Managing Relations Between R&D and Marketing in NPD Projects PDF

6 J PROD INNOV MANAG 1988;5:6-19 0000 Managing Relaticus Between R&D and Marketing in New Produt1 Development Projects William E. Souder This article examines the R&D/marketing inter- Introduction face conditions found within an extensive data...

6 J PROD INNOV MANAG 1988;5:6-19 0000 Managing Relaticus Between R&D and Marketing in New Produt1 Development Projects William E. Souder This article examines the R&D/marketing inter- Introduction face conditions found within an extensive data Research and development (R&D) and marketing base of new product development innovation personnel depend on each other fOi the creation projects. The incidence of different types of prob- of new product innovations. Yet R&D and mar- lems between these two important functions are keting departments have frequent misunderstand- analyzed and the effects of these observations on ings and conflicts. project outcomes are discussed. The article con- Many managers have first-hand experience tains a number of recommendations for increas- with R&D/marketing interface problems and be- ing the success rates of innovation projects by haviors between R&D and marketing groups using a model that improves conditions at the have been carefully studied [3,5,7-11]. However, R&D/marketing interface. much more information is needed about this complex and important topic. This paper exam- ines the R&D/marketing interface conditions found at 289 new product development innov~­ tion projects. Based on these findings, strategies and guidelines are presented for improving the relationships between R&D and marketing groups. Methodology of This Study This study was carried out on a comprehensive data base of life cycle information on 289 new product development innovation projects. The data were collected through ten years of intensive field research at 56 consumer and industrial prod- uct firms [7-10]. Exhibit 1 and Table 1 present the methodology and the project outcome meaSllrc- ment scales used in collecting that data base. The 289-project data base contains numerous ::!~!~iied descriptions and ratings of key events, Addres§ cor~espondencc to William E. Souder, Technology activitie~:;, attitudes and bchavior~ of the R&P Management Sllldje~ In~titute, Department of Industrial Englllel'r- ing. 1048 Benedum Engineering Hall, University of Pillsburgh. and marketing personnel who worked on each Pittsburgh. PA 15261. project. As part of the content analyses, stalisli- ID 19:18 Elsevier Sci~!lcC I'ubli~hing Co., Inc. 52 Vanderbilt Ave. New York. NY 10011 I:.D/MARKETlNG INTERFACE IN NEW PRODUCTS J PROD INNOV MANAG 7 1988;5:6-19 these descriptors are "Frequency of Joint Meet- BIOGRAPmCAL SKETCH ings," "Frequency of Joint Customer Visits," Wm. E. Souder is Professor of Industrial Engineering and Professor of "Degree of Perceived Need to Interact," and Engineering Management. and Director of the Technology Manage- ment Studies Institute at the University of Pittsburgh. A graduate of "Degree of Regard for the Other Party's Com- Purdue University (B.S. with Distinction) and St. Louis University petency. " (MBA and Ph.D.). Dr. Souder has published over 90 research papers Each of the 289 projects was rated on each of and five books on R&D 3I\d new product management. He serves as a director of a new venture firm. a scientific advisor to a human factors these descriptms. Some of these ratings came laboratory and a member of the editorial boards of several prominent directly from the instruments, while others were journals. Prior to joining academia. Dr. Souder spent 12 years in developed through content analyses of the inter- various industrial R&D and new product management positions within views [1,2,7-10]. Redundancies were built in &.t the chemical industry. several points. For example, the "Frequency of Joint Meetings" was primarily measured by ques- tionnaire items that asked for the number of times d reduction &nd factor analyses of this large per year that joint meetings were held. Details ata base [1,2,4], these items were reduced to 42 about these joint meetings were solicited during ttitudinal and behavioral descriptors of the the interviews. Differences greater than 10% in &D/marketing interface. Some examples of the questionnaire responses of the marketing, Exhibit 1. About the Data Base Data: Life cycle data were collected on 289 new produc. development innovation projects at 53 consumer and industrial product firms [7-10]. The data collection focused on project events, with detailed attention given to organization structures, environments, climates, behavioral processes and project success/failure factors. The ultimate objective was to understand development processes for new produt;t innovations. Sample of Firms: Using published statistics, an industry by industry compilation was made of firms with significant new product activities in either consumer or industrial goods. Target firms were then randomly selected from this list, based on a compromise design that carefully considered the cost of traveling to distant sites and the need to maintain representivity on several i:nportant Jimews!ons [7-10]. Approximately five firms were selected from each of the following ten industries: