Technological Entrepreneurship Integrating Technological And Product Innovation

Technovation 20 (2000) 115–127 www.elsevier.com/locate/technovation

Technological entrepreneurship Integrating technological and product innovation Anne Tomes a

a,*

, Rosie Erol a, Peter Armstrong

b

Design and Innovation Research Unit, School of Cultural Studies, Sheffield Hallam University, Psalter Lane Campus, Sheffield S11 8UZ, UK b Department of Management, Darwin Building, Keele University, Staffordshire ST5 5BG, UK Received 30 January 1999; received in revised form 15 June 1999; accepted 22 June 1999

Abstract Using case study material from an advanced chemical technology, this paper identifies two forms of connection between the developer/producer of a new technology and the manufacturers of consumer products. In the first, the connection is direct, whilst in the second it is effected through an intermediary company which has developed an expertise in tailoring the new technology to particular applications. Whilst the direct connection has resulted in a number of NPD failures, the mediated connection has been consistently successful. Through an examination of two NPD programmes within each chain of connection, the paper examines the reasons for these differential success rates, and discusses the implications for the organization of NPD within advanced technology fields.  2000 Elsevier Science Ltd. All rights reserved. Keywords: Technological entrepreneurship; Technological innovation; Product innovation; Networks; NPD chains; Intermediary companies

1. Introduction This paper is concerned with the distinction between technological and product innovation, and with the effectiveness of different forms of inter-company network in achieving the latter. The idea that the two forms of innovation may best be achieved through different forms of co-operation is not entirely new. An exploratory study by Moenaert and Caeldries (1996) suggests that technological learning may be improved by the facilitation of internal communication flows within an R&D team whereas market learning may be advanced by contacts with other R&D teams. The present — equally exploratory — study compares two product development programmes carried out within a company specializing in the development and production of an advanced technology with two carried out by a company specializing in applications of the same technology. Though the programmes in the producer company achieved a measure of technological success, they were, in spite of this, mar-

* Corresponding author. Tel.: +44-114-225-2661; fax: +44-114225-2603. E-mail address: [email protected] (A. Tomes)

ket failures. Those carried out in the applications company, on the other hand, were market successes. The failures in the producer company illustrate the importance for the success of NPD programmes of assigning due weight to product, as opposed to technological innovation. Where technologies are relatively stable, of course, this distinction is obvious, since much design effort in these areas goes precisely into the creation of products which are innovative in the sense of enabling new lifestyles whilst employing conventional technologies. According to Hoshino (1987) and Evans (1990), this is the situation in some areas of consumer electronics. In fields of rapid technological change however, the distinction between technological and product innovation may be obscured by the concurrence of the two. On the one hand NPD programmes may be set in motion by the opening-up of new technological possibilities, and on the other, programmes of technological development may be driven by new product concepts. Where there is a dialogue of this kind between technological and product development, it is easy to mistake the one for the other, particularly where the product of one company in a supply chain is a technology for its customer. As has been pointed out by Afuah and Bahram (1995), innovation may appear differently to the innovat-

0166-4972/00/$ - see front matter  2000 Elsevier Science Ltd. All rights reserved. PII: S 0 1 6 6 - 4 9 7 2 ( 9 9 ) 0 0 1 1 6 - 9

116

A. Tomes et al. / Technovation 20 (2000) 115–127

ing entity than it does to its suppliers or its customers. For example, a technology which is innovative from the point of view of its supplier may, for certain customers, be nothing more than another means of realizing a conventional product. The dangers of confusing product and technological innovation are illustrated through four case studies of new product development (NPD) in a field of rapidly advancing technology, ‘specialist chemicals’. For reasons of confidentiality we have had to suppress details of the actual technology. The case study approach was deemed appropriate for an exploration of the factors which might account for differential success rates in high technology NPD, since it enables complex causative connections to be traced through a number of forms of evidence (documents, artefacts, interviews and observation). On the downside, case study evidence can only be generalized in the form of hypotheses or theories rather than statistically validated findings (Yin, 1994). Two of the case study NPD programmes were unsuccessful in the sense that the company succeeded in solving a sequence of difficult technological problems, only to find that their product’s advantages over a simpler rival technology were insufficient from their customers’ point of view to justify its greater cost and handling difficulties. The assumption behind these NPD programmes appears to have been that the solution of a series of technological problems would automatically result in an innovative product. Bearing in mind Cooper and Kleinschmidt’s (1993) finding that the single most important determinant of NPD success is a product which is clearly different from, and superior to others in the same market, these failures show the importance of maintaining an end-user’s, as opposed to a technologist’s perspective on innovation. A second, related, objective of the paper is to explore the effectiveness of different forms of inter-company network in the NPD process, particularly in its crucial early stages (Cooper, 1988). The advantages and problems of networks as a means of pooling resources of technology, intellectual property and finance have been extensively discussed in a variety of high-technology industries — for example, pharmaceuticals [(Bower, 1993), computing (Noren et al., 1995), chemical processing (Hutcheson et al., 1995) and home automation (Tidd, 1995)]. This body of work, however, has yet to produce an accepted vocabulary for different patterns of inter-company interdependence. Consequently the question of which pattern might most effectively promote NPD has scarcely been posed. The present paper offers a beginning in this respect, since the data suggest that the relationship between a company specializing in technological development and the manufacturers of consumer products may be more effective when mediated by a company specializing in applications of the technology.

