International Journal of Innovation Management
Vol. 4, No. 4 (December 2000) pp. 491–528
© Imperial College Press
KNOWLEDGE-INTENSIVE BUSINESS SERVICES AS
CO-PRODUCERS OF INNOVATION
PIM DEN HERTOG
Dialogic Innovatie & Interactie,
Wilhelminapark 20,
3581 ND, Utrecht,
The Netherlands
e-mail: denhertog@dialogic.nl
Received 20 December 1999
Revised 3 February 2000
Accepted 3 February 2000
In the unfolding knowledge-based economy, services do matter. But while they are increasingly
seen to play a pivotal role in innovation processes, there has been little systematic analysis
of this role. This essay presents a four-dimensional model of (services) innovation, that
points to the significance of such non-technological factors in innovation as new service
concepts, client interfaces and service delivery system. The various roles of service firms in
innovation processes are mapped out by identifying five basic service innovation patterns.
This framework is used to make an analysis of the role played by knowledge-intensive business
services (KIBS) in innovation. KIBS are seen to function as facilitator, carrier or source of
innovation, and through their almost symbiotic relationship with client firms, some KIBS
function as co-producers of innovation. It is further argued that, in addition to discrete and
tangible forms of knowledge exchange, process-oriented and intangible forms of knowledge
flows are crucial in such relationships. KIBS are hypothesised to be gradually developing
into a “second knowledge infrastructure” in addition to the formal (public) “first knowledge
infrastructure”, though there is likelihood of cross-national variations in the spill-over effects
from services innovation in and through KIBS, and in the degree to which KIBS are integrated
with other economic activities. Finally, some implications for innovation management and
innovation policy are discussed.
Keywords: service innovation, innovation patterns, business services, knowledge
infrastructure, innovation systems
491

492 P. den Hertog
Introduction
This paper aims to contribute to a better appreciation and understanding of the
role of services in innovation. Following a brief presentation of existing approaches
to service innovation, the third section outlines a four-dimensional model that
allows for a better understanding of service innovation. This allows us to identify
five basic service innovation patterns. Following this, an analysis is made of
the role played by a sub-category of services that play a considerable role in
innovation, especially that involving their client firms, namely, knowledge-intensive
business services (KIBS). It is argued that KIBS function either as a facilitator,
carrier or source of innovation. Through their almost symbiotic relationship with
client firms, some KIBS function as co-producers of innovation. It is further
argued that in addition to discrete and tangible forms of knowledge exchange,
process-oriented and intangible forms of knowledge flows are crucial in this
relationship. A three-tiered model is postulated in which KIBS are gradually
developing into a “second knowledge infrastructure” in addition to the formal
(public) “first knowledge infrastructure”. This model serves as a hypothesis that
needs to be tested and refined, especially since as important differences between
individual countries can be noted as to the degree to which KIBS actually function
in this way. Finally, the paper considers some wider implications of the analysis
for innovation management and policy.
While the analysis is presented in largely conceptual terms, it is rooted in
empirical research1 conducted throughout the 1990s on services, services innovation
and the role played by KIBS in innovation processes of client firms. The various
examples used here are taken from this research base.
Existing Approaches to Service Innovation
One helpful answer to the question “What does it mean to produce a service?”
is provided by Jean Gadrey and collaborators (1995):
“to produce a service […] is to organise a solution to a problem (a
treatment, an operation) which does not principally involve supplying a
good. It is to place a bundle of capabilities and competences (human,
technological, organisational) at the disposal of a client and to organise
a solution, which may be given to varying degrees of precision”.
1See,
for example, Bilderbeek and den Hertog (1992; 1996; 1997; 1998); Bilderbeek et al. (1994);
den Hertog and Bilderbeek (1997; 1999); den Hertog et al. (1997; 1998); Hofman et al. (1998);
Miles et al. (1995); Vlaar et al. (1997).

Knowledge-Intensive Business Services as Co-Producers of Innovation 493
This definition makes it clear that apart from technological capabilities, human
and organisational capabilities are also important for providing services.
Additionally, it allows for a differentiation between highly standardised service
products or service formulas with quasi-good characteristics (e.g. fast food chains),
and more customised services, often based on more tacit forms of knowledge.
Moreover, the latter services often emerge as a result of co-production between
the actual service provider and its client, as many consulting and advisory services
show.
The analysis of services and service innovation has progressed quite remarkably
over the last decade. Although the attention given to services by researchers
grew from the 1970s on, they were long discounted in terms of technological
innovation. In 1984, Pavitt — introducing his sectoral taxonomy of technological
change — labelled the service industries as mainly supplier-dominated sectors.2
Similarly, the important theoretical contributions of Barras (1986; 1990) portray
most service sectors as initially supplier-dominated, and as receiving an impetus
from manufacturing in order to be able to embark on subsequent phases of
innovation processes.
As the field of service innovation studies has expanded, two results of significance
have emerged here. First, it has been recognised that although many services
deliver a — sometimes substantial — contribution to innovation processes, they
are not merely passive recipients of others’ innovations. Second, the emphasis
on technological innovation has been somewhat moderated by the recognition
of the importance of non-technological elements of, and approaches to, service
innovation. This resulted in a better understanding of the “peculiarities” of services
(Miles, 1993), service management (Norman, 1991; Quinn, 1992), the significance
of interaction with clients (and of clients’ competences; cf. also Kline & Rosenberg,
1986), the importance of recombination of existing elements in new services
(Henderson & Clark, 1990) and other such points. More recently, Gallouj
et al. (1997) outlined six innovation models that could be used for describing
service innovation. They distinguish between radical innovation, improvement
innovation, incremental innovation, ad hoc innovation, recombinative innovation
and formalisation innovation.
A comprehensive model for understanding innovation in services, and better
still understanding innovation in services and other sectors in the same framework,
is not yet apparent in this literature. Such a model is increasingly needed as
2Soete and Miozzo (1989) provided a more differentiated picture of the service industries distinguishing
between supplier-dominated, scale-intensive physical networks and information networks, and
specialised/science-based services. Although this may be a helpful extension of the Pavitt taxonomy,
it still remains a mostly technical (and sectoral) taxonomy.

494 P. den Hertog
service functions can be seen to be prominent all over the economy, and thus,
as service innovations are relevant — although to different degrees — in all
industries. Service innovation is equally relevant for manufacturing firms, which
increasingly use innovation in service functions and features to differentiate their
products. In a similar vein, some business processes in service firms resemble
those in manufacturing, for example, administrative processing in back offices.
A continuum rather than a strict distinction between manufacturing firms and
service firms — and the innovation models used for them — seems appropriate
when discussing firm innovation.
The available models for thinking about service innovation do help identify
those dimensions which appear to be most relevant here. Accordingly, the next
section sets out an (admittedly eclectic) conceptual model as a heuristic tool
that will enable us to discuss service innovation in general terms.
Conceptualising Service Innovation
Service innovation is rarely limited to a change in the characteristics of the
service product itself. Innovation often coincides with new patterns of product
distribution, client interaction, quality control and assurance, etc. But there are
huge differences in the specific patterns involved: what is important for introducing
one new product onto the market might be totally irrelevant for others. Offering
a completely new service may differ considerably from offering an existing service
using a new distribution channel. In practice, most innovations appear to be
a mixture of major and minor changes and adaptations of existing (service)
products.
In order to discuss, map and analyse the diversity of innovations in greater
detail and in a structured way, a four-dimensional model of service innovation
is introduced (Fig. 1). Although conceptual, it is concrete enough to map service
innovation and discuss the practical development of new services or service
innovation policies, and has proved useful in this respect in discussions with
policy-makers and service entrepreneurs (see, e.g. Bilderbeek et al., 1998a; 1998b).
The four dimensions are presented first, followed by the linkages between them.3
Dimension 1: The service concept
Manufactured products (and processes) are typically highly tangible and visible,
unlike most services. Admittedly, some service innovations are highly visible,
3Although most examples provided below are taken from service industries, service innovation is
equally relevant for manufacturing industries.

