On the novelty dimension in project management

Klaus Brockhoff
WHU – Otto Beisheim School of Management, Germany

Projects might be characterized by different degrees of novelty perceived by those who are involved in them. This can influence success, as shown in a number of contributions (for a summary, see: Schlaak, 1999, p. 102). In this paper, we draw primarily on earlier empirical research to develop an understanding of the consequences of different degrees of novelty for project management. At the same time, we want to introduce results from German studies that have not yet been reflected in the non-European scenes of project management practice or the literature on project management research.

Project

A project is defined in many different ways in the literature. It is not intended here to discuss these differences , sometimes resulting from distinctions in purpose for the definitions chosen. To some, it may come as a surprise that, in Germany, even an industry norm sets out to define a project. Since 1980, DIN (German Industry Norm) 69901 considers a singular combination of objectives, resources devoted to reaching the objectives, duration, separation from other activities, and a specific organization as constituent elements of a project. Separation from other activities will become of interest later, when measurement of novelty comes into play. Even without this particular perspective, it is obvious that in complex, large and frequently also highly innovative projects, interaction among teams which contribute to one overall goal is mandatory (Hoegl, Weinkauf, & Gemuenden, 2004). This increases complexity beyond the projects performed by one team, and it differs from the allocation problem arising from one team serving many projects. It leads to the differentiation of project level versus team level management functions (Hoegl & Weinkauf, 2005).

In other definitions, the complexity of projects and their “relative novelty” are mentioned (Frese, 1980). This element, however, is not further specified. It becomes immediately clear that adding the element of relative novelty to the aforementioned project definition might lead to difficulties. As in radically new projects, neither objectives, resources, nor duration might find a clear enough fixation to meet the definition. This is a first indication of the importance of the dimension of novelty in project management. Radically new projects, however, are not only difficult to subsume under standard project definitions. These projects also call for management approaches that differ from the routine. As a first requisite to choose and apply such approaches, one has to come up with an understanding of novelty.

Novelty

The subjectivity issue

Novelty, newness or innovativeness, which are widely used as synonyms, are at least as difficult to define as the term project itself. Novelty cannot be defined in strictly objective ways. Therefore, various “perspectives” (Daneels & Kleinschmidt, 2001) are considered. However, even one perspective, namely that of a firm engaged in a project, cannot be considered to be homogeneous. Rather, because novelty is imagined, it is a subjective phenomenon. Excluding one-person projects, the definition derives from the imaginations of a group of people who are related as contributors to a project. Because separation of project activities from other activities is part of the project’s definition, this group of people should be identified. Only if perceptions of all these individuals are identical might we speak of an objectified notion of novelty. However, such agreement cannot be assumed as a rule. By abstraction, let us assume that a group of people related to a project can be partitioned into one individual and the remaining members of the group, where the individual might agree or disagree with the rest of the group with respect to the novelty of a particular project. It is not uncommon to find a powerful project promoter to assume that a project is not really novel, while experts despair of the level of novelty they encounter. A similar observation with respect to top management as the promoter was made by Green (1995). This leads to four major types of projects (see Figure 1).

Type 1 projects are routine projects. All people involved agree on the zero-to-low degree of novelty. With respect to the knowledge necessary to perform the projects of this type, one might say that “we know” how to do it. Standardized project planning can be applied.

Type 2 projects are considered routine by all but one individual (or a minority of a few). This individual considers the respective project to be rather novel. The reason might be that only “they know.” Should “they” be right, the remaining individual could catch up by learning. This tells us that the spectrum of activities necessary for successful project completion is different from the Type 1 projects: Learning needs to be included in the project planning and activities. Should “they” be wrong, the project is likely to fail if the routines applied do not match with the requirements of novelty. This fate, then, is recognized by one person alone.

Type 3 projects are essentially the opposite of the Type 2 projects. Except for one individual, who considers this project type not to be rather novel, all others lack the special knowledge that the one individual seems to have. “I know,” this person could say. Lotka’s Law describing the skewness of expertise distribution among a group of people (Lotka, 1926) gives an explanation for the divergent perceptions of novelty. However, the problem for project management is that the disagreeing person might either be a genius or a swindler. It would be nice to have signals, such as exams, papers, references etc., to tell the one from the other. But even signals might not be foolproof. Johann Kunckel (1630 to 1703) was a chemist who, among other things, successfully completed projects to produce ruby glass in large quantities, and who re-discovered phosphorus. He wrote a number of well-received books, in particular on glass making, but never received an academic education. Obviously he had promised to the Saxon king that he was also able to produce gold. After the king’s death, he fought for outstanding remuneration. However, the chancellery told him that “if He could produce gold He does not need a salary; but if He cannot produce gold, why then should He be remunerated?” With respect to this project type, supervision might therefore be of primary importance. Defining milestones and checking on project performance are mandatory activities. Incentives or controls that prevent opportunism resulting from the asymmetric distribution of knowledge are mandatory. Documentation of the project progress and the cooperation of many with the knowledgeable person may help to transfer eventual tacit knowledge. This can help the larger number of people involved in learning.

Type 4 projects are those in which – if not in a very strict sense – “nobody knows.” These are truly innovative projects, on which everyone involved agrees. As compared with the Type 1 projects, this might require a different management approach. Information on the choice and application of appropriate project management tools is largely missing. Flexibility accompanying project progression is important. Evaluations of intermediate results are necessary as a basis for continuation or termination decisions.