The two themes of the paper are connected in that the respect in which the mediated relationship was superior to the direct one was in the integration of product and technological development. In the mediated relationship, product concepts originated with consumer goods manufacturers rather than with either the developer of the technology or the application specialist, thus ensuring that due weight was assigned to the product innovation aspect. In this connection, survey data by Karakaya and Kobu (1994) show the advantages of using customers as the source of new product ideas, whilst the importance of involving customers in product design at a more general level has been confirmed by Gemunden et al. (1992), Shaw (1993) and Hutcheson et al. (1996). Besides assigning due weight to the product innovation aspects of NPD, the mediated relationship possessed the additional advantage that the expertise of the application specialist was focused precisely upon the integration of product and technological innovation. This is consistent with Iansiti’s (1995) finding that differences in NPD performance in fields characterized by discontinuous technological change are correlated with skills and routines aimed at technology integration which are successful in merging deep technical knowledge with a detailed understanding of the specific environment in which new technologies are applied. The difference between the cases reported here and those described by Iansiti is that these integrative skills were supplied by an intermediary company, rather than developed within the organization which developed the base technology. The implication is that this form of network may be an effective way of managing the integration of product and technological development.

2. Case studies 2.1. Multichem Multichem is a medium sized multinational company which produces a range of products within the chemical industry. The UK company has an annual turnover of around £100 million, and employs about 1000 people. The company’s involvement in specialist chemicals began about 40 years ago, in collboration with educational and defence establishments. The original chemicals that were developed were unstable and difficult to use, and further research work was carried out to produce a more stable form. They succeeded in developing a synthetic material which was much more stable and predictable in its performance than the original materials based on natural products. These new materials were seen as being novel, highly versatile materials with new optical properties which appeared to the company to have the potential for a wide range of applications. Although the synthetic version is still not easy to handle,

A. Tomes et al. / Technovation 20 (2000) 115–127

this achievement opened up opportunities for the development of new products. Currently, the development and production of specialist chemicals is carried out by the Specialist Chemical Group. This group, consisting of about a dozen research staff plus the marketing and production teams, is responsible for the development of consumer applications as well as the technology itself. We talked with staff for a total of nine hours, with the data coming from taped interviews with the Marketing and Business Development staff of the Specialist Chemical Group. Using case study material from an advanced chemical technology, this paper identifies two forms of connection between the developer/producer of a new technology and the manufacturers of consumer products. In the first, the connection is direct, whilst in the second it is effected through an intermediary company which has developed an expertise in tailoring the new technology to particular applications. Whilst the direct connection has resulted in a number of NPD failures, the mediated connection has been consistently successful. Through an examination of two NPD programmes within each chain of connection, the paper examines the reasons for these differential. It will be evident that Multichem’s early presence in the market for specialist chemical applications was built on successful R&D programmes. Perhaps for this reason, and perhaps because the supply of scientific apparatus to universities and colleges is still a major part of the company’s business, there remains a strong element of R&D-push in Multichem’s search for consumer markets: ... a lot of new products come out of R&D. It’s people playing with the vivid effects, and thinking we must be able to do something with this. And then it’s given to the marketing department to do something with it. Multichem began to combine different chemical components into mixtures to produce new effects. Unlike research laboratories, however, commercial customers could not be expected to master the delicacies of handling the specialist chemicals, since they are very sensitive to contamination. However, if these new effects were to be exploited in commercial applications, it was necessary to find some means of insulating the specialist chemical component from other components of the final mixture. The answer, developed over a number of years of R&D work was to process the specialist chemicals further. This solution therefore, represented a major R& D investment for Multichem before the search for consumer applications could even begin: We were making specialist chemicals, and then because everyone said they found them so difficult to use, we took it on to the next stage ourselves, and started encasing them. Many other people who have

117

wanted to use these wonderful effects couldn’t do the modification. Now that’s not surprising, it’s very tricky. So we started talking to ink makers, and paint makers, surface coatings people. “Here’s some encased chemicals. See if you can make some inks and some surface coatings.” This was pretty hard work, because most of them still couldn’t use the encased chemicals. Basically what they were doing was taking them, sticking them into an existing system, and hoping that would make the product. Now that would never work, and it never did work. So we employed ink specialists, and we learned how to make inks, so we could basically do it ourselves. Having developed its range of inks, Multichem began looking for consumer applications. The Business Development Manager describes the sense of anticipation at that time. It was really a very exciting time because we had these [inks] which people were able to use, just about, with a lot of technical assistance, but they could do it. The general understanding was that only volume markets were worthwhile. Because of the problems of handling specialist chemicals, even in capsule form, niche markets would have demanded disproportionate customer support. Each market you sold the specialist chemicals to required a large amount of technical back-up, due to the difficulties of working with them. Working with a lot of smaller markets would have required too much technical assistance on the part of the company. The markets which were considered were all large enough markets to justify the company undertaking the products.

2.1.1. Clothing The idea behind the clothing project was to print garments with the special inks. Clothing manufacturers were contacted by the marketing department to obtain an initial reaction to this product concept. This was favourable, and so the decision was made to go ahead. This immediately set the agenda for an R&D programme: the further development of the specialist chemical inks so that they would be suitable for screenprinting onto textiles. Washability was a problem: ...the water resistance was generally atrocious, after five washes you’d basically got rid of it... so we had to do a lot of research into how we could make these screen printing inks stable enough to put onto fabric. We spent a lot of time with cross linking systems to try and get it washable.

118

A. Tomes et al. / Technovation 20 (2000) 115–127

—and the loose weave of cheap mass market clothing fabrics: The thing about [these materials] is that they are generally very low quality, very stretchy... with lots of holes in. And we had an ink which didn’t stretch and didn’t wash very well... also the [brightness of the] effect depends very much on getting a single layer of the chemical on a black background. If you’ve got holes everywhere, you don’t get this single layer..., and you don’t get the pure effect. —and the contamination of the specialist chemicals by the other ingredients of screen-printing inks: ... the ink system we have had its limitations, you can’t just stick your normal additives into the ink system, you can’t use the normal binder systems... so you’re very limited by the ingredients you can put in. —and, for a cheap, mass-produced clothing, the cost of the specialist chemical-based inks was a chronic anxiety. In spite of all these problems, the project pressed ahead: So the market situation was wrong, there were [competitive products] at the time, the cost was too much, and there were technical problems as well. But we weren’t to be stopped on that one... basically the size of the market was very attractive. [So] we spent more on R&D... To some extent the problems of washability and of printing onto stretchy fabrics were solved. ...It was really very close to what we needed. You still only got ten washes out of it, and they all had to be hand washed. You couldnt stick it in a machine. So it was getting a bit dodgy for the volume market, but people liked it. In the light of the continued high cost, however, the limited washability which had been achieved was beginning to look insufficient: If it’s a very cheap sort of material, then OK, ten washes is fine, but with it being a specialist chemical product it wasn’t cheap... we were really having to charge quite a lot of money.

