tit

M

ity Mnive

26 Mine 1

nalimpcesprodly i

underpin this trend.

tion, embodiment or use and hence deriving implications for the

certified repairers and so on. In this network, there are a variety ofservice providers and different types of service interactions that

or a service is determined by what Coase (1992) loosely called

of examining the emergence of the service-based economy, haveprovided valuable insights into the nature of services. In thefourth section, we examine a classification of services based onthree demand rationales and in the fifth section we look at the

Industrial Marketing Managementmanagement of service as opposed to product-based firms.In this paper, we argue that the quest for foundational differ-

ences between products and services is misguided. First, this isbecause many firms in industrial markets are both product- andservice-based, andwe need to understand better how products andservices interact within firms, markets and supply networks. Aproduct such as laptop computer, for example, inserts itself into anetwork comprising manufacturers, retailers, software firms,users with a set skills and expectations, Internet service providers,

the institutional structure of production, namely the institutionalarrangements that govern production and exchange in a partic-ular industrial system. The distinction between products andservices, we argue, relies on these structures rather than on anytechnical criteria involved in their production or use.

The paper is organized as follows: in the second section wereview the literature on the serviceproduct distinction, focusingmainly on marketing contributions. In the third section, we lookat a neglected contribution from economists who, in the course1. Introduction

This paper revisits the long-standing duality betweenproducts and services that has been a key point of departure inthe marketing literature. Traditionally, the literature has focusedon establishing clear differences between products and servicesbased on technical characteristics associated with their produc-

accompany the life of the product from the moment it rolls out ofan assembly line to its final disposal. These services are highlyheterogeneous, with some that are tightly scripted, requiringminimal interaction between producer and user (e.g. connectionto the Internet) and others that are more customized, open-endedand interactive (e.g. detection and repair of faults).

Our second main argument is that what counts as a product 2006 Elsevier Inc. All rights reserved.

Keywords: Product and service definitions; Tradability of services; Manufacturing move into servicesServices, products, and the ins

Luis Araujo a,,a Department of Marketing, Lancaster Univers

b Department of Management Science, Lancaster U

AcceptedAvailable onl

Abstract

This paper revisits the productservice distinction from an institutiofocused on the intrinsic characteristics of services with a view to deriveis that the quest for foundational differences between products and servinature of produceruser interactions and the institutional structure ofFurthermore, we develop the argument that services play an increasing Corresponding author. Tel.: +44 1524 593915.E-mail addresses: L.Araujo@Lancaster.ac.uk (L. Araujo),

M.Spring@Lancaster.ac.uk (M. Spring).1 Tel.: +44 1524 593739.

0019-8501/$ - see front matter 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.indmarman.2006.05.013utional structure of production

artin Spring b,1

anagement School, Lancaster LA1 4YX, UKrsity Management School, Lancaster LA1 4YX, UK

ay 20063 July 2006

perspective. Much of the literature in marketing and management haslications for the management of service-based firms. Our key argumentis misguided. What counts as a product or a service is dependent on theuction rather than on any essentialist feature of products or services.mportant role in manufacturing firms and we explore the reasons that

35 (2006) 797805tradability of services. In the sixth section, we look at recent callsfor manufacturers to expand into services and argue that anexpansion in service provision is the inevitable by-product of an

rketincreasing division of labor and the disaggregation of corporatehierarchies. The paper concludes with a number of implicationsfor firms offering productservice combinations rather than pureproduct or services.

2. The serviceproduct distinction

The distinction between products and services has a longhistory in marketing and other disciplines, namely operations(see e.g. Johnston, 1999) and general management (see e.g.Bowen & Ford, 2002). In marketing, the focus was firmly onproducts, with services often regarded as little more than aperipheral activity that added value to a product. Kotler (1980),p. 374, for example, noted cryptically a service is not aphysical thing but rather an energy expenditure.

In industrial marketing as far back as the late 1970s (Grnroos,1979; Gummesson, 1978), there were calls for services to betreated as part of companies' marketing strategies. The focus ofthe service literaturewas often on the classification of services andthe marketing implications stemming from their idiosyncraticqualities (Lovelock, 1983) or the development of servicepackages according to market segments (Boyt & Harvey, 1997).Fisk, Brown, andBittner (1993), p. 68, for example, remarked thatthe four features intangibility, inseparability, heterogeneityand perishability provided the underpinning for the case thatservices marketing is different from goods marketing.