One of the implications that the four project types have is for project planning. If novelty perceptions diverge, preferred planning modes will diverge as well for the different people involved. This could become a source of conflict. Problems of a similar nature might arise if multiple team projects are considered in which each team perceives its task as involving a different degree of novelty. Securing cooperation among the teams to enhance team innovation (Tjosvold, Tang, & West, 2004) might be difficult to achieve.

Quasi-objective or subjective perceptions of novelty and project types

Figure 1. Quasi-objective or subjective perceptions of novelty and project types

Since novelty evaluations such as in Figure 1 apply to a specific point in time, project progress will lead to a change of the respective evaluations (Lange, 1993, p. 125 et seq.). There are indications that this calls for phase-specific project management (Hoegl & Weinkauf, 2005) or, putting this differently, novelty-specific project management. While both technical uncertainty and market uncertainty are reduced in successful novel product development projects, market uncertainty is not reduced as frequently in unsuccessful projects (Lange, 1993).

The project types sketched in Figure 1 have another important implication, namely for empirical research as well as for reporting. The potential disagreement by project contributors on the evaluation of novelty tends to produce substantial single-informant errors in empirical research. Therefore, data should be collected from many informants in order to control this error. This is of particular interest for Type 2 projects and Type 3 projects. In the most encompassing review of new product success studies to date, Ernst (2001, 2002) identified only two out of 51 studies that explicitly and separately studied the hierarchical or functional levels of respondents as a source of error. In his own study, he finds that the informant bias is responsible for roughly 30% of the total variance in product success measures. The same author makes it clear that indirect as well as direct indications show substantial respondent errors with respect to variables that are associated with novelty in many of the ways in which this is measured. These different approaches of measurement need to be explained:

Measuring novelty

Novelty has no meaning by itself, but only in relation to some characteristic. Even if we limit further considerations to the Type 1and Type 4 projects, the characteristic to which novelty relates needs to be made explicit. This is one subject of measuring novelty. Even if we limit ourselves to product innovation projects, this field has produced a substantial number of approaches, ranging from the earlier and simple “new/not new” dichotomies—which, in their most simple form, do not even address which novelty characteristic is indicated via one-dimensional scales—to the multi-variable approaches of today (for instance: Garcia & Calantone, 2002; Daneels & Kleinschmidt, 2001, p. 359; Hauschildt, 2004, p. 14; Schlaak, 1999, p. 91). With respect to the latter approach, four contributions are of major importance. These are very shortly reviewed, keeping the definitional problems in mind.

(1) Green et al. (1995) consider four constructs to define novelty characteristics: technological uncertainty, technical inexperience, technology cost, and business inexperience. These are operationalized by a total of 17 items, one of which has been eliminated for statistical reasons. While the first of the constructs, technological uncertainty, offers an almost objective view of how the scientific community is able to handle the project, the remaining three items refer to the performing firm alone.

(2) Daneels and Kleinschmidt (2001) use a total of 19 items to operationalize four constructs: familiarity of the firm considered with the technology and familiarity with the market, as well as the fit or adequacy of the firm’s resources in the marketplace, and with respect to its technological capabilities to achieve the project objective. This firm-specific approach has some overlap with the firm-specific items of Green et al. (1995). However, except for “technological cost,” which is operationalized in a way that displays the adequacy or fit of technological resources, the remaining two constructs cut across the fit and familiarity dimensions chosen by Daneels and Kleinschmidt (2001).

(3) Schlaak (1999) substantially broadened these concepts. Starting out from Leavitt’s (1965) four dimensions of organizational change—task, structure, technology, and actors—he expands technology into a resources construct, and substitutes processes for the actors’ dimension. These constructs are operationalized by 40 items, 20 of which can be related to the familiarity construct and another 17 to the fit construct. Very careful analysis of a set of project data narrows down the number of items to 24, grouped into seven factors. Three of these represent the fit, while four represent familiarity. Consequently, Schlaak (1999) argues that his approach is different from that of Green et al. (1995). However, the seven factors used to measure novelty reach out far beyond the earlier two dimensions of market and technology. In fact, they point at a strategic issue of the project definition. As already mentioned, this includes “separation from other activities,” mostly meaning other projects performed in parallel. Schlaak (1999) shows how a project can have substantial influence on the whole organization into which it is embedded. Even a product innovation project can lead to novel procurement and production processes, and to changes in the informal or formal organization. In this sense, there is no separation from other activities, particularly if high degrees of novelty are observed. Such emanation effects, as we want to call them, can be a source of opposition against projects, because not everybody loves innovation. Explicit consideration of emanation effects might move a project from Types 1 or 2 to Types 3 or 4. The research by Schlaak (1999) also casts doubt on the proposition of independence of the familiarity and fit dimensions. As observed in many factor analysis studies, the seven factors he identifies are not completely independent of each other. This could mean that fit and familiarity are not totally independent of each other. For instance, novelty to the sales market is a familiarity dimension, while the fit dimension is represented by resource costs in R&D, production and marketing. Both are correlated at a lower level. Similarly, the “higher order” factor of technology and production is composed of two familiarity factors and one fit factor.