Knowledge-Intensive Business Services as Co-Producers of Innovation 495
characteristics of actual
characteristics of existing
marketing & distribution
and competing services
and potential clients
NEW SERVICE capabilities NEW CLIENT
CONCEPT INTERFACE
(DIMENSION 1) (DIMENSION 2)
TECH-
NOLOGICAL
OPTIONS
org apab
ies
cap HRM
(DIMENSION 4)
ilit
c
an iliti
ab
isa es
tio
na
l
NEW SERVICE
DELIVERY SYSTEM
(DIMENSION 3)
capabilities, skills & attitude of existing © Dialogic
and competing service workers
Fig. 1. A four-dimensional model of service innovation.
especially where delivery of the product is involved (ATMs, etc.). Many others
involve more intangible characteristics, such as new ideas of how to organise
a solution to a problem. Although a particular service concept may already be
familiar in other markets, the key thing is that it is novel in its application
within a particular market. As usual, in innovation research, there are thorny
problems concerning when a product, function or concept is really new. Judgements
can vary according to whether and when it is new to the providing firm, new
to the client, to the regional, national or global market and whether it involves
new logic or scientific knowledge.
Some examples of “conceptual innovations” are:
• call centre services — these install, organise and recruit staff for their client’s
call centre — which have emerged from temporary staffing offices on the
basis of their initial involvement with providing temporary labour for call
services;
• IT consultants who offer their client firms semi-standardised and incremental
plans for implementing e-commerce;
• Benetton’s development of a particular style of shopping outlets to give the
brand name its own character, to create a specific shopping environment that
is recognisable for their clients.

496 P. den Hertog
Although not all service innovations have a strong novel conceptual element,
conceptual innovations are much more likely to be found in service firms (or
better service functions) than in pure manufacturing firms. Such innovations
are usually highly intangible — meaning that while in some cases, the service
itself may have quite tangible elements, the new features have less to do with
material artefacts.
Dimension 2: The client interface
A second element of service innovations is the design of the interface between
the service provider and its clients. These interfaces are the focus of a good
deal of service innovations, though innovation studies, with their focus on mass
manufacturing, have tended to overlook the changes occurring in these interfaces.
As a quite general phenomenon across a wide span of services, product offerings
are increasingly marketed and even produced in a client-specific way (even with
client-specific pricing) and delivered electronically as far as they have informational
components.
In business services in particular, clients are often also part and parcel of
the production of the service product. The way the service provider interacts
with the client can itself be a source of innovation. Increasingly, there is no
clearly identifiable point where the producer’s activity stops and the user’s activity
begins. This is of course particularly true where the business service itself is
offering support for innovation, e.g. in R&D and design services. With the high
degree of co-design and co-production of service products, it may be difficult
to locate the innovation within service supplier or client: it is not unusual, for
instance, for service firms to site their staff within client organisations for periods
of time.
Examples of “client-interface innovations” include:
• the large-scale introduction of account management systems in professional
organisations, such as economic consulting or IT firms, can in some cases
be interpreted as a renewal of the client interface;
• electronic data interchange (EDI), which represents an effort to establish
common formats for electronic documents that allow for a wide range of
interactions to be partially automated — including various elements of design
as well as ordering and invoicing. Organisational challenges have made the
take-up of EDI slower than anticipated but a substantial industrial use has
developed.

Knowledge-Intensive Business Services as Co-Producers of Innovation 497
Dimension 3: The service delivery system/organisation
This dimension, involving service delivery system and organisation, is often
directly related to the preceding dimension — the linkage between the service
provider and its client (the client interface). Delivery is indeed one specific
type of interaction across the client interface (others include financial transactions,
design inputs, after-sales, and so on). However, dimension 3 is different. It refers
to the internal organisational arrangements that have to be managed to allow
service workers to perform their job properly, and to develop and offer innovative
services. It is closely related to the question of how to empower employees,
to facilitate them so that they can perform their jobs and deliver service products
adequately. On the one hand, new services may require new organisational forms,
(inter)personal capabilities and skills. On the other hand, an organisation can
be designed and employees can be trained so as to leave room for innovations
and non-conventional solutions to practical problems.
Examples of “delivery system and organisation innovations” include:
• the large-scale introduction of home shopping services — or consumer
e-commerce — may cause a substantial change in the ways in which service
provider and client relate;
• introducing e-commerce in business processes may require serious business
process reengineering. E-commerce may not only have a substantial impact
on the way in which the actual commercial transactions occur, but also the
processes preceding and following the transaction;
• in more traditional shopping environments, the lengthening of retailer opening
hours may have serious consequences for the type of customers it attracts,
the type of products offered, the immediate availability of sales and after-
sales service of different types, etc.
Dimension 4: Technological options
The fourth dimension is the centre of much analysis and debate, especially
concerning the degree to which service firms themselves in practice are giving
shape to technology development. Clearly, service innovation is possible without
technological innovation; technology is not always a dimension. Nonetheless,
in practice, there is a wide range of relationships between “technology” and
“service innovation”. These vary from technology mainly playing a role as a
facilitating or enabling factor, to something much closer to supply-push, technology-
driven innovation.
Service firms also differ in their awareness of relevant available technological
options, the degree to which they dispose of the necessary technology themselves

498 P. den Hertog
or have access to the necessary knowledge and the degree to which they consequently
can act as demanding customers and articulate their technological needs. Many
innovations are driven by downstream service sectors and can surely be considered
user-dominated. In fact, users may play a crucial role in developing and implementing
new services, although some of the required technologies may come from suppliers.
Although IT is certainly not the only relevant technology in service innovation,
IT is particularly pervasive. The numerous information processing tasks to which
it may be applied include many that are intrinsic to almost all economic activities.
IT is thus often perceived as the great enabler of service innovation. Many
commentators who recognise the profound implications of IT for services still,
however, consider this technology as typically supplier-dominated. It is true that
many smaller and less innovative service firms are relatively less proactive when
it comes to incorporating new IT, though even here there is rarely the purely
passive process of absorption implied by the term “diffusion”. However, in many
larger and/or more advanced firms there is an extremely active process of
technological development going on.
Examples of “technological innovations” include:
• large retail stores increasingly resemble financial services in their IT use. For
instance, the UK supermarket Tesco has set up an Internet service provider,
and many supermarkets are offering banking and insurance services;
• tracking & tracing systems enable transport service providers to monitor the
progress of their fleet and thus to manage their transport services more closely.
These examples of IT utilisation illustrate that service firms are not necessarily
supplier-dominated. This is especially true in the case of IT services themselves,
like software houses. To a certain degree, software firms have to adapt their
activities to new products from hardware companies, e.g. new generations of
chips. This involves near continual updating and — typically — expansion of
software to exploit the facilities of new equipment. But the process of developing
new applications, new functionality, new interfaces, etc., is much more in the
software firms’ hands. It is also evident that sectors with a long experience
of IT investment are major sources of innovation — in the shape of new
configurations of hardware, new software and applications, new interfaces, etc.
A good example is the financial sector, which is a huge employer of software
and networking staff.
Linking the four dimensions
Any service innovation involves some combination of the above-mentioned
dimensions of service innovation. A complete new service will usually mean

Knowledge-Intensive Business Services as Co-Producers of Innovation 499
that a new service delivery system will have to be developed, employees will
have to change the way they work or relate to customers (the client interface),
the way IT is used in business processes, while a new service concept may
also be involved. In addition to the meaning of these four dimensions as separate
as discrete vectors of change, the linkages between these dimensions may be
of even more significance.
Often these cross-linkages are forged in practice by those responsible for
marketing, organisation development and distribution. For instance, launching
a new service concept (for existing or new clients) requires marketing expertise.
Similarly, creating an adequate interface with clients and adapting the service
delivery system require knowledge of how services are distributed (both in terms
of where they are produced and of how they are delivered). The decision as
to whether to develop new services requires organisational knowledge: can the
current organisation deliver the new service? What organisational changes might
be needed?
A particular service innovation may display one dominant feature related
to one of the four dimensions; quite likely, this particular feature will prompt
a set of changes in other dimensions, in order to bring about a successful innovation.
Consider an example from the retail service sector. Intelligent cash registers
and advanced data warehousing, as widely used by large companies, are basically
technological innovations (i.e. dimension 4). They allow for the creation of detailed
client profiles and personalised product offerings. But these applications cannot
be bought from the shelf and be readily implemented. They need to be combined
with the specific shop formula employed (dimension 1), the way the retailer
wants to communicate with its clients (dimension 2) and to train its employees
(dimension 3), etc.
In practice, it may be the combination of the four dimensions that ultimately
characterises each particular service innovation. The weight of the individual
dimensions and the importance of the various linkages between them vary across
individual services, innovations and firms. Similarly, the inputs required to link
the dimensions in practice differ according to the type of service, and the extent
to which the search and selection process (inherent in all innovation processes)
is formalised.
Mapping Service Innovation Patterns
Just as service innovations are extremely varied, so is the role of service firms
in innovation processes diverse. The dominant view of innovation in services
portrays the process as supplier-dominated, with service firms being dependent
on their suppliers for innovative inputs which are to different degrees transformed

