridge, MA: MIT Press.
Söderlund, J., & Andersson, N. (1998). A framework for analyzing project dyads—The case of discontinuity, uncertainty and trust. In R.A. Lundin, & Midler, C. (Eds.), Projects as Arenas for Renewal and Learning Processes (pp. 181–189). Norwell, MA: Kluwer Academic Publishers.
Weil, B. (1999). Conception collective, coordination et savoirs. Les rationalisations de la conception automobile. These de doctorat spécialité ingénierie et gestion, École Nationale Supérieure des Mines de Paris, Paris.
Proceedings of PMI Research Conference 2000