(4) An even broader perspective with respect to the consequences of project novelty is taken by the measurement approach of Gemünden and Salomo (2005). Their four dimensions of novelty relate to the market, the technology, the organization into which the project is embedded, and the external environment as defined by regulatory or societal aspects. A total of 20 items is used for construct definitions. Again, familiarity and fit cut across the four constructs. The broadening of the perspective is not only manifest in the environmental construct, but also, for instance, in the market aspects of novelty. The authors ask project representatives to what degree they think that customers need to change their behavior and attitude to appreciate the project outcome. Other than originally imagined, the perspective taken in the evaluation is not represented by a member of the group who holds a specific perspective similar to that of the firm hosting the project.

What can be learned from this? Even though great progress has been made in measuring novelty, no standard has yet been established. This is even more disturbing as the approaches referred to above focus only on product innovation projects. Novelty with respect to other types of projects should be measured by the same methodological rigor. The broadening of the novelty perspective adds more realism to the evaluation, but at the same time might be an additional source of variance.

Plausible concepts of novelty are not easily replicated in follow-on studies. Constructs cannot be considered generally valid, although authors tend to use about 20 items for their measurement and at least some basic agreement exists as to the necessity of including technology and market characteristics in measuring novelty. Multi-informant approaches are called for, but are not common. These could help to control for single informant errors. Perceptions matter, as can be seen by referring to the project categories of Figure 1.

A standard measurement procedure would establish an important step towards better comparison of the results of empirical studies, and it would help management to better classify their projects.

How novelty works: The principles of influence

Another important aspect is that of whether novelty is a moderator or an independent variable (Daneels & Kleinschmidt, 2001). These can be considered as two different principles of influence that novelty may have on projects. Assuming strictly linear relationships, and using {y, x, z} as variables and {a, b,..., f} as parameters, the two principles can be illustrated very easily. The variable y indicates a particular project outcome, such as one dimension of project success or a composite measure of project success. The variable x indicates a project management characteristic, while z is a measure of novelty. The latter implies a characteristic, which could also be represented by a vector with respective consequences for the parameters. Error terms are neglected here. Then, novelty as an independent variable is observed in

(1) x = a + bz

(2) y = c + dx = c + d(a + bz).

In (2), z is simply substituted for x. As a moderator, novelty is modeled in

(3) d = e + fz

(4) y = c’ + x(e + fz).

It is easily seen that (2) and (4) are conceptually different, and also that (4) is the more general approach. In empirical estimation of unknown parameters, it will not be possible to differentiate between c’ and c + da, both being considered as an increment. Thus, in (4) an additional parameter (e) is estimated. While this consumes another degree of freedom, it adds explanatory power if e is estimated as significantly different from zero. An even more expanded approach is represented by a combination of (2) and (4), which leads to

(5):

(5) y = c’ + x(e + fz) + d’z

with d’ = db. Because plausibility, occasionally supported by scatter plots, has it that some independent variables can have nonlinear effects on the dependent variable beyond the multiplication of x and z, linear approaches can be overly simplistic. Comparing a nonlinear expansion of (1) with an approach such as (4) or (5) might therefore be of interest.

From this short presentation, it is concluded that further empirical research should adopt the moderator principle in trying to explain influences of novelty on project management. Furthermore, nonlinearities should attract more attention than they have to date.

Novelty effects on project management

The following presents selected empirical findings of effects of novelty on project management. This has two major limitations, besides the inability to cover all studies. First, the measurement issues raised above limit comparisons. Second, we draw on studies that advance the hypothesis of a causality which is mostly tested by non-causal approaches. Thus, the possibility of project management influencing novelty rather than the other way around is not explicitly covered (see Figure 2). The dotted line shows this causality. However, this relationship does not appear to exist. Rather, both project management and novelty might be determined by underlying factors, such as the governance of the organization into which the project is embedded. In equations (1) through (4), this is implicit in a lack of time indices at the variables. The bold lines and the dashed line show the relations modeled in (5).

The causality issues between project management and novelty

Figure 2. The causality issues between project management and novelty

In the following, three aspects shall be considered: project management structure, clarity of project objectives and autonomy of project management, and the project manager.

Project management structure

Project management structure reflects a hierarchy of top management, with project manager and project team as the acting persons. Involvement of top management, competencies assigned to the project manager, and expertise of the project team have positive influences on project success. Lechler (1999) presents an overview of the results that support this. The same author shows in his LISREL study of 257 successful and 191 unsuccessful projects that the total effect on project success of these individuals or groups of people can vary with project type. Three types of projects identified by Lechler (1999) exhibit – among other characteristics – increasing “innovativeness” and “technological risks.” This observation helps in integrating a stream of research on the management structure for innovative projects, which was not originally integrated with the project management literature. In this descriptive, rather than normative, literature, the roles of sponsor, champion and expert together with their cooperation are identified as crucial for project success. Lechler (1999) associates top management with the role of sponsor, the team with the role of the experts, and project management with the role of the champion.

The role model was originally adopted by Witte (1973). He studied projects set up to manage first-time purchases of computers in companies. He identified a sponsor’s role (called power promoter) and an expert’s role as the two important roles that need to be present and represented at the same time to enhance the chance of project success. This research was expanded into integrating the role of the project champion (process promoter), and more recently to the gatekeeper for the transfer of knowledge as well as the relationship promoter to support project relations with other organizations beyond the project’s host (Chakrabarti & Hauschildt, 1989; Hauschildt & Gemünden, 1999). The promoter roles are associated with particular hierarchies of their representatives as well as with particular resources that their representatives can mobilize for project success. A recurrent result of empirical research is that the roles should all be represented in project management and that cooperation among the representatives of these roles is decisive for project success.