500 P. den Hertog
into more or less innovative service offerings. However, our studies confirm
that the contribution of service firms to innovation is broader than this, especially
if we include the non-technological aspects of innovation. Some services even
play a role in the innovation processes of their clients, e.g. R&D, design &
engineering services and some IT services. A typology4 of five innovation patterns,
in each of which service firms play a different role, is presented below, and
displayed in Fig. 2. Each of the patterns displays a different mix of linkages
between three types of actors:
• suppliers of inputs (equipment, capital, human resources, etc.);
• the (innovating) service firms, and;
• the clients of the innovative service product (another service or manufacturing
firm in the case of intermediate products, or final users).
Going from pattern 1 to 4, the influence of the client firm or final consumer
on the innovation process gradually increases. Pattern 5 represents a somewhat
different situation as all actors in the value system contribute to a particular
innovation or are forced to accommodate it.
Supplier-dominated innovation (pattern 1)
Service innovations have traditionally been depicted as following this pattern:
innovations (as a rule technological innovations) are derived from hardware
industries in particular. These innovations from external suppliers are disseminated
and implemented by service industry users, who in their turn satisfy the needs
of their clients. Examples of this pattern include:
• microwave ovens in catering, whose introduction has greatly extended the
possibilities for food preparation (and reheating) in cafes and restaurants;
• cash registers and mobile phones that have been assimilated into many small
firms which otherwise use little new technology.
There are many similar examples with a clear “technology push”. Typical
for this pattern is, at least initially, little scope for user industries to influence
the actual product supplied by the supplier. The adopting firm often has to bring
about some organisational changes in order to be able to use the innovation
— to adapt its organisation, train its employees, etc. — and to offer more efficient
and higher quality services as a result.
4This is to be considered a mapping device. Quite possibly, more patterns can be found.

Knowledge-Intensive Business Services as Co-Producers of Innovation 501
ROLE OF:
INNOVATION SUPPLIER SERVICE FIRM CLIENT FIRM EXAMPLE
PATTERN (service or
manufacturing)
Introduction of
Supplier- interactive TV
Dominated equipment; IT
Innovation goods; medical
robots,
tomography
Introduction of
Innovation in new shop formula,
Services new pension and
saving schemes
Green banking
Client-Led services, door-to-
Innovation door transport
services
Engineering
Innovation services helping
through Services oil & gas firms in
designing new oil
rigs, etc.
Multifunctional
Paradigmatic chip cards, sub-soil
Innovation transport services
Legend
Inputs for Supplier- Implementation
Service Product Domination of new Service
(push) Product/
Organisation
Locus of User User of
Innovation Domination Innovative
Activity (Pull) Service Product
Fig. 2. Patterns of services innovation.

502 P. den Hertog
Innovation within services (pattern 2)
In this pattern, the actual innovation and implementation take place in the service
firm itself. Such innovations may be technological, non-technological, or (as
in many cases) a combination of the two. Typical examples of this pattern involve
a new product, product bundle or delivery system, that is thought up in the
service firm itself (e.g. by a new business team), and implemented throughout
the organisation, possibly with “innovation support” from outside.
Client-led innovation (pattern 3)
In this case, the service firm is responding to needs clearly articulated by its
clients. While, in a sense, every successful innovation is a reaction to a perceived
market need, for some service innovations, this is more clear-cut than for others.
Some examples:
• door-to-door public transport services aimed at the business traveller, a clear
answer to the often-heard complaint “we would like to use public transport
(the train) more often, but that pre- and post-train transport is too time consuming”;
• green banking services, to appeal to a growing number of individuals who
want to invest their (saved) money in a “socially responsible” way.
In these cases, the demands are expressed by segments of mass markets.
In many other cases, the influence may come from a single client, which is
often the case in business services: for instance, a client may propose that a
training firm back up its face-to-face sessions with computer-based aids.
Innovation through services (pattern 4)
In this more complicated pattern, service firms influence the innovation process
taking place within the client firm. The provider of intermediate services may
provide knowledge resources that support the innovation process in various ways,
such as:
• providing an expert project manager with the necessary skills to implement
an innovation;
• providing an innovative tailor-made software package;
• providing training or written advice regarding product selection and
implementation;
• providing advice on how to conduct the innovation process, or providing support
tools to foster creativity among teams in the client organisation.

Knowledge-Intensive Business Services as Co-Producers of Innovation 503
Different modes of support coexist in providing such business services
(— often KIBS). An engineering consulting firm supported an oil and gas company
that wanted to drill and explore in a “protected” area, helping them to find
new operational methods to meet the strict environment protection rules by reviewing
existing practice, proposing new operations, designing new methods, training
the operators of the client firm, actually heading the drilling operations, etc.
Despite these inputs, much if not all of the actual innovation takes place at
the client site, and with its personnel (cf. Hoffman et al., 1998). In this case,
the engineering firm facilitated the innovation process at the client firm. Across
different circumstances, the precise role (source, carrier, facilitator) of the
intermediate service provider, and its degree of interaction with the client firm,
may vary substantially (Bilderbeek & den Hertog, 1997).
Paradigmatic innovations (pattern 5)
This pattern involves complex and pervasive innovations affecting all actors
in a value chain profoundly: when driven by fundamentally new technologies,
they are labelled technological revolutions or new technology systems (Freeman
& Perez, 1988). But they may also be driven by regulations, resource constraints,
and other dramatic changes that require innovation to take place across many
elements of the value chain, implying completely new infrastructures, new types
of knowledge and adaptation on the part of intermediate and final users. For
example:
• if in a very densely populated area, the regular transport of goods is no longer
possible and the decision to switch to underground transport was taken, parties
across the value chain would have to innovate and change practices. Manufacturers
of transport equipment would have to provide completely new transport
equipment; transport companies would have to change their service offerings,
retrain their personnel, market their product in different ways; users would
have to change their behaviour and use of transport facilities;
• similarly, the switch from a few public TV channels towards multi-channel
pay-per-view regimes require innovations and change of behaviour on many
fronts;
• the large-scale introduction of multi-functional “smart cards” would be another
example of a paradigmatic innovation.
The typology demonstrates the wide variety of roles of service firms in innovation
processes. Since the way service providers and client firms interact is central
to the process of service innovation, both factors have to be taken into account.
Further innovation patterns may be identified by, for example, taking additional

504 P. den Hertog
variables into account. One such factor might be the role of government as
a trigger for innovation, a role that can be quite important — as when innovation
is promoted by R&D funding and/or procurement decisions, or through new
regulations (e.g. those fostering environmental innovation). Another variable may
be the degree to which end-users are given the opportunity (or are forced) to
co-produce particular services: the shift to “self-servicing” is an important aspect
of services innovation. Many variations on these innovation patterns can be
found in practice. Sometimes, for example, innovation takes place in individual
service functions (irrespective of whether it is a service or manufacturing firm),
that might or might not be subsequently outsourced to specialised service firms.
Two such situations suggest additional patterns of service innovation. First,
innovation in a firm’s internalised service function is quite a common pattern.
All firms engage in a range of service activities — and sometimes this involves
innovation. A capital goods manufacturer may have an innovative leasing scheme,
dealer organisation, after-sales service or training sessions. Increasingly,
manufacturing firms realise that the package of services offered around the actual
goods may be crucial to competitiveness. Quite frequently, the added value realised
with these services is much higher than the margins realised on the capital goods.
A second case is innovation in an outsourced service function, such as facilities
management, catering and cleaning, or even more strategic functions (e.g. temporary
staff, management, R&D). In most cases, more specialised service firms take
over these functions.5 In many outsourcing relationships, activities may be precisely
specified and cost competition may be intense; in such cases, innovation is less
likely. But in other cases, there is sufficient level of specialisation and scope
for economies of scale to provide incentives for innovative solutions. For instance,
companies increasingly hire temporary labour, and increasingly, the troublesome
task of managing these temporary workers and the associated paperwork can
be outsourced to a temporary employment agency. In their turn, such agencies
increasingly understand the human resources required by particular client
firms, and may even offer to completely take over the human resources
management function, training and hiring personnel, helping displaced staff to
find new jobs, etc.
The Role of KIBS
Knowledge-intensive business services (KIBS) form a category of service activities
which is often highly innovative in its own right, as well as facilitating innovation
5This pattern resembles the previous one but it goes a step further. In this case, the client firm influences
the innovation taking place in the outsourced service function.