metals, glass, transportation (includes automotive and mass transit), plastics, machinery, electronics (includes computers and instruments). chemicals, food, aerospace and pharmaceuticals. Sample of Projects: Using carefully specified definitions, the population of new product innova- tion projects initiated during the preceding five years was assembled at each firm. A random sampling of equal numbers of success and failure outcome projects was taken from these populations at each firm, while maintaining a range of types of technologies, types of innovations, degrees of difficulty of projects, central vs. divisional R&D efforts, and several other important dimensions [7-10]. Several ongoing projects whose success or failure outcomes were unknown at that time were intentionally included in this sampling. Following these procedures, approximately 10% of each film's portfolios were selected into the 289 project sample studied here. Data Collection: A total of 27 instruments, numerous telephone interviews and 584 in-depth face-to-face interviews were carried out on each project to record the life cycle histories and extract the relevant data on each of the 289 projects [7-lQ). A cascading interview procedure was used to cross-validate information cellected from eact, marketing, R&D and other subject on each I project 17··· 101.  J PROD INNOV MANAG W. E. SOUDER 8 1988;5:6-19 Table 1. Project Outcome Measurement Saile characterized by a low frequency of meetings between the R&D and marketing personnel, Success Outcomes highly speciaiized and organizationally separated Degrees of Descriptors a R&D and marketing functions, and low degree Success Technical Outcomes Commercial Outcomes of perceived need to interact. Therefore, the label "Lack of Interaction" was coined to describe High Breakthrough Blockbuster this R&D/marketing interface state of affairs. Medium Enhancement Above expectations Met the specs Met expectations Twenty-two of the 289 projects, or 7.6% of the Low sample, exhibited this state. Similarly, the other Failure Outcomes states and percentages shown in Table 2 were Descriptors found and accordingly labeled. Degrees of Several firms that experienced the Lack of Failure Technical Outcomes Commercial Outcomes Interaction, Lack of Communication and Too- Low Learned a lot Below expectations Good Friends problems on the projects studied Medium Gained some Protected our position here avoided these states on some subsequent technology but lost money projects. Follow-up studies with these firms High Complete dud Took a bath we won't forget showed that they overcame these states through modest efforts. These efforts included more fre- Other Outcome quent joint meetings, joint involvemeni.s in plan- SE = Stopped the effort early due to poor progress ning proposed projects and increased sharing of information. Moreover, though these problems often lowered the organization's new product development effectiveness, they were not totally R&D and other personnel on the same project disruptive and they seldom led to major project were reconciled during the interviews. As an- failures. Therefore, as shown in Table 2, these other example, the "Degree of Perceived Need to problems were labeled 'mild'. By contrast, the Interact" was primarily measured by asking Lack of Appreciation and Distrust problem states pointed questions of the respondents during the interviews. The information from the interviews was then checkea against the Likert-type scale Table 2. Incidence of Harmony and ratings supplied on the questionnaires. Apparent Disharmony States disparities were resolved by returning to the Percentage of Projects subjects to clarify their responses and ratings. States Experiencing Each State This type of multi-method, multitrait measure- Mild Disharmony ment approach is commonly used to maximize Lack of interaction 7.6% the validity of social science measurements. Lack of communication 6.6 A profile of ratings was thus developed for Too-good friends 6.3 each of the 289 projects. Some of these profiles Subtotal 20.5 appeared to be very similar; others appeared to be very dissimilar. Statistical cluster analyses Severe Disharmony techniques were then applied to the profiles in Lack of appreciation 26.9 order to exhaustively cluster the projects by tl1e Distrust 11.8 various types of profiles that were found [2,4]. Sllbtotal 38.7 Disharmony total 59.2 Seven R&D/Marketing Interl'ace States Harmony Using a 95% statistical significance level, seven Equal partner 11.7 different clusters were found from the cluster analysis. Each cluster was then labeled according Dominant partner Harmony total -29.1 40.8 to its ob,,;;:. ,;;.1 items. For example, a review of Overall total 100% the items in one cluster showed that it was &D/MARKETING IN'rERFACE IN NEW PRODUCTS I PROD INNO~ MANAG 9 , ]988;5:6-19 v-ere labeled 'severe'. Follow-up studies showed needs. This occurred because neither party felt nat these types of problems were not easily the other had much information of significant Ivercome, they usually ~aused operating dis- value. And neither felt it was important to inform uptions, consumed many hours of managerial the oth~r 0f the details of their own work. This filent in moderating disputes, delayed key ?C~­ state wae: aptly summed up in the comments of lons and important decisions and led to pr0Ject one respohdl"nt. "If we told them aJ! tit!IS, they iillures. wouldn't k~10W what to do with it.... We know Many other projects were found that did not more abo'.l E than they do. Our best source of xhibit either mild or severe disharmonies. As information ('omes ~rom right here, from our- hown in Table 2, these projects were considered selves." Note how this state of affairs is different o be in a 'harmony' state. from the above Lack of Interaction syndrome, where the perceived urgency of pursuing their own activities caused the parties to neglect each ~haracteristies of the Mild other. Here, both parties harbored negative feel- )isharmony States ings about the worth of the other that stood in the wa} of interaction. ,ack of Interaction Though various causes of the Lack of Commu- n this state of affairs, there were very few formal nication state were observed within the data nd informal meetings between the R&D and base, two experiences repeatedly lead directly to larketing personnel. Both parties were deeply this problem. One was the perceived theft of oncemed with their own narrow specialties and credit. When either party took what the other.either saw any reason to learn more about the thought was undue credit for meritorious project ,ther's work. Neither party saw the need for achievements, this inevitably led to a Lack of lose interaction. R&D expected marketing to Communication problem. The impression that se whatever they gave them, and marketing the other party had taken unfair advantage was xpected R&D to create useful products. long remembered. Another experience that fre- This state resulted more from simple neglect quently led to the Lack of Communication state ~all from any strong an;mosities between the was top management's uneven use of accolades. 'arties. For example, one subject noted: "You If top management praised one party and did not et busy and you don't stop to think about praise the other, rivalry invariably developed that {hether or not they should know about this or shut off some future communication. As one ~at.... when you have to get your part of the subject noted: "If we don't tell them anything, )h done." Another subject said: "If you don't they can't go to management and take credit for et used to seeing each other you don't miss each it. " ,ther, and if you don't think about each other you on't make any effort to get together. And you lways have to make an effort." It may be noted Too-Good Friends ~at several projects experiencing the Lack of In this state of affairs, the R&D and marketing nteraction state were in older, commodity prod- personnel were too friendly and mruntained too ct firms that were attempting to develop new high a regard for each other. They enjoyed each mduct lines. Most of these firms had no his!of- other's company so much that !hey frequently ~s of close R&D/marketing interactions. met socially, outside the work environment. Th'!se social affairs often included the individu- al's families, e.g., family picnics and Sunday ,ack of Communication afternoon socials were common. In most of the n this state, the two parties purposely main- Too-Good Friends cases, work and social aspects lined verbal, attitudinal, and physical distances were commingled, e.g., joint visits to customer rom each other. R&D purposely did not inform facilities might also involve a round of golf and larketing about their new technologies until very the Sunday afternoon socials always included lie in the development cycle. Marketing pur- some informal discussions of business. Each osely did not keep R&D informed about market party felt that the other had their own area of  W. E. SOUDER 10 J PROD INNOV MANAG 1988;5:6-19 exclusive expertise, and that the other was be- pression of disharmony and strife. Professional yond reproach. This inhibited each party from disagreement appears to be a very heaithy and challenging the other's assumptions and judg- enlightening climate for its members. At times, ments. Consequently, important information and such disagreements may seem to become very subtle observations were overlooked that were heated and destructive. Yet if these discussions significant for the project. are confined to the issues and do not become What factors led to this type of problem? personally threatening to the participants, they Surprisingly, past successes sometimes led the can actually strengthen the R&D/marketing inter- team members to become too-good friends. face. Thus, it is the lack of professional dis- Teams of R&D and marketing personnel who had agreement (Too-Good Friends) that constitutes worked together successfully for long periods of disharmony, and not its presence. time often became complacent. Their pctency appeared to decline once they had achiev~d com- plete harmony. Apparently, they needed some Characteristics of the Severe conflicts or the challenge of building harmoni(ms Disharmony States relationships to maintain their alertness. A re- lated factor was a kind of blind faith in the Lack of Appreciation correctness of the counterpart person. As one This state was characterized by strong feelings respondent observed: "You are always sort of that the other