If one accepts Lechler’s association of people and roles, it is possible to study the importance of each one of the three roles for project success (Table 1). Here, this is of particular interest with respect to the novelty dimension. For both, the project manager (champion) and the project team (experts), their effect on project success increases with the two novelty variables. The effect of top management is almost identical, with a slight reduction of influence on success for the most innovative projects. The top management (sponsor) influences success directly as well as indirectly via its influence on project management and teams. The project management influences project success only via its influence on the teams. For the most novel projects standard instruments of project management play a very reduced role. It is suggested that these projects require considerable new knowledge, which consequently increases the importance of the team for project success. Also, because these projects are relatively complex and consume more resources than any other type of project, the role of the project management is stressed. The slight reduction of the importance of top management’s role is explained by the assumption that this type of project is so prominent within the organization that it does not need as much top management attention as might be necessary to keep a less prominent project on a path to success.

  Project Type A (n=192) Project Type B (n=120) Project Type C
(n=102)
 
Novelty Low Medium High  
Risk Low Medium High  
Project size Medium Small Large  
Total impact of top management .60 .64 .59  
Total impact of project management .12 .16 .23  
Total impact of project team .41 .40 .63  

Source: Lechler (1999, pp. 205 & 207)

Table 1. Impact on project success by three types of role representatives and project characteristics

In this study, an overly simplistic—namely, static—view is presented. In reality, the roles of promoters are evaluated differently with respect to different project phases. The people representing different roles can change, and the relevance of certain roles for project success can vary with novelty (Folkerts, 2001). In particular, if sponsors’ support in highly novel projects is discontinued, these projects tend to fail. Furthermore, the interaction or cooperation of the three role bearers is not studied. In addition, the inclusion of the novelty dimension is a rather crude one. Both of these aspects are covered in a much more elaborate study in which the novelty dimension is the major issue (Papies, 2005). At the same time, this study looks at the same projects in different phases of their development.

With respect to novelty, Papies (2005), with some restrictions, follows the above-mentioned concept of Schlaak (1999). Furthermore, he looks at the same new product development projects in the three phases of concept development, advanced development and testing or market introduction. In the first two phases, interactions of the roles of sponsor, champion and expert are studied together with top management support and cooperation of the promoter’s roles. The sponsor, the champion and the expert roles are all identified and considered with respect to their influences on the success variables. Thus, a very elaborate approach is chosen. Novelty is considered as a moderator, as in (5).

In Tables 2 and 3, a summary of results is presented. In these tables, – or + signs represent a significant parameter estimation in the respective direction, while a 0 stands for an insignificant result. It is found that in the two relevant project phases, novelty itself (as measured by the parameter d’ in (5)) is perceived as having mostly significant negative effects on project efficiency and project effectiveness, but positive effects on the contribution to the project’s USP. The moderating effect of novelty (represented by the parameter f in (5)) is significant in only a few situations.

  Phase Success measure (y) Promoter type / Variable (x) d’ f  
  Concept development Effectiveness Sponsor - 0  
    Efficiency Sponsor - 0  
    USP Sponsor + 0  
    Effectiveness Champion - 0  
    Efficiency Champion - 0  
    USP Champion + 0  
    Effectiveness Expert - +  
    Efficiency Expert - +  
    USP Expert 0 0  
    Effectiveness Cooperation - 0  
    Efficiency Cooperation - +  
    USP Cooperation 0 0  
    Effectiveness Top man. support - 0  
    Efficiency Top man. support - +  
    USP Top man. support + 0  
    Effectiveness Team quality - 0  
    Efficiency Team quality - +  
    USP Team quality + 0  

Source: Own construction according to results by Papies (2005)

Table 2. Novelty as moderator in structuring project management in the concept development phase

  Phase Success measure (y) Promoter type / Variable (x) d’ f  
  Advanced development Effectiveness Sponsor 0 0  
    Efficiency Sponsor - 0  
    USP Sponsor + +  
    Effectiveness Champion 0 0  
    Efficiency Champion - 0  
    USP Champion + 0  
    Effectiveness Expert 0 0  
    Efficiency Expert - 0  
    USP Expert + +  
    Effectiveness Cooperation 0 -  
    Efficiency Cooperation - 0  
    USP Cooperation + 0  
    Effectiveness Top man. support - +  
    Efficiency Top man. support - +  
    USP Top man. support + 0  
    Effectiveness Team quality 0 -  
    Efficiency Team quality - -  
    USP Team quality + 0  

Source: Own construction according to results by Papies (2005)

Table 3. Novelty as moderator in structuring project management in the advanced development phase

In the conceptual phase of the projects studied, a significantly positive moderating effect occurs five times. This means that, although novelty might reduce the value of the success variables considered, this reduction can be more than compensated for by higher degrees of novelty if an expert is involved or if cooperation among sponsors is insured, or if top management lends its support, or if team quality can be assured. In four of these five cases, this occurs with respect to efficiency, and only once with respect to effectiveness. Otherwise, insignificant results are noted for this phase of project development.

In the advanced development phase, it is notable that higher degrees of novelty have a negative influence on efficiency in all cases. Obviously, high degrees of novelty are correlated with high degrees of uncertainty, which in turn lead to surprises, spoiled well-laid-out plans, etc. The moderating effect is similar to that of the preceding phase with respect to top management support. It is even stronger with respect to the influence of a sponsor or an expert on USP. However, effectiveness of team quality and cooperation, as well as efficiency of team quality, are all negatively influenced by higher novelty during this phase. This may result from the weakness of team management in dealing with a situation as complicated as that of high novelty.