Knowledge-Intensive Business Services as Co-Producers of Innovation 505
in other economic sectors, including both industrial and manufacturing sectors.
In an earlier study (Miles et al., 1995), the innovative role of this category
of services (KIBS) was substantiated, both through literature research and case
studies. KIBS industries were defined as:
• private companies or organisations;
• relying heavily on professional knowledge, i.e. knowledge or expertise related
to a specific (technical) discipline or (technical) functional domain; and,
• supplying intermediate products and services that are knowledge based.
KIBS actually cover a rather wide range of services.6 For instance, many
professional services can be considered knowledge intensive, be they technology
based or not. In other service sectors, specific segments with a relatively high
professional component can be identified as intensive users of knowledge. Table 1
gives an overview of KIBS.
The analysis of the role of KIBS in innovation processes brings into focus
the ways in which knowledge is produced and used in the economy, and the
role of KIBS in these processes. The production of services is often the result
of a joint effort of the service provider and client (be it the client is the ultimate
end-user, some intermediary in the client firm or in another firm altogether).
In this process of co-production, the quality of the resulting service product
largely depends on the nature of the interaction between the service provider
and client, and the quality of the communication process that is involved. One
important role for KIBS is providing a point of fusion between more general
scientific and technological information, dispersed in the economy, and the more
local requirements and problems of their clients. KIBS operate as catalysts which
promote a fusion of generic and quasi-generic knowledge, and the more tacit
knowledge, located within the daily practices of the firms and sectors they service.
One result of this interaction is that feedback from clients can shape innovations
in service firms, just as much as service firms can influence their customers’
innovation. “Interactive learning” and “user-producer linkages” are important
notions here.7 This two-way learning process is prominent where KIBS are
concerned, including the category of wholesale and retail trade of machinery
6In case studies performed since 1994, we have focussed on those KIBS that derive their intermediate
function primarily from the production and transfer of technology-related knowledge, among others,
engineering services and IT service firms. This sub-category of technology-based KIBS (T-KIBS) is
only occasionally considered within the existing bodies of research literature and policy practice on
technological innovation.
7Thus, the literature on user–producer relationships (see, e.g. Lundvall, 1988; 1992) is especially
relevant for the study of service (mediated) innovation.

506 P. den Hertog
Table 1. Knowledge-intensive business services: an overview.
• Accounting and bookkeeping
• Management consultancy (not only concerning new technologies)
• Specific building services (e.g. architecture, surveying, construction engineering, etc.)
• Facility management services
• Technical engineering services
• Research and development services (excluding university-based R&D)
• R&D consultancy services
• Design (not only concerning new technologies)
• Environmental services (e.g. environmental law, elementary waste disposal services,
remediation, environmental monitoring, scientific/laboratory services, etc.)
• Computer and information-technology-related services (including software services)
• Legal services (note that technology-related legal services form a growth area!)
• Marketing & advertising
• Exploitation and trade in real estate
• Training (not only concerning new technologies)
• Specific financial services (e.g. securities and stock-market-related activities)
• Temporary labour recruitment services
• Press and news agencies
Based on: Miles et al., 1995.
and equipment.8 IT support services, management consultancy and technical
engineering, for example, typically work with their clients in highly interactive
ways. Client firms and KIBS providers work together to find solutions to problems
and challenges. Through the interaction, the client’s knowledge base changes
while the KIBS provider also gains more experience, learning more about the
characteristics of a specific industry. The KIBS provider is thus enabled to refine
and differentiate the services offered and methods used, to learn about new
business opportunities, upgrade his/her track record, etc.
As innovation processes involve parties with various gaps in resources and
in innovation management capabilities, intermediaries (including various KIBS)
may be employed directly to fill these gaps or less directly to help bridge them
(Bessant & Rush, 1998). The type of bridging required varies:
• expert consulting, providing particular solutions to particular problems;
• experience sharing, transferring what is learned in one context to another.
(The metaphor here might be that of bees cross-pollinating as they fly from
flower to flower);
8The services “surrounding” the hardware are often as important as the hardware itself for implementing
especially complex systems and machinery.

Knowledge-Intensive Business Services as Co-Producers of Innovation 507
• brokering, putting different sources and users in contact across a wide range
of services and resources. (The metaphor here is that of marriage broker/
dating agency trying to set up the most appropriate match between two partners.
The degree to which the innovation agent is perceived as independent is important
here, as his/her credibility is likely to diminish if there are particular standard
solutions being proposed or particular sources of supply advanced.);
• diagnosis and problem clarification, helping users articulate and define the
particular needs in innovation. Many user firms lack the capability to understand
or prioritise their problems into a strategic framework for action and outside
agencies may be able to assist in this process. (A metaphor here would be
the general medical practitioner, whose main task is diagnosis and who then
prescribes a range of specialist treatments such as medicines, physiotherapy,
diet, surgery, which are then actually delivered by others.);
• benchmarking, where the process of identifying and focussing on “good practice”
can be established through an intermediary;
• change agency, where organisational development can be undertaken with help
from a neutral outside perspective.
Earlier, a differentiation was made — more specifically focussing on KIBS
— between three roles played by KIBS in supporting innovation in client firms,
namely: facilitator, carrier and source of innovation (Bilderbeek & den Hertog,
1997; Miles et al., 1995). These three roles are briefly characterised below.
Facilitator
A KIBS firm is a facilitator of innovations if it supports a client firm in its
innovation process, but the innovation at hand does not originate from this KIBS
firm nor is it transferred (from other firms) by this KIBS firm to the client
firm. Examples include:
• a management consultant helping a client to introduce a new account management
system or developing a new service distribution channel;
• a technical engineering firm seconding a team of its engineers to work with
the technical engineers of the client to co-produce an innovative solution in,
e.g. offshore platform construction or subsoil building.
Carrier
A KIBS firm is a carrier of innovation if it plays a role in transferring existing
innovations from one firm or industry to the client firm or industry even though

508 P. den Hertog
the innovation in question does not originate from this particular KIBS firm.
Examples are:
• an IT firm implementing and customising advanced and innovative ERP software
(SAP, BAAN) in a client firm;
• a management consultant specialising in CAD/CAM applications helping a
major client (a shipyard) to specify the exact user needs and technical
specifications of a new CAD/CAM programme, and subsequently to
implement it.
Source
A KIBS firm is a source of innovation if it plays a major role in initiating
and developing innovations in client firms, usually in close interaction with
the client firm. Relevant examples here include:
• an advertising agency developing and implementing a completely new campaign
for a client;
• a provider of call centre solutions advising and actually implementing a new
call centre for a client.
Knowledge Resource Flows
In addition to discrete and tangible forms of knowledge exchange, process-oriented
and intangible knowledge flows are crucial in their relationship between KIBS
and client firms. This section draws on various contributions to identify
discriminating dimensions which can help us grasp these processes, and relate
these to the four dimensions of service innovation considered in the third section.
The Nonaka and Takeuchi model of knowledge creation
Nonaka and Takeuchi (1995: 56– 94) draw on the well-known distinction, pioneered
by Polanyi (1967), between tacit knowledge, which essentially represents “know-
how” (the subjective knowledge), and explicit knowledge, “knowing about”
(the objective knowledge). Explicit knowledge (available in the form of formulas,
technical specifications, or embedded in equipment, computer programmes, and
so on) is relatively easy to transfer and store.9 On the contrary, tacit knowledge
9Hales (1997: 4) criticises in this context “addressing knowledge as some kind of ‘stuff’ which can be
held and deployed in various ways”. Strictly, it is information that is being stored and communicated,
and this information is only transformed into knowledge by an active cognitive process on the part of
the user.