party was relatively u:-eless. M~r­ reluctant to challenge and question what your keting felt that R&D was too sophisticated, while colleague tells you. He's the expert in that area. R&D felt that marketing was too simplistic. Mar- And you don't expect that he'll play politics with keting felt that R&D should be prohibited from you, so there's no reason to question his integ- visiting customers because they would talk over rity. And you figure he's the best man you've got, their heads. R&D felt that marketing did not have so he probably won't steer you wrong." a good grasp of the market needs. In this state, A detailed examination of the other clusters of the marketing groups often purchased their R&D projects showed that past successes and great work outside the firm rather than use the in-house faith in each other also characterized effective R&D group. R&D often independently moved R&D/marketing interfaces, i.e., the Equal Part- ahead with its own ideas, by-passing marketing ner Harmony state in Table 2. What were the and attempting to launch their own new products. distinguishing fa..:tors? The answer appears to be The~e efforts seldom succeeded, and the failures a matter of interpersonal dynamics. The parties were usually rationalized by the R&D personnel to an effective interface always challenged and as marketing's fault for failing to assist them! penetratingly questioned eath ether. They ap- What caused the Lack of Appreciation? No peared to enjoy and thrive on this aspect, some- single cause was identified. Some cases had long times with impish good humor. When one partner remembered histories of ineffectiveness by one found a g?P in the other's logic, both partners party, e.g., R&D failed to develop the promised were suddeniy ener6ized to close that gap. Such product or marketing failed to correctly identify experiences further strengthened their relation- the market. Sometimes, the organizational cli- ship. The partner who committed the logic gap mates fostered a lack of appreciation. For exam- never seemed to suffer any loss of pr~stige in the ple, several respondents indicated that they other's eyes. Rather, the ambience was described "never see any signals from management that by one partner as "a climate where we look kr collaboration is desired." Other respondents flaws, and it's not important who committed the noted that "management has not indicated t'1at flaw. We just want to find it and work together to we are expected to cooperate with them." It is fix it." This is clearly a different climate from the interesting that management must make a special above Too-Good Friends state. effort to enC(lurage cooperation: it does not seem It should be noted that such professional dis- to be automatic. agreements and challenging hehaviors. that often The organi7:,tion of R&D and marketing into characterize effective R&D/marketing interfaces, separate depi~nments with separate budgets and may give the outside ohserver the mistaken im- 0~~rations often fostered a lack of appreciation. lD/MARKETING INTERFACE IN NEW PRODUCTS J PROD INNOV MANAG II 1988;5:6-19 ,s evidence, consider the following sampling of before I came into this group. So, YOil see, you atements from personnel at five firms in the really have to watch out for him." This type of ack of Appreciation state. "We don't have any institutionalized Distrust was found surprisingly IputS into their plans and budgets". "They have often. leir own operations and so do we". "We get our :wards from doing our thing and they get theirs 'Om something else". "No one is responsible for Characteristics of the Harmony States ()w it all comes together". "We just go our ~parate ways." Equal Partner Harmony In this state, each party appeared to share equally in the work loads, activities and rewa" :s. Each listrust party felt free to call joint meetings on almosL any listrust is the extreme case of deep-seated jeal- issue. These meetings were characterized by an llsies, negative attitudes, fears and hostile be- open give and take offacts, opinion~ any r~dings. iviors. In this state of affairs, marketing felt that No issues were left unresolved and consensus &D could not be trusted to foHow i"structions. was sought by everyone. Study committees and ~D [elt they were blamed for failures, but task forces with joint memberships were com- tarketing was credited for successes. Several mon, with the task force chairmanships rotated &D groupo; in this state feared that marketing between the R&D and marketing personnel. 