In software development, project team performance can be moderated by novelty of the projects assigned to a team. Interestingly, high levels of novelty impact on the relationship between team quality and efficiency, but do not significantly impact on the relationship between team quality and effectiveness as seen by team leaders and managers; team member ratings are non-significant overall (Hoegl, Parboteeah, & Gemünden, 2003). Non-significant negative impact of higher novelty levels on team quality can be compensated by the moderating effect. This result seems to contradict earlier findings, and suggests managing team quality in response to novelty rather than using unified approaches across the board.

Clarity of project objectives and autonomy of project management

Some of the negative direct influences of high novelty on efficiency and effectiveness of projects might be related to a lack of structuring objectives in such projects. Prescriptive literature suggests that project objectives should be non-complex, measurable, specific, tangible and easily verified, among other characteristics (Kerzner, 1984, p. 344). Primarily, this seems to apply to Type 1projects (See Figure 1). There is nothing wrong with trying to achieve these characteristics, even in novel projects (of Type 4 in Figure 1). However, such objectives should not be upheld irrespective of the path of the knowledge accumulation in the project. Case studies show that this request, while naively natural, is frequently not met, particularly if commitments among different contributors to a project have to be agreed upon (Hauschildt & Pulczynski, 1992; Hauschildt & Pearson, 1994). Penalty for breach of contract at the institutional level, or face-saving at the individual level, might work towards inflexibility, but also, unfortunately, towards project failure. To achieve flexibility in project management without totally losing sight of the original objectives, a particular set of characteristics of the project manager or the champion could be called for, or a certain level of autonomy in managing the project. Results on both of these issues can be presented.

As mentioned before, separation from other activities is considered a definitional characteristic of projects. This could help to achieve the level of flexibility needed for successful high-novelty projects. The same idea seems to have led quite a number of researchers to suggest that a high degree of novelty might best be managed if project managers enjoy a high level of autonomy with respect to organizational structure and physical separation from routine operations, resource availability and use (Krieger, 2005), as well as social autonomy. The latter is a short description for either offering personal, face-to-face cooperation among team members or for team members remaining in their home institutions with only virtual cooperation (Gemünden & Salomo, 2005). A fine overview of these suggestions and findings is given by Krieger (2005, p. 37). Another quite similar expression for this suggestion is the high-powered project management (Wheelwright & Clark, 1992).

As before, the novelty dimension can be studied as moderating possible direct relationships between variables measuring various dimensions of autonomy and project success. This casts doubts on earlier findings of extending autonomy in projects with increasing novelty to achieve project success (for instance: Christensen & Overdorf, 2000; Simon, Houghton, & Gurney, 1999; Gerwin & Moffart, 1997). Several arguments can be advanced to explain findings which do not show positive effects of structural autonomy and limited effects of resource autonomy on success (Krieger, 2005). Increasing autonomy of the project management can lead to cutting ties with the originally supporting organization together with insignificant effects of the novelty moderator. Thus, for instance, recourse on knowledge or other resources might be cut off. Face-to-face interaction as an indicator of social autonomy of a project is, however, significantly positively associated with success, and the novelty moderator has a positive influence. This calls for co-locating teams that work on highly innovative projects within the organization to benefit from the richness of personal information exchange and an easier way of motivating identification of team members with the project.

With respect to another autonomy dimension, namely autonomy in formulating objectives, it is found that too much autonomy might paralyze the synergies that the supporting organization, hopes for when starting the project (Brockhoff & Schmaul, 1996). A loss of focus can be the result. Even the internet pages that report on project failures are abundant with not only the usual criticism of lack of resources, weakness of project management or failure of top management support, but also with respect to overambitious goals, project objectives driving away from the vision of the supporting organization or goal changes by top management without proper communication (for example: Rossi, 2005; Hedman, 2005). Again, the novelty dimension may come into play as a moderating factor. Because some projects are driven by the availability of resources (means), including knowledge, and other are driven by the availability of a clear view of purposes (ends) (for the means-end-classification of innovative projects, see: Hauschildt & Pearson, 1994), the change of objectives can have substantially different success influences. It is quite plausible to assume that such changes can be helpful and supportive for resource-driven projects, while it can kill purpose-driven projects, unless the purpose itself undergoes respective parallel changes.

The findings call for a complicated multidimensional optimization of autonomy as a means to achieve the flexibility necessary when novelty increases. First, the optimization has to be achieved with respect to the novelty dimension, where we have indications that with respect to the market and technology dimensions more autonomy can be allowed as compared with the organizational and environmental fit dimensions. This can be concluded from the observation that the level of market or technological novelty in the sense of fit does not significantly correlate with success measures, while some negative correlations can be observed with respect to organizational and environmental novelty dimensions. These come close to a familiarity dimension. The management task here was nicely phrased a few years ago by Yasutsugu Takeda, then head of research at Hitachi Corp. He did not like “blue sky” research projects, but rather “north star” projects: almost equally removed from the present technology and market, but organizationally much better focused. Second, the optimization has to be achieved with respect to the dimension of autonomy, where issues of organizational structure, resource availability, and freedom to dispose of resources as well as social autonomy have to be considered. Third, optimization has to be achieved with respect to the success dimension and the phase of project work, which might serve as signals for the ultimate but hardly controlled financial success in the market.