Knowledge-Intensive Business Services as Co-Producers of Innovation 509
(made up of ideas, experience, facts, etc.) is highly personal and difficult to
convert; it cannot easily be codified, and can often only be observed through
application and acquired through practice and experience.
Nonaka distinguishes individual knowledge, possessed by single operators,
from organisational knowledge. The latter includes both knowledge possessed
by individuals, and the procedures, routines, strategies, etc., which are developed
at a corporate level and shared by the members of the organisation. The process
of knowledge generation is directly associated with the interaction between
individual and organisational knowledge: “human knowledge is created and
expanded through social interaction between tacit knowledge and explicit
knowledge” (1995: 61). Four modes of knowledge conversion are identified,
as depicted below in Fig. 3 (1995: 62– 69):
(i) Socialisation — the development of tacit knowledge from tacit knowledge.
This is a process of sharing experiences and thereby creating tacit knowledge
such as shared mental models and technical skills (e.g. to teach somebody
how to use a machine). Often, an individual can develop tacit knowledge
through observation, imitation and practice, without much use of instructional
language.
(ii) Externalisation — development of explicit knowledge from tacit knowledge.
This involves the rationalisation of tacit knowledge and its articulation into
explicit concepts and formal models (e.g. to write an instruction manual).
Externalisation is the core of the knowledge creating process, the
“quintessential” knowledge-creation process, in which tacit knowledge becomes
explicit, taking the shapes of metaphors, analogies, concepts, hypotheses
or models.
(iii) Combination — development of explicit knowledge from explicit knowledge.
This involves the systematisation and conversion of one system of
formalised concepts into another one. It may involve combining elements
from different knowledge systems together into a new framework, the
“reconfiguration of existing information through sorting, adding, combining,
and categorising of explicit knowledge”. (Associated skills will be those
which involve the ability to obtain a new formula, procedure or software
from existing ones.)
(iv) Internalisation — the move from explicit to tacit knowledge. This occurs
when explicit knowledge is converted into specific know-how (e.g. discovering
how to use a piece of equipment for specific or new needs). Internalisation
is related to “learning by doing”: it is a process of embodying explicit
knowledge into tacit knowledge and the individuals’ tacit knowledge bases
in the form of, for example, shared mental models or technical know-how.

510 P. den Hertog
TO:
Tacit Knowledge Explicit Knowledge
Socialisation: Externalisation:
Tacit share experience, Articulate experience
Knowledge discuss ideas, In formal models;
opinions Embed experience into equipment,
software, etc.
FROM:
Internalisation: Combination:
convert models and formulas into reformulate
Explicit tacit skills; formal models and data, convert
Knowledge learn/teach how codes, etc.
to use equipment
Fig. 3. Four modes of knowledge conversion and the “knowledge spiral” as conceptualised by Nonaka
and Takeuchi (1995: 62/71).
A dynamic element is added to this taxonomy so as to view organisational
knowledge creation as a continuous process, shaped by shifts between the various
modes of knowledge conversion. Nonaka and Takeuchi (1995: 70–71) see this
process as involving:
• first, a socialisation mode, usually starting with construction of a “field” of
interaction, within which the sharing of members’ experiences and mental
models is facilitated;
• second, an externalisation mode, triggered by meaningful “dialogue or collective
reflection”. The use of appropriate metaphor or analogy helps team members
to articulate hidden tacit knowledge that is otherwise hard to communicate;
• third, a combination mode is triggered by “networking” newly created knowledge
and existing knowledge from other sections of the organisation, thereby
crystallising them into a new product, service or managerial system;
• finally, “learning by doing” triggers internalisation.
The tacit knowledge of individuals is seen as the basis of organisational
knowledge. The process of organisational knowledge creation requires an expansion
of the space of interaction through which knowledge is fused and created, so

Knowledge-Intensive Business Services as Co-Producers of Innovation 511
that it additionally takes place at the group, organisational and inter-organisational
level. Thus, a spiral of knowledge creation is established, involving the four
modes of converting tacit and explicit knowledge.
The Nonaka and Takeuchi model of organisational knowledge creation provides
several insights relevant to the KIBS–client interface:
• First, it emphasises the combination of these different forms of knowledge.
Various case studies of the relationships of KIBS to their clients (Bilderbeek
et al., 1994; Miles et al., 1995) indicate that the more tacit forms of knowledge
flows are at least as important as the explicit, codified forms of knowledge
exchange.
• Second, it draws attention to interaction between individuals, team members
and employees from various organisations in creating knowledge new to the
firm. Studies of the functioning of, for example, engineering services (Hofman
et al., 1997; Vlaar et al., 1997) provide many examples of the knowledge
base of the client firm being constantly enriched by confrontation with the
knowledge base of the KIBS provider. This mainly involves personal interactions
between professionals (and the client firm must possess professionals who
can maintain and enrich this dialogue).
• Third, it draws attention to the dynamic nature of the knowledge-conversion
processes. The constant mixing, redefinition, linking, exchanging, reshaping
and enriching of various forms of tacit and explicit knowledge in the course
of interaction10 is what typically happens where KIBS and their clients interact.
KIBS can trigger and strengthen processes of knowledge conversion in clients
(and vice versa). Often, when a client hires a KIBS, new project teams are
set up, employees are forced to interact, to make tacit knowledge explicit,
to think about new combinations of knowledge and their mental models are
challenged. KIBS, in other words, contribute to firm-level learning processes.
They can provide new knowledge certainly, but they may also act as catalysts,
which help internal communication and knowledge conversion.
10KIBS — especially management consultancies — are often (and sometimes justly) criticised for
reselling the client’s own knowledge again to the client. However, the process of conversion and
reconversion of knowledge may not be entirely fruitless! The sheer act of interacting with the KIBS
can help processes such as socialisation, externalisation and combination. The interaction between
clients and KIBS can establish a field in which knowledge resources are exchanged and dialogue
established between various functions/experts (e.g. creating multidisciplinary project teams). This
can allow for combining existing pieces of knowledge already present inside and outside the company.
And the tacit tricks of the trade can be interchanged just through the process of performing tasks
together (e.g. when in-house software developers are working together in project teams with external
IT consultants).

512 P. den Hertog
A multitude of knowledge flows
In practice, knowledge flows between KIBS and their clients are manifold. Some
examples are given in Table 2. Sometimes these resources — solutions to a
(perceived) problem — are (in part) very concrete and tangible. This is the
case, for example, when the service product delivered is a software programme,
written report, drawing or design, advertisement campaign, temporary expert,
project plan, benchmark, or advice on a new organisation structure. However,
the outcomes of the interaction between the service provider and client firm
may be much more complex and hard to pinpoint. More fuzzy outcomes or
process-oriented forms of knowledge exchange can be important by-products
of the more concrete resources just mentioned, since explicit knowledge is often
accompanied by tacit knowledge. In other cases, knowledge resources are developed
and interchanged in the course of the cooperation between the KIBS provider
and client firm. Examples of the more fuzzy results that this can achieve are:
improvement of a management team’s internal communications (they have been
able to build knowledge of shared language, metaphors, visions, objectives),
better understanding of potential markets, know-how for applying equipment/
systems, improved negotiating capabilities for discussing plans and actions with
partners, creation of R&D collaborations, building increased support (inside or
outside the firm) for a solution to a problem, improved reputation, new personal
contacts, introduction to expert network, knowledge institutions or policy-makers.
Many services cannot be packaged in the form of a written report or a piece
of software, but even when they can, it is common to find that implementation
requires various forms of more direct interaction. The content and quality of
the service provided by a KIBS are substantially defined by the quality of this
interactive process and the degree to which service professionals in client and
service provider relate to each other.
Four discriminating dimensions can be used to provide a more structured
view of the ways in which KIBS providers and client firms interact, and the
various kinds of knowledge resource flows involved, namely:
(i) discrete/tangible versus form or process oriented/intangible knowledge;
(ii) human embodied versus non-human (capital, written information) embodied
forms of knowledge resources;
(iii) explicit/codified knowledge versus tacit/non-codified knowledge;
(iv) contractual versus non-contractual forms of knowledge.
These dimensions of knowledge flows can be related to the four dimensions
of service innovations identified earlier.