thing ever, but when we actually came to do it, they all went a bit lukewarm. You can describe [the effect] on the phone, and they get very excited about it, but when they actually see the end product... it’s not like they’d been expecting. So there was always that disappointment factor. What finally killed the project was competition from a rival technology. Not anticipated at the launch of the specialist chemical development programme, these dyes turned out to be cheaper, easier to apply and gave a better visual effect. It’s basically a different type of material... And they are easier to use generally... I wouldn’t say they are any more stable, they’re just [give a better effect], easier to use...so they came at just exactly the same time as we were trying to make people interested in [our product]. And they came in at a low price.

2.1.2. Food labels A special effect food label was another product concept which emerged from Multichem’s search for mass markets for their inks. The idea was proposed at a joint meeting of the R&D and marketing departments: ...Sure enough we started talking to the food people, they said, ‘Yes, that’s what we want. We must have it.’ On the basis of this positive initial response to the concept, Multichem began work on developing inks and a printing system suitable for this application: Screen printing is slow, so we had a real challenge. We had got to get our ink systems into a new printing technology. And the only way we were going to be able to develop a new printing technology was by going back to encasing the chemicals. All the faster cheaper printing techniques were really putting down much lower film weights of inks. So you need much smaller chemical units. So we had an intensive R&D campaign to reduce their size.

Meanwhile, the progress towards an achievable product on the technical front appeared to have the effect of clarifying the customers’ requirements:

Besides reducing the size of the capsules, Multichem needed to formulate a chemical mixture which would attract the attention of the consumers, as well as produce the desired effect. This specialist chemical mixture is a very particular mixture. They spent ages working on this. The problem of printing the labels at speed was tackled by forming alliances:

When you talk to people about [this product] they get really excited, saying it’s the most wonderful

... we teamed up with an ink manufacturer, and a label manufacturer. It was quite easy to persuade

A. Tomes et al. / Technovation 20 (2000) 115–127

them, when we told them what we were planning ‘We’ve got a huge market here.’ So I teamed up with these guys, and between us we developed a new ink. And basically it worked brilliantly. We were concentrating on being experts on the specialist chemicals, they were concentrating on being experts on the ink. And that way we worked really very well together. On this basis, Multichem succeeded in developing what appeared to be a saleable system: And when we had a system that we were both happy with, we went and practically knocked on the door of the biggest food manufacturer in the world, well the UK arm of it, and said ‘Hi guys, this is who we are, this is what we’ve done, we think maybe you’ll be interested’. And they were. They turned round and said, ‘Yes, we’re very interested, we basically want to use this as a marketing tool.’ At this point, the food manufacturer became sufficiently committed to involve itself in the development of the printing process: ... they supported the work. They bought the ink. They really pushed it. We ran trials at the biggest printers in the UK. We started to move into the huge league, printing millions and millions of these labels, so everything became a lot more serious and a lot more expensive. At this point, there was considerable excitement at Multichem. There was talk of building a new plant if the product took off. Against this, there were still fundamental, and apparently intractable problems: We were still not too happy with the visual effect, and the technical problems were really that it was a water based ink. They were used to solvent based inks. Water based inks take a lot longer to dry [and] what they found was... they had to slow the machine down, which they didn’t like. Also they couldn’t get enough [ink] down with one hit. You had to put it on twice. So you had to print it, dry it, print it and dry it, and also try and print all the other solvent based colours on the same sheet. Still keen on the concept, the food manufacturer began evaluating the same competitive technology which had defeated Multichem in the previous project. At the time, Multichem believed that their own specialist chemicalbased labels had sufficient technical advantage: We thought we were going to be OK with the [competition this time], because... [they couldn’t pro-

119

duce such a controlled effect]... So we thought, ‘OK, we’ve really got the edge here.’ Unfortunately for Multichem, this turned out to be a misreading of the manufacturers’ real needs. In the light of the advantages of compatibility with solvent-based inks and speed of printing enjoyed by the other technology, their shortcomings turned out to be acceptable after all: But at the end of the day... [the competitive technology was] cheaper and easier... And in the end we ended up running trials side by side, with the [competitor] company. And the story continues that [the food manufacturer] then decided to go with the other method. And the product went into the shops. The business development manager reflects on the experience: The market opportunities were real. You can tell by the [competitive technology] going in and taking the market from us. It’s a low quality product, but at the end of the day, it did what was necessary. You can say ours is a much better product, much more sophisticated,... and much cleverer, but they’re not interested in all that stuff. I think one of our failings has been concentrating too much on how clever we are and what we can do with our wonderful specialist chemical, we really are focused on these. If we were more a marketing company than a chemical company, we would have found a way of securing an agreement with the manufacturers [of the competitors technology]...but it wasn’t the core technology for us... and we wanted to just concentrate on the specialist chemicals.