It is obvious that the project management knowledge to date does not offer enough information for such an optimization. This is particularly so, because the study by Krieger (2005) which opened this presentation of the optimization problem was based on 104 relatively radical or very novel types of product innovation projects. To what degree the results can be corroborated with respect to other types of projects, we do not know.

Project manager

To what degree an able project manager can make good for missed optimization is another interesting question. Even a casual look at the project management literature reveals the extremely demanding set of characteristics that the “ideal” project manager should meet. As speculated above, this may be even more important in the case of high-novelty projects because the multidimensional optimization has to be negotiated with the management of the supporting organization, and because maintaining the optimal degree of internal flexibility—for instance, with respect to the project objectives—is very demanding. Indeed, the accumulated characteristics that practitioners and the academic project management literature use to define “ideal” project managers has led one researcher to speak of the ideal project managers as “heroes” in the classical sense of the word or even “archangels” (Drumm, 1996). Friedman, Fleishman and Fletcher (1992), looking at R&D projects, question the necessity of technological qualifications for project managers and favor their managerial competencies to achieve success. A similar indication, particularly from the point of view of Japanese managers as compared with German or US managers, supports this view (Brockhoff, 1990, p. 87), although with no particular reference to projects. By application of conjoint analysis, Keim (1997, p. 219) identified the relative weight of competencies requested for R&D managers (see Table 4). It is apparent that the past experience on the job, together with systematic and analytical thinking, counts less than the combination of creativity, motivation, and the abilities to plan and organize. The highest level of each characteristic is always preferred to lower levels, with one exception. For creativity of the project manager, both the lowest and highest level receive almost equal weight, while medium levels find small numbers of supporters.

  Characteristic Mean weight  
  Past development experience 20.8  
  Systematic and analytic thinking 18.4  
  Ability to plan and organize 24.7  
  Ability to motivate 17.6  
  Creativity 18.5  

Source: Keim (1997, p. 234)

Table 4. Mean weights for characteristics of “ideal” R&D project managers

In reality, hardly any project manager will be able to live up to the highest expectations with respect to each and every one of the characteristics mentioned in Table 4. Therefore, 24 items were identified from the literature on project manager characteristics to describe reality (Keim, 1997, p. 151). By employing factor analysis, these were assigned to seven factors, clustered to identify five types of project managers. These, in turn, were related to project success. Success was mainly measured as technical success, but all projects had already been introduced into the markets, some showing a successful market life of three years. Again following the analysis by Keim (1997), Table 5 summarized her results.

From Table 5 we can conclude that a project manager who interacts with his team, who can motivate and who can critically or logically evaluate project performance, achieves the best project results. Positive, but not outstanding, values with respect to the other characteristics certainly help to do such an outstanding job. This characterization is particularly interesting if confronted with the results for the group of the least successful project managers. They stand out with respect to their problem-solving capabilities, but they miss out on all other characteristics. One can imagine that this characterizes managers who are used to exercising power, to put high pressure on a team with little regard for the specific difficulties of highly uncertain and novel jobs. It was mentioned before that the dominance of a power promoter in novel projects is not a favorite driver of success. Type 2 represents the most frequently observed project manager. This manager comes close to the mean of characteristics of all project managers with respect to five out of the seven characteristics, but with lowest values for two of them. One can imagine that working under the leadership of this type of manager means to live with a certain degree of chaos, but the average business experience and the average learning ability might still create acceptable results. The wheeler-dealers of Type 3 have to fight against their lack of other characteristics, specifically learning and more reflective responses to problems that arise. Type 4, finally, seems to lack experience, which cannot be fully compensated for by their learning, initiative, motivation and reflection. This type seems to lack a frame of reference for the positively marked characteristics.

Characteristics Project Manager
Type 1
Project Manager
Type 2
Project Manager
Type 3
Project Manager
Type 4
Project Manager
Type 5
 
Can interact
with team
1.47 0.12 -0.47 -1.06 -0.15  
Has business 0.23 0.04 0.92 -0.57 -1.01  
Has problem
solving
capability
0.49 -0.55 -0.15 0.07 1.33  
Shows
initiative and
creativity
0.49 -0.62 0.95 0.68 -1.25  
Can learn and
organize
           
  0.12 -0.03 -0.99 0.47 -0.08  
Can motivate
and cooperate
           
  0.62 -0.35 -0.0 0.40 -0.21  
Reflective,
logical thinker
           
  0.33 -0.22 -1.15 0.32 -0.48  
Relative share
of project
manager types
20% 38% 11% 20% 11%  
Project
success scale
values
5.6 4.8 4.7 4.5 3.8  

Source: According to Keim (1997, pp. 160–161, & 189)

Characteristics of project managers are evaluated by T-values, based on standardized factor values. They can be interpreted as deviations of one cluster of project managers from the overall mean. Extreme positive and negative T-values per each characteristic are highlighted for better reading.

Table 5. Realistic types of R&D project managers, their frequency of appearance, and success.

While these results offer a great deal of insight into the characteristics of project managers for novel projects, it is unclear whether they can be applied to other project types. Also, because all of the projects considered in the study had been introduced into the market, no major project failures are included in the study. Their inclusion might sharpen the picture that could be drawn, or introduce a new type of project manager who scores below average on all characteristics.