Knowledge-Intensive Business Services as Co-Producers of Innovation 513
Table 2. Knowledge flows between service provider and client firms.
More or Less Discrete/Tangible More or Less Process-Oriented/Tangible
• Training • Routine problem-solving as part of everyday
• Benchmark project work
• Project management • Improved capability to collaborate in project
• Software package teams
• Advertisement campaign • Instruction when installing new machinery
• Written report (e.g. feasibility study) • Articulation/specification of needs (e.g. for
• Project plan procurement)
• Drawing/design • Sparring partner (testing of ideas client firm)
• Advice (e.g. on new organisational structure) • Introduction to new networks of
• Computation professionals/user groups
• Diagnosis • Information on performance of competitors
• Product documentation (manuals etc.) • Market information
• Secondment of a temporary expert • Coupling to new partners
• Use of an R&D facility • Knowledge on how to create support for
innovations
• Insight on how to access research and
technology organisations, higher education
institutes and government
Dimension 1: Discrete/tangible versus process oriented/intangible knowledge
resource flows
As is already clear from the foregoing, intangible — often process-oriented —
knowledge resource flows are as important as their more tangible equivalents.
Often the two are co-produced. A KIBS provider that offers a software solution
to a client firm will not only produce a knowledge flow in the form of a ready-
made software package. The software developers will learn about the firm in
which it will be applied, establish a working relationship with the in-house experts,
possibly “en passent” advice on other topics, introduce the client to a network
of other users of similar software, etc. The software provider may use the client
as a reference (helping to maintain his reputation), fine-tune the software package,
etc. Table 3 outlines some tangible and intangible knowledge flows using the
four dimensions of the service innovation model described in the third section.
Dimension 2: Human-embodied knowledge versus non-human knowledge
resources
The second dimension refers to the degree to which knowledge is “embodied”
in humans. Human-embodied knowledge flows require face-to-face interaction

514 P. den Hertog
Table 3. Examples of tangible and intangible knowledge resource flows between service provider
and client on the four dimensions of the service innovation model.
Service Concept Client Interface Delivery System Technological
Options
Tangible Campaign of an Report delivered Marketing training Installation of a
knowledge advertising by marketing for front office new data mining
flows company for bureau on market service employees. software
positioning a new prospects for an programme.
shop formula on the electronic home
market. shopping service.
Intangible Experience of hired Invitation to Hired expert act as a In house software
knowledge expert on similar present the new mirror e.g. by con- team brainstorms
flows campaigns in the service on an fronting the client with hired expert
industry. international firm with the quality on new business
marketing of the “service en- opportunities
conference (new counter” as perceived using the new
networks, new by competitors. software.
contacts).
Table 4. Human- and non-human-embodied knowledge flows between service provider and client:
Examples in terms of the four dimensions of the services innovation model.
Service Concept Client Interface Delivery System Technological
Options
Human- A management Organise a user Employees of the Instruction by a
embodied guru gives a panel with a client firm receive an maintenance
knowledge vision on firm’s clients to on the job training worker on how to
flows electronic test a prototype (by external experts) handle the new
commerce-based service. on dealing with copier.
service concepts customers.
in a face to face
meeting.
Non-human- Reading a report Install a website An action plan by a A CD ROM
embodied on state of the art to communicate management containing an
knowledge innovation with (potential) consultant for interactive demo of
flows strategies in clients. reorganising the firm the e-commerce
service firms. into well focused encounter.
SBUs.

Knowledge-Intensive Business Services as Co-Producers of Innovation 515
between service provider and client firm. Disembodied knowledge flows are
typically written down (a report, an action plan, an article in a magazine, an
electronic database) or incorporated in capital goods or a piece of equipment.
Human-embodied knowledge flows are generally thought of as relatively important
in services in general. However, from the example, it can be gathered that written
communication and technology do play an important role as well, most often
in combination with human-embodied knowledge flows. In Table 4, further examples
are given.
Dimension 3: Explicit/codified knowledge versus tacit/non-codified knowledge
The Nonaka and Takeuchi model stresses the conversion processes in which
tacit knowledge becomes explicit, recombined and is again internalised (in an
enriched version). Discussion of economic transactions usually brings explicit
knowledge to mind. But while there is rarely a price tag on the exchange of
tacit forms of knowledge (which are much harder to pinpoint), they are at least
as important in the interaction between KIBS and their clients. Table 5 gives
some examples.
Table 5. Examples of explicit and tacit knowledge flows between service provider and client firm on
the four dimensions of the services innovation model.
Service Concept Client Interface Delivery System Technological
Options
Explicit/ Read a chapter on Purchase the SAP Obtain the Read the product
codified launching new customer relations requirements for documentation on
knowledge service products in module from obtaining an ISO how to handle the
flows the latest service BAAN company. 9000 certificate for new colour
management book. the service photocopier.
organisation.
Tacit/non- During a weekly Sharing the feeling Participate in a one- Engineers of the
codified cafe visit, two between an external day seminar on data contracted
knowledge friends — one and internal inter- warehousing and engineering firm
flows working for a active designer of discuss new and oil company
insurance firm, the what “feels” as an opportunities with a share best
other in space appealing web site software sales practices during
research — discuss design. representative. their two months
arrangements for at sea installing a
financing satellites. new oil rig.

516 P. den Hertog
Dimension 4: Contractual versus non-contractual forms of knowledge
Another important feature of the knowledge flows taking place between KIBS
and client firms is the degree to which these are part of a contractual relationship
or simply occur without a contract between KIBS and the client firm (Table 6).
Most often, contractual and non-contractual forms of knowledge exchange coincide,
and especially when KIBS have a more or less steady relationship with a client,
the contractual knowledge flows are liable to be supplemented with more informal
types of knowledge flows. This is not only the result of KIBS trying to link
client firms, but also a matter of experts or professionals of both KIBS and
the client firm developing (trustful) relationships.
Table 6. Examples of contractual and non-contractual knowledge flows between service provider
and client firm on the four dimensions of the services innovation model.
Service Concept Client Interface Delivery System Technological
Options
Contractual Contract an Hire a marketing Order consultancy Contract an
knowledge external designer research firm to firm to improve client engineering firm
flows to design a new as-sees how many friendliness of after to help procure a
line of differently customers might sales service piece of
positioned switch to department. machinery/
products. e-commerce. hardware.
Non- Discuss new A software bureau A trainer discusses Experience as an
contractual business specialised in “call after the training with expert the
knowledge opportunities centres” suggests to the management the advantages of an
flows during a meeting contact a situations he/she electronic
of a professional specialised experienced by boardroom
association. temporary work competitors. session, and
agency for pool decide it to use it
management. in one’s own
projects.
There is little statistical evidence available illustrating the role KIBS play
in co-producing innovations. Table 7 presents data from the Dutch Statistical
Office, on the development of computer and related IT services, economic advisory
services (business consultancy) and engineering and technical advisory services,
over the period 1987–1994. It displays the speed at which the number of firms
and turnover of these industries have developed, and provides a substantial estimate
of the turnover realised from knowledge transfer to the client firm or organisation.

Knowledge-Intensive Business Services as Co-Producers of Innovation 517
Table 7. Number of firms (× 1000), turnover (in million guilders) and percentage
of turnover with knowledge transfer in three KIBS industries in the Netherlands
for the period 1987–1994.
Computer and Related IT Services
Year Number of Firms Turnover Turnover with Knowl.
Transfer
1987 3,3 4 688 48
1988 3,9 5 163 54
1989 4,6 5 862 56
1990 5,0 6 608 56
1991 5,5 7 307 58
1992 5,8 7 550 56
1993a 6,9 8 008 59
1993b 8,5 8 777 58
1994 9,1 9 071 59
Economic Advisory Services (Business Consultancy)
1987 4,4 1 936 96
1988 5,5 2 216 96
1989 6,8 2 535 90
1990 7,7 3 070 92
1991 10,0 3 489 93
1992 11,6 3 924 95
1993a n.a. n.a. n.a.
1993b 13,9 4 026 84
1994 15,5 4 420 83
Engineering and Technical Advisory Services
1987 n.a. n.a. n.a.
1988 n.a. n.a. n.a.
1989 7,5 7 191 85
1990 7,9 7 959 87
1991 9,0 9 076 95
1992 9,8 9 913 94
1993a 11,3 11 772 96
1993b 10,8 10 224 98
1994 11,4 10 392 98
Note: Comparisons over time are complicated by a shift in the Standard Industrial
Classification in 1993. Here, we include the figure for (a) the “old” and (b) the
“new” industrial classification scheme. For each of the three industries, CBS
defined more precisely what activities involved knowledge transfer. For example,
in computer and related IT services, database services and selling of hard- and
software are seen as activities in which not much knowledge transfer is involved,
as opposed to activities, such as systems development and secondment of IT
personnel.
Source: CBS (1996: 94/95)