2.2. Chemtech Though a small company, with only 60 or so employees worldwide, Chemtech headquartered in the US is the product of a complex series of take-overs, mergers and demergers. This is well illustrated in the story of how the company originally became involved with its principal specialist chemical technology. About 25 years ago, a precursor company — then a specialist stationary manufacturer — incorporated a novelty specialist chemical product into one of its standard products. The popularity of this item encouraged the company to develop similar products for other applications. By the 1980s stationary had been quietly forgotten as Chemtech’s interest in specialist chemical technology diversified into the more technically demanding applications. Currently the company supplies specialist chemical products into

120

A. Tomes et al. / Technovation 20 (2000) 115–127

a wide range of application areas though the tradition of novelties lives on in the form of promotional products. Chemtech is one of the industry’s successes. Current worldwide annual turnover is about $10 million, increasing at about 20% per annum. The UK operation was set up about 5 years ago and currently employs 15 people, many of whom have worked for Multichem at some point in their careers. Though the company has interests in other technologies, about half of the UK employees are directly involved in specialist chemicals. The form of this involvement is rather different to that typical in Multichem. Chemtech neither develops nor manufactures specialist chemicals on its own account. Most of its raw materials, in fact, are purchased from three or four chemical manufacturers including Multichem. Rather, it is a company which has built up a particular expertise in formulating specialist chemical mixtures which will deliver the properties required in consumer applications. Surprisingly perhaps, in a science-based industry, this expertise has been acquired through experience, rather than formal scientific study. Our data are from interviews, totalling 10 hours, with the company’s research, marketing and production staff. The quotations are from taped interviews with the UK Director: The industry [is] founded really on empiricism. People go out and find a bunch of chemicals, and then mix them or process them in some way or another, and the ones that work you keep, and the ones that don’t you throw away. So if you go and search the patent literature or the open literature the isn’t very much written about anything that anyone does in the industry. There are patents there if you read them [which] tell you some rudiments, but there’s nothing of any significance if you were to look seriously at getting into the industry. So there’s quite a significant barrier to entry based on the high level of empiricism.

(1994). The following examples are typical of the history of Chemtech. We added a downstream processing capability in 1982 when we hired a guy from [one company] which was a competitor... We also hired another guy who was doing [work for] medical [applications]. So he had some experience of novel specialist chemical formulations when he came to Chemtech. Though the company is well aware of the importance of marketing the capabilities built up in this fashion, the industry as a whole has a reputation for uncertain quality. For this reason the company’s ability to develop specialist chemical applications is best publicized by results and by the education of potential customers rather than advertising as such: We don’t do a lot of advertising... you’ve got to go through an education process to build an new image for the products and the markets... a lot of our competitors actually make products where the quality actually isn’t what it should be, so you get a raw deal in terms of the image in the marketplace. Particularly in the UK market, the consumer level [of expectation] is a result of our efforts to ensure the quality of the products that we supply. A very important consequence of the reputation built up by Chemtech is that product concepts normally originate with its customers rather than the company itself: Interviewer: So what’s the process then? Does someone come to you with an idea, or do you have an idea for a new product? No usually people come to us with ideas. We’ve got a good track record of making big things happen.

This means that Chemtech’s ability to adapt specialist chemical mixes is relatively unique within the industry: ... people that are in the industry who compete with us basically rely on their raw material suppliers to provide them with technical support. None of them really have the capability to develop anything that is outside of the ordinary. The two basic processes through which Chemtech has built up its portfolio of capabilities are acquisition and recruitment. In an industry characterized by great deal of churning, this demands opportunism informed by active networking within the industry. This ability of innovating entrepreneurs to use personal networks and prior relationships to glue together sources of knowledge and expertise has been noted by Lipparini and Sobrero

2.2.1. Toy blocks This happened in the case of a multinational producer of toy building bricks. The idea was to include a part which used the special effects produced by these chemicals in one of their toy sets. Chemtech became involved as a result of the toy manufacturer’s search for a partner who could develop such products: Large companies undertake global searches for people within this business... And they scan through the various trade publications, publications about company services and capabilities, and they’ll end up with a shortlist of about a dozen people that might be worth contacting around the world, who appear to know something about the technology.

A. Tomes et al. / Technovation 20 (2000) 115–127

In its initial approach, the toy company had in mind a fairly definite specification for the brick. It is an important part of Chemtech’s approach, however, that this was treated as a basis for negotiation rather than as an unalterable NPD target. The aim was to reach an acceptable compromise between the customer’s requirements and what could realistically be achieved through the formulation of specialist chemical mixtures: I think what they wanted was basically somebody who understood what they wanted to try and do. When we first set out to describe what was possible, we had to persuade them that they don’t actually need the device to satisfy every one of their design prerequisites all at the same time. You end up working backwards from the ideal product to a product that is going to be acceptable to them that can be produced to meet their tolerances, is technically possible and fits also their budget...you have to communicate with [the customers] and let them understand the technology of your business. If they understand, you’re not just fobbing them off with something that’s second best, you’re actually taking what’s available and customizing it to what they particularly want. To Chemtech’s director, this initial negotiation is a key phase of the NPD process: Customers usually ask us to do something which is close to being impossible, on the boundary between what you can and cannot do technically. You’ve got to bring it back inside that boundary as far as you can so that you can give yourself a realistic shot of producing the thing reproducibly, and hitting it, not spot on every time, but certainly within the levels of tolerance that are acceptable to the customer. Everything that Chemtech has ever done with big multinationals has always involved some kind of dialogue with the customers. You listen to what they tell you to find the best shot you can give them. Because the negotiation, on Chemtech’s side, is based on an assessment of technological feasibility, it also serves as a vetting mechanism. The company does not involve itself in speculative R&D programmes: ... while you take the company forward, you have to turn your back on a lot of things as well. One of the reasons Chemtech has been successful is that we turned our back on more projects than we decided to pursue.

121

At the time of our research, the toy project looks like a success for Chemtech. This is not only important in itself: it will also feed back into the company’s reputation-based approach to marketing: This [toy company] thing is a prestigious piece of work. It’s just [these chemicals] on a unit, a little widget which produces a certain effect. It’s not that sophisticated a product really when you look at it, but it’s made to a very demanding quality. Our ISO9000 qualification was an important factor in taking the project forward. The new product’s going to be launched in July, well we’ve made a few shipments, but when that appears on the market it will open people’s eyes to the fact that high spec devices can be made to satisfy the most demanding clients using [this] technology. That’s one of the things which our competitors around the world would find very difficult to do.