Souder and Jenssen (1999, p. 198) concluded that “more exacting new product development practices are required to achieve success in unfamiliar market environments.” First, this means that methods should be differentiated with respect to the degree of novelty and novelty characteristics. Second, this suggests that the application of more exacting practices and methods needs particular project manager abilities or characteristics. This could well add a third dimension to Table 5. Rubenstein et al. (1976, p. 18) once said that one should not believe that “organization structure, control mechanisms, formal decision-making processes, delegation of authority, and other formal aspects of a so-called well-run company are sufficient conditions for successful technological innovation. ” In view of the results in Table 5, one might say that these instruments are necessary, but they need to be applied by project managers who know when to choose which instrument, and to determine the level of its application if this is possible beyond a simple alternative of choice.

In a separate part of her study, Keim (1997, p. 214) discusses the question of associating the characteristics of the five types of project managers can be associated with the promoters. She concludes that Type 1 can be associated with the champion and Type 3 has a profile that corresponds with the expert. This follows the same associative reasoning as is already known from Lechler’s (1999) study. We are critical of such associative, not empirically tested, reasoning, particularly because the low levels of learning and reflective abilities of Type 3 project managers cast doubt on Keim’s (1997) conclusions. It would not be surprising if—in further, more expanded, analyses—one would find that, for instance, Type 5 is the sponsor of a pet project that he or she personally manages. Type 4 has a number of characteristics that could more convincingly characterize them as experts.

Furthermore, no information is available on the cooperation among the team members. Thus, the extant work is no true test of the promoter model and its hypotheses. In all fairness, it was not planned as such a test.

Conclusion

In establishing efficient and effective project management, a multi-dimensional optimization problem has to be solved. One of the driving forces to be observed in striving for a solution is the moderating effect of novelty, which is itself a multidimensional concept. It is difficult to assess novelty, since it is a subjectively held concept related to dimensions that can be company-internal or -external. No standard measurement approach has yet evolved. In fact, the advance of measurement concepts is not without conflicts. This limits comparisons across studies. Furthermore, the way that novelty is empirically modeled can have influences on results with respect to the success dimension of projects. It is shown that the approach to consider novelty as a moderator is more general than the approach to consider it as an immediate success variable alone.

One of the contributions of the present paper is to present a few results of empirical research on project management done in Germany, and to relate this to some of the issues of project management that are discussed internationally. Here, concepts as that of the promoters of project management might prove to be fruitful in other research as well. The moderating influence of novelty can be combined with the promoter concept to explain project success. It can also be used to identify more formal project management methods as being adequate for the less novel projects, while certain characteristics of project managers might be more important than project management methods for achieving success of highly innovative projects.

Brockhoff, K‥ (1990). Stärken und Schwächen industrieller Forschung und Entwicklung [Strength and weaknesses of research and development]. Stuttgart, Germany: C. E. Poeschel Verlag.

Brockhoff, K., & Schmaul, B‥ (1996). Organization, autonomy, and success of internationally dispersed R&D facilities. IEEE Transactions on Engineering Management, 431), 33–40.

Chakrabarti, A., & Hauschildt, J. (1989). The division of labour in innovation management. R&D Management, 191),161–171.

Christensen, C. M., & Overdorf, M. (2000). Meeting the challenge of disruptive change. Harvard Business Review, 782), 67–76.

Daneels, E., & Kleinschmidt, E. J. (2001). Product innovativeness from the firm’s perspective: Its dimensions and their relation with project selection and performance. Journal of Product Innovation Management, 186), 357–373.

Drumm, H. J. (1996). Das Paradigma der Neuen Dezentralisation [The Paradigm of new Decentralization.]. Die Betriebswirtschaft, 561), 7–20.

Ernst, H‥ (2001). Erfolgsfaktoren neuer Produkte. Grundlagen für eine valide empirische Forschung [Success factors of new products: Foundations of valid empirical research]. Wiesbaden, Germany: Gabler Verlag, Deutscher Universitätsverlag.

Ernst, H.. (2002). Success factors of new product development: A review of the empirical literature. International Journal of Management Reviews, 41), 1–40.

Folkerts, L. (2001). Promotoren in Innovationsprozessen. Empirische Untersuchung zur personellen Dynamik [Promoters in innovation processes. Empirical research of the human resource dynamics]. Wiesbaden, Germany: Deutscher Universitätsverlag.

Frese, E. (1980). Projektorganisation [Project organization]. In Handwörterbuch der Organisation (2nd ed.). (pp. 1960-1974). Stuttgart, Germany: C. E. Poeschel Verlag.

Friedman, L., Fleishman, E. A., & Fletcher, J. (1992). Cognitive and interpersonal abilities related to the primary activities of R&D managers. Journal of Engineering and Technology Management, 93), 211–242.

Gemünden, H. G., & Salomo, S. (2005). Der Einfluss der Projektautonomie auf den Projekterfolg [The influence of project autonomy on success]. In J. Amelingmeyer & P. E. Harland (Eds.), Technologiemanagement & Marketing – Herausforderungen eines integrierten Innovationsmanagements (pp. 63-82). Wiesbaden, Germany: Gabler Verlag.

Garcia, R., & Calantone, R. (2002). A critical look at technological innovation typology and innovativeness terminology: A literature review. Journal of Product Innovation Management, 192), 110–132.