518 P. den Hertog
Table 8 is possibly even more intriguing: based on the make-and-use tables
in the national accounts, it gives some insight on the use of these three KIBS
industries by various client categories. Some remarkable patterns can be noticed,
bearing in mind the size of the user industries (as revealed by their added value
and mean level of investment). Economic advisory services are used above average
by manufacturing, whereas banking and insurance firms use a relatively above
average amount of computer and related IT services. It is interesting to see
that even in the well-developed Dutch service economy, the services of research
and scientific institutions (in the fourth column — these are largely public services)
are geared relatively more towards the manufacturing than to the service industries.
This brings us to the next question, i.e. whether KIBS are developing into a
second, private, knowledge infrastructure partly complementing and partly
competing with the first, established, public knowledge infrastructure.
KIBS as a “Second Knowledge Infrastructure”
KIBS have come to play a central role in transferring and, in many cases, creating
and combining, knowledge resources in innovation systems. They are doing this
directly through their provision of services and indirectly through facilitating
the mobility of highly educated personnel.11 The direct roles played by KIBS
have much in common with the roles of organisations within the public knowledge
infrastructure, such as research and technology organisations (RTOs) and higher
education institutions (HEIs). The latter two also play a role in diffusing knowledge
to the various firms and organisations they work with through contract research,
educating students and providing training to personnel of client firms.
Given their role as co-producers of knowledge and innovation with client
firms, the rise of KIBS can be seen as contributing to a new knowledge infrastructure.
This new knowledge infrastructure is no longer largely constituted by RTOs
and HEIs of different kinds, but also includes KIBS. In the context of the
development of the knowledge infrastructure, we have postulated two hypotheses
concerning the role of KIBS in innovation processes (den Hertog & Bilderbeek,
1997b: 31):
(i) KIBS will in practice gradually develop into an informal (private) second
knowledge infrastructure or knowledge base, partially complementing and
partially taking over the intermediary role traditionally played by parts of
the more institutionalised, formal (public) first knowledge infrastructure.
11KIBS have high shares of highly educated staff, apparently with high levels of mobility.

Knowledge-Intensive Business Services as Co-Producers of Innovation 519
Table 8. Use of KIBS by manufacturing and non-manufacturing industries in the Netherlands.
The use of Services Provided by: Size of Using
Industries
Industry Computer Engineering Economic Research % of % of
Service/ & Advisory & Total Investment
IT Firms Technology Firms Scientific Value (Average
Consultancy Institutions Added Level
Firms (1993) 1986–92)
Manufacturing 1 375 966 954 1 324 19 17
— High tech 237 99 140 316 3 2
— Medium tech 693 410 513 654 7 7
— Low tech 445 457 301 354 9 8
Non-manufacturing 4 643 6 307 2 070 1 953 81 83
— Agriculture & 19 191 12 14 3 5
fisheries
— Mining 161 114 4 15 3 2
— Construction & 140 52 91 29 5 2
install.
— Utilities 122 193 46 53 2 4
— Wholes./retail./ 854 337 379 77 12 9
rep.
— Transp./wareh./ 521 273 36 25 16 11
comms.
— Financial 470 218 2 29 4 4
services
— Business 1 245 3 742 1 078 105 16 30
services
— Other services 1 031 1 041 416 1 601 19 17
Consumption in 0 212 0 0
private households
Consumption in 0 30 0 129
public sector
Exports 1 405 1 534 1 321 1 197
Total 7 423 9 049 4 345 4 603 100 100
Source: CBS (1996: 96)

520 P. den Hertog
This implies that this category of service firms would function as a diffusion
agent or even as a source of innovation for their clients — contrasting
with the popular image of service firms as innovation followers.
(ii) The traditional distinction between public and private knowledge-based
(advisory) services will gradually disappear. This could lead to a development
in which not firms and institutions but networked service professionals —
irrespective of the formal organisation to which they belong — will increasingly
act as carriers and sources of knowledge. This process of blurring of boundaries
will eventually result in a more flexible capacity of external KIBS professionals
cooperating with internal KIBS professionals in providing knowledge-intensive
business services.
The notion of the development of a second knowledge infrastructure forms
the basis of a three-tiered model of the evolving role of KIBS in innovation
processes, as given in Fig. 4 and elaborated in Table 9. This model needs further
development and testing. While the shift from phase 1 to phase 2 is relatively
well-substantiated, phase 3 is much more an extrapolation from a few apparent
vanguard developments. For instance, networks are emerging in which professionals
operate rather loosely between organisations, sometimes combining various
assignments, and there seems to be increasing mobility of personnel between
the various organisations in the (now broadly defined) knowledge infrastructure.
Given these trends, and the gradual shift from the first phase to the second
phase of KIBS development; it will be well to be prepared for developments
towards the third phase.
There is some evidence of a blurring of the boundaries between services
offered by the public knowledge infrastructure and KIBS services. Nevertheless,
the two infrastructures generally play different roles within innovation systems.
Universities primarily have relations with large R&D-intensive manufacturing
firms and (in the case of social and administrative knowledge) the public sector.
KIBS firms have a much broader spectrum of clients, including public authorities
and some smaller firms. Large firms and other organisations benefit
disproportionately from both knowledge infrastructures. SMEs, with their relatively
low levels of internal competence and limited financial resources, often lack
capabilities for making effective use of KIBS, and typically rely on public or
semi-public sources for external knowledge.
The picture is complicated by evidence confirming that national innovation
systems differ considerably in the degree to which KIBS function as important
co-producers for other industries. At a macro-level, Windrum and Tomlinson
(1999: 402) showed that the degree of integration between KIBS and other economic
activities differs considerably between the United Kingdom, Netherlands, Germany

Knowledge-Intensive Business Services as Co-Producers of Innovation 521
Phase 1: Embryonic stage of KIBS development
KIBS
Predecessors
Public Private
Knowledge Knowledge
Base Base
Phase 2: KIBS as a 2nd knowledge infrastructure
Public Private
KIBS KIBS
specialised KIBS func- specialised KIBS func-
tions provided internally tions provided internally
Public KB Private KB
Phase 3: Networked KIBS service professionals
external(-ized)
KIBS professionals
internal internal
KIBS professionals KIBS professionals
Public KB Private KB
Fig. 4. Three-tiered model of the role of KIBS in innovation processes.

522 P. den Hertog
Table 9. Some characteristics of the individual phases in the three-tiered model.
Phase 1: “Embryonic” Stage of KIBS Development
• Limited interaction between the public and private knowledge base
• Limited number of intermediary knowledge institutions (mainly public) and firms
• Emphasis in innovation processes on generating new knowledge
• Predecessors of KIBS functions mainly coupled to well established categories of professional staff
(R&D, accounting, marketing, legal affairs, etc.) and to a substantial degree provided within the
firm
• Knowledge mainly interpreted as formal technical expertise (R&D)
• Sectoral knowledge orientation dominates
• Innovation policies mainly focused on supporting R&D/increasing knowledge bases
Phase 2: “KIBS as 2nd Knowledge Infrastructure”
• Interaction between public and private knowledge bases increasingly considered essential
(“economies of scope in S&T”)
• Increasing number of intermediary knowledge institutions and firms
• Emphasis in (interactive) innovation processes on generating new knowledge and diffusion
• KIBS increasingly identified as a separate category of knowledge generators/diffusers, although a
clear separation between public and private KIBS remains
• Explicit “make or buy” decision concerning provision of KIBS functions
• Broadened definition of knowledge, various kinds of formal knowledge and tacit knowledge
(intangibles)
• Knowledge orientation crosses sectoral boundaries. Network and cluster perspective starts to develop.
• Broadening of innovation policies (more aspects, more actors)
Phase 3: “Networked ‘KIBS’ Service Professionals”
• Increasingly non separable and overlapping public and private knowledge bases
• KIBS recognised as significant intermediate actors in innovation processes in public and private
sectors
• Increasing combination of innovative service functions in new products and services. Normalisation
of innovation in service functions.
• Public and private organisations & firms develop knowledge management systems and seek actively
the help of KIBS
• Well developed user-producer linkages between internal-external KIBS professionals
• KIBS professionals increasingly combine various roles (entrepreneur, scientist, consultant, staff
member) and function in a network of service professionals
• Tasks traditionally performed by public policy-makers increasingly performed by semi-public (at
arms length) or private KIBS professionals