2.2.2. Smart indicators Chemtech’s involvement in smart indicators began in much the same way as the toy brick project. A development company producing products for the catering industry thought up the concept of incorporating ‘smart’ chemicals into food packaging, which can react to the condition of the contents. This company too carried out a scan for technological expertise: The most important thing in the history of that project was first of all they asked [us] to do something, ‘Can you make an indicator’ [that reacts to the state of the food] and I said, ‘Well, once you sort the... technology....’ Interviewer: So they came to you in the first place? Yes, they basically found us, and they were talking to other people at the time saying ‘This is what we want you to try and do for us, is it possible to do it?’ I think they undertook a global search... they basically found us, and they were talking to other people at the same time, saying this is what we want you to try and do for us, this is what I want to do, is it possible to do it?

Interviewer: So what factors influence your decision about whether or not to follow an idea?

Chemtech’s approach was to assess the requirements of the development company in the light what was achievable through specialist chemical technology. Having done so, the company proceeded to the ‘working model’ stage:

Well I think a lot of it’s gut feeling... the company depends a lot on me to make the judgement calls on what’s worth chasing and what isn’t worth chasing.

I said, ‘Well what is the... precision? What sort of tolerances have we got?’... And again the key thing there was it’s not possible to buy a chemical mixture

122

A. Tomes et al. / Technovation 20 (2000) 115–127

off the shelf with the right properties [so] the most important thing was to actually make a prototype to... see whether it works. And we did that. The next problem was mass production. The solution drew on Chemtech’s unique ability to formulate specialist chemical mixtures: So now the issue becomes we want to make something like three million or five million of these things every month. How are you going to apply them automatically to the packaging? [The solution] was to produce something that was on a roll that could be applied automatically. It’s not the easiest thing to do because we had to adapt the mixtures for one process to a different technology. And we did that eventually, and unfortunately it wasn’t quite as good as the original one, but it worked. I think again we had about 10 different chemicals, 10 or 11 in the first one, and six or seven different chemicals in the second one. And none of the chemicals were the same in either mixture. Again that exemplifies the type of thing that Chemtech can do that our competitors can’t, and also it exemplifies the technical approach that we have to making new products possible. And to the best of my knowledge our competitors couldn’t even get close to doing that. As with Multichem’s projects, competing technologies were a potential alternative route to the food packaging indicator. Partly because of the company’s interest in this competitive technology, however, Chemtech’s director was aware of this alternative throughout the project, and remained confident that the advantages of their own specialist chemical technology would prove decisive in this application. I think there are some applications where the [alternative] technology is better... So there are various areas where it would be difficult for our own technology to compete with alternative technologies. But for the most part, our principal business is geared around the advantages of our materials, which is to pursue applications which generally speaking would not be possible with any other technologies.

3. Discussion It is tempting, at this stage, to make a straightforward comparison between the relatively successful NPD process at Chemtech and the problems encountered by Multichem. Clearly the processes in the two companies differ in the source of the product concepts, the preliminary evaluations of their prospects and the monitoring

of competing technologies. On the basis of such a comparison, it would not be difficult to put forward explanations of the companies’ different success rates, pointing towards the conclusion that companies competing in fields characterized by rapidly evolving technologies ought to do it like Chemtech and not like Multichem. The difficulty with this apparently logical approach is that Multichem and Chemtech are not altogether independent. Although the companies compete as the suppliers of specialist chemicals adapted to specific new products, these chemicals are developed from generic materials supplied by Multichem, amongst other manufacturers. Whilst Chemtech is not locked into its relationship with Multichem, nor even into the particular technology, the fact remains that its present reputation as a developer of specialist chemicals for consumer applications depends upon the existence of suppliers of the ‘generic’ chemicals. In a sense, therefore, the problems of Multichem — at least insofar as they stem from its situation as a supplier company rather than from strategic ‘errors’ — are part of the cost of Chemtech’s relative success. For this reason, we believe is better to view the NPD processes within these two firms as part of larger chains of relationships connecting the development of specialist chemicals with the consumer end-point. From this point of view, the case studies are windows into two different product development networks. The first (pattern 1) is a direct connection, in which the producer/developer of the base technology sets out to adapt it to the requirements of product manufacturers. In the second (pattern 2), this connection takes place through an intermediary company which has developed a specialist expertise in the adaptation of the base technology. It is the effectiveness of these two forms of connection between technology and application which is in question, rather than the performance of the two companies concerned. 3.1. Pattern 1: the direct connection Pattern 1 began with the physical properties of the specialist chemicals. Conceiving of the problem as one of creating mass markets for their products, Multichem set out to develop a generic method of applying them to consumer items. Thus the initial development of inks based on specialist chemicals preceded the search for specific applications of these inks. Once viable inks had been produced, the company’s R&D and marketing staff looked for existing mass-produced items (clothing and food labels in the cases described above) which might be enhanced in some way by the chemicals’ properties. The resulting product concepts were then tested against the reactions of relevant manufacturers. Where these were positive, Multichem then launched into a secondary adaptation of the specialist chemical inks and printing technology to the chosen consumer products. Though