'anted to liquidate them. R&D lamented that Moreover, it was part of the Equal Partner cul- larketing often attempted to dictate exactly ture to involve R&D and marketing personnel 'hat, where, when and how to do the project, jointly in all customer visits, customer follow- lowing no room for rebuttal and no tolerance for ups, customer service, new product planning and leir suggestions. Marketing lamented that when forecasting, project selection and product strat- &D got involved the project disappeared and egy formulation activities. ley never saw it before it was completed, at Three features were common to all the Equal 'hich point it was seldom what they wanted. Partner casl.!s. One, the marketing personnel everal cases were found where R&D initiated were technically trained. They all had undergrad- lany projects and kept them secret fr{J'll market- uate degrees in science or engine{!ring. Two, the 19 "so marketing wouldn't kill them before they marketing personnel had prior careers in R&D. lined enough strength to move along on their Thus, personnel Wl;re often successfully ex- wn. " Cases were found where marketing changed or rotated between the R&D and mar- ['Ought R&D into the picture only after the keting functions. Three, the R&D and marketing roduct specifications had been finalized "in or- personnel had a strong sense of joint partnership. ::r to avoid any (I.rguments from R&D about how As evider~ce, note the following sampling of do it". quotes collected from R&D and marketing per- What caused the Distrust state? Though no sonnel in Equal Partner states. "We couldn't get ngle cause was found, several important contri- along without them". "We're on the phone with Jting factors were isolated. AH the Distrust each other constantly". "I feel like I've known lses began as either a Lack of Appredation or a them a long time". "We've been through 'thick ack of Communication problem that evolved and thin' together." to Distrust. Many of the Distrust cases were laracterized by personality conflicts that top Dominant Partner Harmony anagement had allowed :0 exist for a long time. I some cases, these conflicts had become so In this state, one of the parties was content to let stitutionalized that even personnel who had not the other lead. Both R&D-dominant and market- ~en involved harbored feelings of Distrust. As ing-dominant cases were found. For example, 1 example. note the following quote from one one R&D subject in a marketing~dolTlinant case :spol1dent, referring to his counterpart in an- noted: "We have no idea at all what the market her depart nwr!l "He Ohl.::; did ~;ome thing~ to needs arc. But if they'll tell us what they want I. I'm not sure what they were. It all happened and supply the specs we can sure make it for 12 J PROD INNOV MANAG w. E.SOUDER 1988;5:6-19 them. A marketing respondent in an R&D- ished the effectiveness of the end products. In dominant case said: "We can usually sell what many of the projects experiencing the Lack of R&D gives us. We don't really know what they Communications problem, the new products ei- are able to come up with. They know what it ther did not match the market needs or failed to takes to make a good penorming product better meet some important customer specification. In than we do." about half of the projects with Lack of Interaction It may be noted that the dominant partner problems, the end products either did not per- cases seldom involved complex technologies, ex- form as originally planned or arrived too late to acting customer needs or large R&D efforts. capture a rapidly changing market. Thus, Mild Most of these cases involved developmental ef- Disharmonies generally depreciated the degree of forts as opposed to research efforts. This reinfor- success of the end products. But they seldom ces the notton that problems at the R&D/mar- resulted in dismal product failures. keting intf'rface escalate as the technology or the By contrast, in a majority uf the projects user's environment become more complex. experiencing Lack of Appreciation problems, the end products either failed to perform or they were not cost-effective. In many of the projects where Inddence, Severity and Consequences Distrust occurred, the products did not perform of Disharmony at all. Thus, Severe Disharmonies resulted in a As the percentages in Table 2 show, a surpris- high frequency of rather dramatic failures. More- ingly high incidence of R&D/marketing dishar- over, it should be noted that Severe Dishar- mony was found. Nearly two-thirds (59.2%) of monies were very difficult to overcome. Attempts the projects studied here experienced some type by management to ameliorate them through nego- of R&D/marketing interface disharmony. More- tiation, reorganization, bargaining or personnel over. it is especially disconcerting that over one- transfers often left deep scars and sowed the third (38.7%) of the projects studied here experi- seeds for a renewed outbreak of similar problems enced severe disharmonies. These results are elsewhere. Thus, the prognosis for firms experi- statistically significant at the 99.9% level of confi- encing Severe Disharmonies is unusually pessi- dence (using the binomial statistical test ). That is, a statistically significant number of projects Table 3. Distribution of Project Outcomes by were found to be experiencing disharmonies. And Harmony/Disharmony States a statistically significant number of these projects had severe disharmonies. PercLntage of Projects in Each State Exhibiting Each O..:tr ,linea But is disharmony disruptive to project suc- cess? Table 3 responds to this question. Most of Partial States Success Success Failure the Harmony projects succeeded. Partial success characterized the Mild Disharmony projects. And Harmony 52% 35% 13% most of the Severe Disharmony projects failed. Mild disharmony 32 45 23 Severe disharmony 11 21 68 As noted in Table 3, these results evidence a statistically significant relationship between the t statistic = 88.84, C =.61, significant at