Knowledge-Intensive Business Services as Co-Producers of Innovation 523
and Japan. They argue that those countries with strong, mutually beneficial links
between business services and other (most notably manufacturing) activities enjoy
higher spill-over effects from service innovation.
Implications for Innovation Management and Innovation Policy
The model just presented needs (further) empirical testing.12 Particular attention
needs to be paid to cross-country differences in terms of institutional structures,
and to the levels of integration of business services (especially KIBS) and
manufacturing. Further, the model should be linked to the distinction between
“mode 1” and “mode 2” knowledge production in science and technology —
provided by Gibbons et al. (1994). Their basic argument is that a distinctly
new set of cognitive and social practices in the production, legitimisation and
diffusion of knowledge is emerging. This so-called “mode 2” is different from
those practices governing the largely Newtonian model of mode 1 or what many
would label as science (Gibbons et al., 1994: 2–3). The argument cannot easily
be summarised in one table,13 but Table 10 outlines some characteristics of
the two modes of knowledge production.
Table 10. Some characteristics of “mode 1” and “mode 2” knowledge production (based on Gibbons
et al., 1994: 3–8).
Mode 1 Mode 2
Problems set and solved in a context Knowledge produced in the context of application
governed by the (academic) interests of a
specific community
Disciplinary Transdisciplinary
Homogenity Heterogenity
Hierarchical organisation Organisational diversity
More socially accountable and reflexive More socially accountable and reflexive
Quality determined mainly through peer More composite, multidimensional quality control
review judgements on a disciplinary basis process
12This is underway, in particular, in the ongoing RISE project studying “RTOs in the Service Economy”
led by Dr. Mike Hales of CENTRIM, University of Brighton.
13Theirwide-ranging analysis covers interaction between those involved in mode 1 and mode 2
knowledge production and the role of IT in this; the expansion of both knowledge providers and
knowledge users; the implications for “traditional” knowledge-generating institutions, such as
universities, government research organisations and industrial laboratories; and the shaping of
innovation policy.

524 P. den Hertog
KIBS can be seen as a specific example of the “mode 2” variety of knowledge
production. They share many of the characteristics of the “mode 2” way of
producing knowledge, most notably: its organisational diversity, orientation to
applications, and transdisciplinarity. The characterisation of “mode 2”, for example,
as “problem-solving capability on the move” (Gibbons et al., 1994: 5) very
neatly describes what KIBS in practice are. This also brings us back to the
more general description of what it means to produce a service — recall the
definition of Gadrey et al. (1995): “to produce a service […] is to organise
a solution to a problem […] to place a bundle of capabilities and competences
(human, technological, organisational) at the disposal of a client and to organise
a solution, which may be given to varying degrees of precision”.
Thus, the rise of KIBS has important implications for both innovation
management and innovation policy, including the following:
Innovation management
Since KIBS play a role as co-producer of innovation in some of the client firms
they serve, questions arise concerning:
• Whether and how industries differ in their actual “consumption” of KIBS;
• Whether KIBS play similar roles in co-production of innovation in all developed
countries;
• Which are the vanguard KIBS industries; and which KIBS industries are most
likely to follow a similar development pattern in the near future?
• How do in-house and outsourcing strategies, both in KIBS and client firms,
affect the scope for co-production of innovation?
• What type of appropriability strategies do KIBS use themselves?
• How do KIBS firms take stock of the lessons learned in the interaction with
client firms?
Innovation policy
As KIBS play a role in transferring knowledge in innovation systems, there
may be consequences for the way in which innovation policy is shaped. Given
indications that (some) service industries are particularly prone to use services
of KIBS, innovation policy-makers might consider the scope for using KIBS
for realising their policy goals. This is especially so as “traditional” institutions
within the public knowledge infrastructure seem less able to function in the
service economy, and are more geared towards servicing established manufacturing
industries. Among the questions that arise in this context are:

Knowledge-Intensive Business Services as Co-Producers of Innovation 525
• How other bridging institutions — such as applied research organisations,
government research labs, transfer agencies and even universities — cope
with the rise of KIBS?
• To what extent KIBS can be used to perform knowledge transfer and innovation
support programmes for government? If so, how?
• How to balance the innovation functions provided by KIBS and “traditional”
(mode 1) institutions in individual industries and clusters?
These challenging research themes will hopefully fuel further debate on the
role of services, and particularly KIBS, in innovation processes and innovation
systems.
Acknowledgements
The insights presented here are the result of discussions with Prof. Ian Miles
and Rob Bilderbeek as well as various colleagues and former colleagues —
especially Sven Maltha, Prof. Ruud Smits and Dr. Jos Leyten — and numerous
fellow “services” researchers abroad. Most notable are those who participate(d)
in the “Services in Innovation, Innovation in Services” (SI4S contract number:
SOE 1-CT96-1015) and the “RTOs in the Service Economy” (RISE contract
number: SOE1-CT98-1115) projects — both sponsored through the European
Union DG XII TSER programme — and the ongoing SIID project on Strategic
Information Provision on Innovation in Services, undertaken together with
Groningen University and financed by the Dutch Ministry of Economic Affairs.
The comments of the two referees have also proved useful. However, the text
presented here is the sole responsibility of the author and the usual disclaimers
apply.
References
Barras, R. (1990) Interactive innovation in financial and business services: the vanguard
of the service revolution. Research Policy, 19, 215–237
(1986) Towards a theory of innovation in services. Research Policy, 15,
161–173
Bessant, J. & Rush, H. (1995) Building bridges for innovation: the role of consultants
in technology transfer. Research Policy, 24, 97–114
(1998) Innovation Agents and Technology Transfer, SI4S Article 09, University
of Brighton: CENTRIM
Bilderbeek, R. & den Hertog, P. (1992) Innovatie in diensten. Position paper. [Innovation
in services. Position paper], Apeldoorn: TNO Strategy, Technology and Policy

526 P. den Hertog
(1996) Innovation in knowledge intensive business services. Lessons from
case studies. Paper presented at the Roskilde Seminar on Innovation in Business
Services. Roskilde University, January 1996
(1997) The interactiveness and innovative role of technology-based knowl-
edge-intensive business services (T-KIBS). TSER-SI4S-project, TNO-SI4S Report
No. 3, Apeldoorn: TNO Strategy, Technology and Policy
(1998) Technology-based knowledge-intensive business services in the
Netherlands: their significance as a driving force behind knowledge-driven innovation,
Vierteljahrshefte zur Wirtschaftsforschung, 67(2), 126–138
Bilderbeek, R., den Hertog, P. & Chehab, N. (1998a) Management van vernieuwing
in diensten. Verslag van een workshop voor ondernemers op 22 april 1998 [Management
of innovation in services. Report of a workshop with entrepreneurs]. Utrecht/Apeldoorn:
Dialogic/TNO Strategy, Technology and Policy
(1998b) Innovation and services, Report of an expert workshop for the
Dutch Ministry of Economic Affairs. Utrecht/Apeldoorn: Dialogic/TNO Strategy,
Technology and Policy
Bilderbeek, R.H., den Hertog, P., Huntink, W., Bouman, M., Kastrinos, N. & Flanagan,
K. (1994) Case studies in innovative and knowledge-intensive business services.
Apeldoorn: TNO Strategy, Technology and Policy (STB/94/041)
CBS (1996) Kennis en Economie [Knowledge and the Economy], Voorburg/Heerlen:
Centraal Bureau voor de Statistiek
den Hertog, P. & Bilderbeek, R. (1997a) Recent innovation patterns in services in
the Netherlands. Report prepared within the framework of the SI4S-project, Apeldoorn:
TNO Centre for Technology and Policy Studies
(1999) Conceptualising service innovation and service innovation patterns.
Research Programme on Innovation in Services (SIID) for the Ministry of Economic
Affairs, Utrecht: Dialogic
(forthcoming) The New Knowledge Infrastructure: The Role of Knowledge
Intensive Business Services in National Innovation Systems. In Services, Innovation
and the Knowledge-Based Economy, ed. I. Miles & M. Boden. London: Continuum
den Hertog, P., Bilderbeek R. & Maltha, S. (1997) Intangibles, the soft side of innovation.
Futures, 29(1), 33– 45
den Hertog, P., Bilderbeek, R., Marklund, G. & Miles, I. (1998) Services in innovation:
knowledge-intensive business services (KIBS) as co-producers of innovation. SI4S
Synthesis Paper No. 3, Oslo: STEP
Freeman, C. & Perez, C. (1988) Structural crisis of adjustment: business cycles and
investment behaviour. In Technical Change and Economic Theory, ed. G. Dosi
et al. London: Pinter Publishers
Gadrey, J., Gallouj, F. & Weinstein, O. (1995) New modes of innovation. How services
benefit industry. International Journal of Service Industry Management, 6(3),
4–16
Gallouj, F. & Weinstein, O. (1997) Innovation in services. Research Policy, 26,
537– 556