A. Tomes et al. / Technovation 20 (2000) 115–127

this was done in partnership with ink and label manufacturers in the case of the food labels, most of the R& D thrust still fell within the field of specialist chemical technology. Consequently it took place within Multichem itself. Provided the interest of the consumer product manufacturers was confirmed, this stage of development was followed by a tertiary phase aimed at developing a fully commercial product and production process. In both of the case study applications, it was during this third phase, and only in the light of the emerging potentialities of the competitive technology that there occurred a clarification of the users’ requirements. In both of the cases described above, this clarification proved fatal to the specialist chemical technology (preference for bright colours, approximate temperature response). 3.2. Pattern 2: connection through an application specialist Pattern 2 began with an approach to Chemtech from a consumer product manufacturer. This meant that two very important preliminary stages occurred within these consumer product companies rather than in Chemtech. Firstly a relatively stable product concept had been developed, which called for properties achievable by specialist chemicals. Secondly the environment had been scanned for suppliers of suitable technologies, as a result of which Chemtech had emerged as a front runner. On these bases, Chemtech could assume a degree of commitment on the part of the manufacturer. Because these processes, as well as the establishment of Chemtech’s reputation, occur offstage in the cases reported above, it is all the more important to be aware of them when considering the requirements for successful NPD within the Pattern 2 connection. Chemtech’s reaction to approaches from interested product manufacturers might fairly be described as argumentative. Unwilling to embark on R&D programmes which involve major extensions of current specialist chemical capabilities, Chemtech aimed to negotiate achievable performance targets before committing itself to development work. This kind of compromise, of course, is not always possible, and it was important to Chemtech’s NPD success-rate that many projects were abandoned at this stage. Those which proceeded were developed on the basis of Chemtech’s expertise in manipulating properties by mixing the basic specialist chemicals. In an area in which the received scientific understanding is not always a reliable guide to the properties of chemical mixtures, expertise of this kind is partly experience-based. As such, it is expensive to acquire and relatively unique to the company — or rather to the individuals presently employed in it. This characteristic of the technology adds to the confidence with which Chemtech can pro-

123

ceed with their development programmes. Customers are unlikely to find alternative sources of the company’s expertise. The actual R&D process in Chemtech is brief, specific to particular manufacturing process, and, like the product specification, subject to negotiation. In the case of the food packaging, it is noticeable that there was a compromise between the manufacturer’s preferred application process and Chemtech’s estimate of the capabilities of the specialist chemical indicator. Whereas the customer’s specifications appeared to become clearer and more rigid for Multichem at this late stage of development, for Chemtech it seemed to become more flexible. 3.3. The two NPD chains compared 3.3.1. Customer involvement in product concept 3.3.1.1. Pattern 1 Although Multichem’s marketing department collaborated with R&D in originating the product concepts, it was a marketing department dominated by research-minded chemists. This meant that the company’s product concepts tended to combine a highly professional knowledge of specialist chemical properties with a decidedly amateur understanding of the relevant markets. Though the company was aware of the importance of marketing, this awareness took the form of testing the product concepts produced within the company against the reactions of relevant consumer product manufacturers. Whilst these reactions were probably authentic, the market sector from which they were obtained had already been defined by an initial ‘R&D’ view of the market. The pattern 1 marketing process, in other words, tended to over-define the question and customers from whom an answer would be obtained. Certainly it failed to draw on the marketing expertise of the consumer product manufacturers. 3.3.1.2. Pattern 2 Chemtech’s product concepts originated with consumer product manufacturers. Since these were large companies, the anticipated competitive advantages of an specialist chemical-based product were likely to be informed by a relatively sophisticated understanding of the market. Thus Chemtech could be fairly sure that a technically successful NPD process would also be commercially successful. Besides this advantage in market intelligence, there is evidence that customer involvement makes for a more efficient product development process. Surveys in the manufacturing, medical instrument and food processing industries consistently indicate that the use of the customer as a source of technical ideas is associated with NPD success (Karakaya and Kobu, 1994; Gemunden et al., 1992).

124

A. Tomes et al. / Technovation 20 (2000) 115–127

3.3.2. Mass markets vs niche markets 3.3.2.1. Pattern 1 Multichem’s product concepts originated in-house. Since the company was fairly large, this meant that they were driven by the need to find outlets for a large-scale production of specialist chemicals. This requirement tended to translate into a scan of (existing) mass products for those which might be enhanced by the chemicals’ properties. Unless the properties in question were unique to the specialist chemicals, therefore, this strategy virtually guaranteed that the company would encounter competition from other technologies. If some feature of a mass product is obviously attractive, it is more than likely that that others will be looking for ways of providing it. It should be pointed out that Multichem has learned from its experiences of competition from apparently inferior technologies. In the case of the clothing application, the company believes that it could have established a partnership with the competitor, so as to exploit the potential of the alternative technology to the advantage of both. 3.3.2.2. Pattern 2 The product concepts suggested to Chemtech by its potential customers (a food indicator and toy brick) might well be have appeared to Multichem as the kind of ‘novelties’ to be ignored in the search for mass markets. The fact that the market for Chemtech’s food indicator later expanded to 3–5 million units annually, could scarcely have been predicted at the outset. The search for mass markets built into the concept stage of Multichem’s NPD process was virtually guaranteed to overlook ‘hidden’ markets of this kind. To some extent, this was a function of the scale of the companies: since Chemtech could prosper in markets too small to support Multichem, it could not only develop products intended for these markets, but also reap the rewards if and when those markets expanded. 3.3.3. Customer commitment 3.3.3.1. Pattern 1 Although Multichem recognized the need for customer commitment after their problems with the clothing application, this was still only achieved at a late stage in the development of the food label. The prototype application was developed unilaterally by Multichem on the basis of nothing more than an expression of interest on the part of potential customer. Even when an achievable process had been demonstrated, the customer’s involvement in developing it to the production stage was largely limited to setting design specifications and running manufacturing trials. This lack of customer commitment to the specialist chemical application meant that it tended to remain on trial throughout the NPD process. Its acceptability depended