To this day, textbooks (e.g. Kotler, 2003) and academic papers(e.g. Bowen&Ford, 2002) rely on these dimensions to convey thenotion that services deserve special treatment. In keeping withsome of these authors, we will refer to these features orcharacteristics as the Intangible, Heterogeneous, Inseparable,Perishable (IHIP) characteristics. The implication that can beextracted from these statements is that services are what productsare not, and that product-based marketing needs to be adapted toservice environments. As Grnroos (1998), p. 235, memorablyput it: The most important change from the product marketingsituation is that the product is missing. Only occasionally haveservice marketing authors acknowledged that importantdifferences exist between service firms, not just between servicefirms and goods firms (Zeithaml, Parasuraman, & Berry, 1985,p. 43).

In addition, the literature has often stressed the outcome-versus-process duality to define services. According to this view, whereasproducts appear as pre-produced at the consumption junction, theproduction of services is performed in a process in whichcustomers interactwith the production resources of the service firm(Grnroos, 1998). This view can be questioned. First, it is notunusual in business markets for users to either request customizedproducts or to actively negotiate and iterate with suppliers thespecifications they require (Araujo, Dubois, & Gadde, 1999;Spring & Dalrymple, 2000). Secondly, it is common for firms inprofessional services such as law or advertising to producecustomized solutions based on repeatable packages either throughstraightforward repetition of tried and tested processes or

798 L. Araujo, M. Spring / Industrial Marecombination of pre-produced modules (Grabher, 2004; Sundbo,1994). Finally, there are many service examples when thecustomer's interaction with the resource structure of the producercan be fairly minimal or inconsequential, as when a utilitiescustomer switches on the lights at night or connects a computerthrough a modem to an Internet service provider.

More recently, service marketing writers have begun to attackthe foundational differences that have helped services marketingdevelop as a separate field. Vargo and Lusch (2004a,b) andLovelock and Gummesson (2004) attempt to debunk the mythsunderpinning the IHIP characteristics. When examined moreclosely, as we have begun to show already, these are foundwanting, and numerous exceptions can be invoked to counter thesupposedly watertight distinction between products and ser-vices. For example, tangible goods are often heterogeneous andmany services can be mass produced without difficulty, as Levitt(1972) noted long ago. The variability of services can becontrolled through strict quality procedures, standardization ofservice modules, and automation of service functions (Lovelock& Gummesson, 2004).

In an argument reminiscent of Levitt's (1972) aphorism thateverybody is in services, Vargo and Lusch (2004a), p. 326,propose that services are more universal than goods. In theirwords, services are the application of specialized compe-tences (skills and knowledge) through deeds, processes andperformances for the benefit of another entity or the entity itself(self-service). The corollaries of the argument that everythingis a service, is that manufacturing and exchange should be seenas services, and products merely as vehicles for service delivery.Furthermore, Vargo and Lusch (2004b) contend that marketinginherited from economics the notion that value is embedded inproducts and usage is directly related to product features.Instead, they propose that through product usage, customers areco-opted in the value creation and delivery process. Exactlyhow value through product usage is created, we are not told.

Levitt (1981) pinpointed the limits of the traditionaldistinction between products and services and focused insteadon the marketing of tangibles and intangibles. The nub ofLevitt's argument is the distinction between those products andservices whose qualities can be experienced and tested inadvance and those that cannot. If it is true that intangibles cannotbe assessed and tested in advance, it is equally true that for manycomplex products the process of installation and appropriationinto the user's value system involves a lot more than the productitself. As Levitt (1981), p. 97, puts it: you won't know how itperforms until it's put to work. The conclusion that Levittextracts is that whereas tangible products must be intangibi-lized to stress benefits in use, intangible products must betangibilized or to create metaphors or surrogates fortangibility (Levitt, 1981, p. 100). Relationships with customersin the case of intangibles need to be managed more carefully andcontinuously than in the case of tangibles although Levittstresses that this is a vital for products too, notably in the case ofnew and complex products.

In summary, neither the productservice distinction foundedon the IHIP characteristics nor one based on process-versus-outcome consumption bears much scrutiny. However, recent

ing Management 35 (2006) 797805contributions suggesting that the balance should be swungtowards services have shied away from examining why, how andwhen particular productservice combinations should be

owner (B) of a car (C) may request a garage (A) to repair the car;or, the owner (B) of his own body (C)may request a doctor (A) toprovide medical repairs. They point out that such situationsalways involve a triangular relationship between A, B and C; thisis illustrated in Fig. 1.

4. Three demand rationales for services

The examples used above to illustrate the service triangleare both what Hill termed requests for assistance or inter-

799rketing Management 35 (2006) 797805deployed to address particular types of demand. The trick ofturning the tables by affirming that everything is a serviceelides the issue of the role of products in service provision. Putdifferently, why are somany services delivered through productsrather than through other means?We shall return to this question.First of all however, we look at approaches to the definitions ofservices that have been used by economists addressing theservice-based economy from an institutional and nationalaccounts' perspective.