Gerwin, D., & Moffat, L. (1997). Withdrawal of team authority during concurrent engineering. Management Science, 439), 1275–1287.

Green, S. G. (1995). Top management support of R&D projects: A strategic perspective. IEEE Transactions of Engineering Management, 42(3), 223-232.

Green, S. G., Gavin, M. B., & Aiman-Smith, L. (1995). Assessing a multidimensional measure of radical technology innovation. IEEE Transactions on Engineering Management, 423), 203-214.

Hauschildt, J. (2004). Innovationsmanagement [Innovation management]. München, Germany: Vahlen Verlag.

Hauschildt, J., & Gemünden, H.. (Eds). (1999). Promotoren. Champions der Innovation [Promoters. Champions of innovation]. Wiesbaden, Germany: Gabler Verlag.

Hauschildt, J., & Pearson, A. W. (1994). Ends, means and innovation management. Creativity and Innovation Management, 33), 162–166.

Hauschildt, J., & Pulczynski, J. (1992). Rigidität oder Flexibilität der Zielbildung in Innovationsprozessen? [Rigidity and flexibility of goal formation in innovation processes]. Zeitschrift für Organisation, 612), 74–81.

Hedman, E. R. (2005). Why technology projects fail. Retrieved September 7, 2005, from www.thespacereview.com/article/372/1.

Hoegl, M., Parboteeah, P., & Gemüenden, H. G. (2003). When teamwork really matters: Task innovativeness as a moderator of the teamwork-performance relationship in software development projects. Journal of Engineering and Technology Management, 203), 281–302.

Hoegl, M., & Weinkauf, K. (2005). Managing task interdependencies in multi-team projects: A longitudinal study. Journal of Management Studies, 426), 1287–1308.

Hoegl, M., Weinkauf, K., & Gemünden, H. (2004). Inter-team coordination, project commitment, and teamwork in multi-team R&D projects. Organization Science, 151), 38–55.

Keim, G. (1997). Projektleiter in der industriellen Forschung und Entwicklung [Project managers in industrial research and development]. Wiesbaden, Germany: Deutscher Universitätsverlag.

Kerzner, H. (1984). Project management (2nd ed.). New York: Wiley.

Krieger, A. (2005). Erfolgreiches Management radikaler Innovationen. Autonomie als Schlüsselvariable [Successful management of radical innovations. Autonomy as key variable]. Wiesbaden, Germany: Deutscher Universitätsverlag.

Lange, E. C. (1993). Abbruchentscheidung bei F&E-Projekten [Stopping decisions for R&D projects].. Wiesbaden, Germany: Deutscher Universitätsverlag.

Leavitt, H. J. (1965). Applied organization change in industry: Structural, technological and humanistic approaches. In J. G. March (Ed.), Handbook of organizations (pp. 1144–1170). Chicago: Rand McNally.

Lechler, T. (1999). Was leistet das Promotoren-Modell für das Projektmanagement? [What is the contribution of the Promotor Model to project management?]. In J. Hauschildt & H.-G. Gemünden (Eds.), Promotoren. Champions der Innovation (pp. 179-209). Wiesbaden, Germany: Gabler Verlag.

Lotka, A. J. (1926). The frequency distribution of scientific productivity. Journal of the Washington Academy of Sciences, 166), 317–323.

Papies, S. (2005). Prozessbegleitende Bewertung von Innovationsprojekten [Process-related valuation of innovation projects]. PhD dissertation. Vallendar, Germany: WHU.

Rossi, S. (2005). Using the insanity clause for failed projects. Retrieved September 7, 2005, from www.cio.com.au/index.php.id;162727491;fp;4;fpid;21

Rubenstein, A. H., Chakrabarti, A. K., O’Keefe, R. D., Souder, W. E., & Young, H. C. (1976, May). Factors influencing innovation success at the project level. Research Management, 19, 15–20.

Schlaak, T. M. (1999). Der Innovationsgrad als Schlüsselvariable. Perspektiven für das Management von

Produktentwicklungen [The degree of innovativeness as a leading indicator. Perspectives for the management of product development]. Wiesbaden, Germany: Deutscher Universitätsverlag.

Simon, M., Houghton, S. M., & Gurney, J. (1999). Succeeding at international corporate venturing – Roles needed to balance autonomy and control. Journal of Applied Management Sciences, 8(2), 145–149.

Souder, W. E., & Jenssen, S. A. (1999). Management practices influencing new product success and failure in the US and Scandinavia: A cross-cultural comparative study. Journal of Product Innovation Management, 16(2), 183–203.

Tjosvold, D., Tang, M. M. L., & West, M. (2004). Reflexivity for team innovation in China. The contribution of goal interdependence. Group & Organization Management, 29(5), 540–559.

Wheelwright, S. C., & Clark, K. B. (1992). Revolutionizing product development. New York: Free Press.

Witte, E. (1973). Organisation für innovationsentscheidungen. Das ptomotorenmodell [Organization for innovation decisions. The Promotor Model]. Goettingen, Germany: Vandenhoeck & Rupprecht.

This material has been reproduced with the permission of the copyright owner. Unauthorized reproduction of this material is strictly prohibited. For permission to reproduce this material, please contact PMI or any listed author.

©2006 Project Management Institute

Advertisement

Advertisement

Related Content

Advertisement

Publishing or acceptance of an advertisement is neither a guarantee nor endorsement of the advertiser's product or service. View advertising policy.