on meeting comparatively rigid performance specifications. This was probably another consequence of Multichem’s strategy of looking for applications in existing mass markets. Almost by definition, the production technologies in such markets will be highly-developed and stable, so that any modification would incur considerable costs. As a result, there appeared to be no possibility of a compromise between the consumer product specification and the properties of specialist chemicals. Noticeably, in fact, the requirements of Multichem’s customers became clearer, and more demanding, as development work proceeded. Again, this problem has now been recognized within the company. The current policy is that NPD work only progresses in collaboration with lead users who are prepared to undertake an equal risk in the development process. 3.3.3.2. Pattern 2 Chemtech’s director was keenly aware that an approach from, rather than to, a customer signaled commitment to the success of a project: ... you know, people’s careers are made on bringing the products we are able to produce for people to market successfully. I mean look at [an electronics-based manufacturer], the first time round they failed to bring the thing to market they ended up firing a few of their research engineers. And I think this is happening with [one company] at the minute in the cosmetics industry. These high volume, high exposure deals are very important to the people that we’re dealing with in the company as our customers, and they can’t afford to drop the ball, because their careers are on the line. Shaw (1993) has noted the importance of product champions in customer companies in facilitating learning by innovating entrepreneurs and in diffusing the costs of NPD programmes. In the case of Chemtech, this customer commitment also meant that their customers were more likely than Multichem’s to be flexible on performance specifications. For this reason, Chemtech was able to negotiate the initial targets for the NPD process, and to continue the dialogue ‘from first contact to first order’, as the Director put it. 3.3.4. Vetting 3.3.4.1. Pattern 1 Vetting consisted of two phases. The first was a preliminary canvass of the product concept amongst potential customers. Where this was positive, the company proceeded to the ‘working model’ stage of product development, thus guaranteeing that the costs of this would be incurred every time a customer was attracted to the concept. The second phase of vetting was tacit in the customer’s trials of the performance and

A. Tomes et al. / Technovation 20 (2000) 115–127

manufacturing compatibilities of the prototype product. In the cases outlined above, NPD on pattern 1 was abandoned only at the point where the product failed these trials, having incurred all of the costs of development work up to that point. Thus vetting in pattern 1 was, effectively speaking, in the hands of the customer. The anticipated difficulties of developing appropriate chemical mixtures appear to have been of little account. They appeared, indeed, to have been regarded as a scientific challenge, of which more in a moment. 3.3.4.2. Pattern 2 Many possibilities were rejected at the concept stage. Chemtech would only embark on NPD on the basis of a ‘gut feeling’ (i.e. experience-based knowledge of what can be achieved by mixing the basic chemicals) that the development track would be short and successful. In this sense the company was conscious of operating at the ‘low tech end of a high-tech field.’ Given that the initial approach in pattern 2 was from customers, and that this signified that that a potential market existed, this meant that vetting was on the basis of technological viability, that it was carried out by the producer company, and that it mostly took place before development costs were incurred. 3.3.5. Doing science vs doing new products 3.3.5.1. Pattern 1 NPD was driven by the R&D mentality. This was probably a function of the history of Multichem, as well as the background of its staff. Much of the initial growth of the company came from supplying scientific apparatus to universities and colleges, and this was still an important side of the business. Appropriately — from the point of view of this side of the business — many of the staff, including the marketing staff, had higher degrees in chemistry. Even in their discussions with customer companies communication tended to be between R&D departments rather than through marketing. What ran like a thread through the Multichem interviews was a relish for challenging R& D problems. Instead of prompting re-evaluations of the NPD programme in which they occur, it seems that the difficulties only added to the company’s determination. Entirely appropriate to an R&D team tasked with the development of a new technology, such an approach is more questionable when carried over into an NPD strategy. It is an attitude which guarantees that expenditure on each NPD programme will end only with its complete success or complete failure. 3.3.5.2. Pattern 2 NPD was driven by the intersection between the customer’s product concept and Chemtech’s expertise in mixing specialist chemicals. The director of Chemtech was conscious that the firm’s approach lacks the glamour of ‘real’ science. NPD tracks appeared to be short, highly specific to the customer’s product (rather

125

than aimed at a generic mass market), and probably inexpensive. In the long run, of course, the commercial successes of Chemtech within pattern 2 could be argued to be dependent upon the scientific knowledge of specialist chemical properties produced within pattern 1. Amongst other things, this was signified by the fact that many of Chemtech’s staff had previously worked for Multichem. From this point of view, the problem might be seen as one of the distribution of the fruits of success within pattern 2, rather than the failures of pattern 1. 3.3.6. Reputation 3.3.6.1. Pattern 1 Multichem’s reputation is as a pioneer and a leading producer of specialist chemicals. Within this field, the company is regarded as a major repository of expertise. Its track record, however, is in developing technological capabilities, rather than the incorporation of these into successful products. Having entered the field of product application as a means of creating markets for its chemicals, the company finds itself faced with the task of establishing a reputation whilst it is still developing the appropriate expertise. Though this is no easy task, its willingness to reflect on its experiences and change its mode of operation are grounds for optimism. 3.3.6.2. Pattern 2 Many of Chemtech’s competitive advantages flow from its track-record in the adaptation of specialist chemical mixtures to customers’ requirements. It was on the basis of this reputation that the company was approached by consumer product manufacturers with concepts to which they were committed in advance. From this it followed that the company was in a position to choose the post promising development projects — and so on. Whilst our case studies clearly illustrate the advantages which flowed from Chemtech’s reputation, they cannot, in their nature show how it was created and maintained. This happened over a long time-scale and in advance of the events described in this paper. It is important, nevertheless to bear in mind the key role played by reputation in Chemtech’s continued success. Without it the advantages of specializing in adapting the base technology to particular products would have counted for little. To put the matter another way: the functional advantages needed to be converted into reputation before they could become actualized as competitive advantage. As Bidault and Fischer (1994) have pointed out, the uncertainties for customers in buying-in unknown technologies means that the reputation of the supplier (established in Bidault and Fischer’s came through prior business contacts) becomes very important to the purchase decision.

126

A. Tomes et al. / Technovation 20 (2000) 115–127

4. Conclusions Evidence from case studies is, in its nature, indicative rather than conclusive. The two successful and two unsuccessful NPD programmes compared this paper, however, point towards two main conclusions. The first is that companies with a successful record in the development of a technology may not be equally successful in the development of products based on that technology. In order to be successful NPD programmes need to aim at new, or markedly advantageous, products as experienced by the user (Cooper and Kleinschmidt, 1993). Product innovation in this sense is not guaranteed by the solution of difficult technological problems. The mindset and company culture adapted to technological innovation may need to adjust its focus in order to give due weight to the issues involved in product innovation. The second conclusion is that intermediary companies may have an important part to play in overcoming this problem. The intermediary company described in this paper had built up an experience-based expertise in the adaptation of a base technology to the needs of consumer product manufacturers. Because of the company’s reputation in this respect, its product concepts originated with the manufacturers of consumer items, thus ensuring that the product innovation aspect had been given due weight. Thus a network in which the technology producer connects to consumer product manufacturers through an intermediary company specializing in applications of the technology may be more effective than a direct connection. That this differential success was not confined to the particular NPD programmes reported in this paper was indicated by the fact that the intermediary company was in a much stronger financial position than the technology producer. In a tacit endorsement of this analysis, the company which produced the base technology had considered the option of abandoning their attempts to develop consumer products directly in favour of supplying specialist chemicals to intermediary application specialists. In the particular field of specialist chemicals, however, there were too few companies with the appropriate capabilities. At the time of writing, the producer company is having some success with an alternative strategy in which the intermediary company is replaced by the research laboratories of large customer companies. Because of its reputation in the science of specialist chemicals, the company is a leading supplier of specialist chemicals to these laboratories, and there are cases in which they have developed successful new products based on these chemicals. References Afuah, A.N., Bahram, N., 1995. The hypercube of innovation. Research Policy 24, 51–76.

Bidault, F., Fischer, W.A., 1994. Technology transactions—networks over markets. R&D Management 24 (4), 373–386. Bower, D.J., 1993. New product development in the pharmaceutical industry: pooling network resources. Journal of Product Innovation Management 10 (5), 367–375. Cooper, R.G., 1988. Predevelopment activities determine new product success. Industrial Marketing Management 17, 237–247. Cooper, R.G., Kleinschmidt, E.J., 1993. Screening new products for potential winners. Long Range Planning 26, 74–81. Evans, W., 1990. The Japanese corporate approach. In: Oakley, M. (Ed.) Design Management: A Handbook of Issue and Methods. Blackwell, Oxford, pp. 393–405. Gemunden, H.G., Heydebreck, P., Herden, R., 1992. Technological interweavement—a means of achieving innovation success. R&D Management 22 (4), 359–376. Hoshino, K., 1987. Semiotic marketing and product conceptualisation. In: Umiker-Sebeok, J. (Ed.) Marketing and Semiotics; New Directions in the Study of Signs for Sale. Mouton de Gruyter, Berlin. Hutcheson, P., Pearson, A.W., Ball, D.F., 1995. Innovation in process plant: a case study of ethylene. Journal of Product Innovation Management 12 (5), 415–430. Hutcheson, P., Pearson, A.W., Ball, D.F., 1996. Sources of technical innovation in the network of companies providing chemical process plant and equipment. Research Policy 25, 25–41. Iansiti, M., 1995. Technology integration: managing technological evolution in a complex environment. Research Policy 24, 521–542. Karakaya, F., Kobu, B., 1994. New product development process—an investigation of success and failure in high-technology and nonhigh-technology firms. Journal of Business Venturing 9 (1), 49–66. Lipparini, A., Sobrero, M., 1994. The glue and the pieces— entrepreneurship and innovation in small-firm networks. Journal of Business Venturing 9 (2), 125–140. Moenaert, R.K., Caeldries, F., 1996. Architectural redesign, interpersonal communication and learning in R&D. Journal of Product Innovation Management 13, 296–310. Noren, L., Norrgren, F., Trygg, L., 1995. Product development in interorganizational networks. International Journal of Technology Management, 105–118. Shaw, B., 1993. Formal and informal networks in the UK medical equipment industry. Technovation 13 (6), 349–365. Tidd, J., 1995. Development of novel products through intraorganizational and interorganizational networks—the case of home automation. Journal of Product Innovation Management 12, 307–322. Yin, R., 1994. Case Study Research: Design and Methods, 2nd ed. Sage Publications, Beverly Hills. Anne Tomes is a Visiting Research Fellow at Sheffield Hallam University. She received her BSc from Birmingham University in 1968, her MSc in Marketing from UMIST in 1971 and her PhD from Sheffield University in 1982. Before her move to academia, she gained industrial experience in new product development with Ferrodo Ltd and Granada TV rental, and worked for 12 years as a marketing consultant with many top organizations, including Whitbread, Batchelors, Texas Instruments, British Rail, Johnson & Johnson, and British Glass. Dr Tomes has researched and published widely within the marketing area and has written two student management textbooks. Her theoretical and practical interests currently focus on design and the development of new science-based products, in which fields she has recently completed two large-scale research projects sponsored by the Design Council. Rosie Erol is a Research Associate in the Design and Innovation Research Unit at Sheffield Hallam University. She received her BSc in Physics at Sheffield University in 1991, her MSc in Instrumentation and Analytical Science at UMIST in 1992 and her PhD in Medical Physics at Sheffield in 1996. She has recently completed work in a research project sponsored by the Design Council on the role of the design imagination in connecting basic research to product applications. Peter Armstrong is a Professor of Management at Keele University. He

A. Tomes et al. / Technovation 20 (2000) 115–127 received his BSc in Engineering from Bristol University in 1960, his MSc in Sociology from Bath University in 1973 and his PhD in Sociology from Bristol in 1988. After seven years as an R&D engineer, he returned to university to study sociology and is now best known for his research on the social aspects of accounting. He has also published widely on indus-

127

trial relations and other aspects of management. Professor Armstrong’s current research is concerned with the role of management accounting in controlling labour and with the cultural interface between management and design.