3. An institutional perspective on service definition

The central concern of the literature on the productserviceduality has been to establish what significant differences existbetween products and services, and hence how product- andservice-firms should be managed. It should be clear by now thatwe suggest that this may be a misguided ambition. Nationalaccounting economists have other reasons for making thedistinction. Given the concern they have with questions such asIs the service economy growing?, they require a basis fordefining what is in the service economy and what isn't. In thisregard, the contributions of Hill (1977, 1999) and Gadrey (2000)deserve a special mention. Hill (1977) argues that a necessarycondition for some item to be a good or a service is that it must becapable of being the subject of a transaction between two ormore different economic units. To identify the characteristics ofgoods or services, the focus should be on the interaction betweenproducers and users.

For Hill (1977), p. 318, a service may be defined as achange in the condition of a unit or a person, or of a goodbelonging to some economic unit, which is brought about as aresult of the activity of some other economic unit, with the prioragreement of the former person or economic unit. The focus onproduceruser interfaces leads Hill (1977), p. 320, to reflect onhow the institutional structure of production is critical for thedefinition of services:

one and the same activity such as paintingmay be classifiedas a good or a service depending purely on the organization ofthe overall process of production amongst the differenteconomic units. [] some changes in the share of serviceindustries in total output or employment may be determinedpurely by changes in which production is distributed amongstdifferent producers and may have no influence whatsoever onthe share of services in the final expenditure.

Crucial to our discussion above, it is clear that technicalcharacteristics of the offering such as IHIP play no part at all inHill's approach to distinguishing between products and services.

Delaunay and Gadrey (1987) and Gadrey (2000) build onHill's work in two ways. First, they develop Hill's definition bypositing that a service activity is an operation intended to bringabout a change in the status in a reality C that is owned byconsumer B effected by service provider A at the request of Band in many cases in collaboration with B, without leading to the

L. Araujo, M. Spring / Industrial Maproduction of an entity that can circulate independently of me-dium C-circulation understood here as the transfer or propertyrights rather than circulation in space. Thus, for example, thevention. He observes that other services have a different ratio-nale, namely the provision of maintained technical capacitiesthat customers can avail themselves of in exchange for payment(e.g. use of a hotel room). Gadrey (2000) adds a third rationale:live performance (e.g. attending the theatre). These differentrationales have direct implications for the types of customersupplier interaction required and the relational as well as spatialproximity of these interactions. In the first and latter cases,relational as well as spatial proximity is required whereas thesecond case does not imply relational or spatial proximity. Forexample, customers can avail themselves of a technical capacitythrough technologically mediated interactions at a distance (e.g.telephone lines, websites) and without relational proximity.

Gadrey's three service logics evoke access to a socio-technicalcapacity embodied in specialized human competencies andtechnical apparatus. In the case of request for assistance orintervention, this socio-technical capacity consists of means forthe performing diagnostics, maintenance or repair to produce thedesired effects on an object owned by the purchaser of the service.In the access to maintained technical capacities logic, thiscapacity may often be invisible to the user and may involve acomplex logic of supply and access (e.g. Internet banking serviceprovision). Under the live performance logic, this capacity isembodied in a complex set of scripts involving both humanperformances and material arrangements (e.g. the stage andseating in a theatre).

Hill's and Gadrey's contributions clarify the nature ofservices and place produceruser interaction at the center ofattempts to distinguish between products and services. However,the notion that services cannot circulate as independent entitiesin a property rights space begs the question of what makesservices tradable at all. In addition, the productservice distinc-tion relies too much on a clear-cut separation, institutional aswell spatial, between production and exchange. For manyFig. 1. After Delaunay and Gadrey (1987).

rketproducts, as we highlighted earlier, there is a more intricaterelationship between production and exchange and closer spatialas well as relational proximity between producers and users.

5. The tradability of services

Hill (1977), p. 336, claims that a service is produced byone economic unit for another, but it is not exchanged betweenthem. A typical definition of market exchange matches theexchange of property rights to the physical transfer of productsin exchange for money (Hodgson, 1988). If that is so, how areservices made tradable?

In Hill's schema, products are not just eminently tradable, buttheir independence fromproduceruser interactions allows a clearphysical separation between production and exchange activities.AsDemsetz (1993) remarked, products can be seen as embodyingspecialized knowledge in a way that is highly advantageous forpromoting the division of labor. Products allow knowledge to bepackaged in a way that makes knowledge transfer between firmsinexpensive and practical since the instructions needed to useproducts do not require in-depth knowledge about how they areproduced. Products thus act as a means of separating productionfrom exchange and delimiting userproducer interaction usersonly have to be instrumentally knowledgeable while substan-tively ignorant about the products they purchase, to use Loasby's(1998), p. 169, felicitous expression.

But two contributions that we now discuss demonstrate, indifferent ways that, once again, the productservice duality isnot so clear-cut. They show that making products tradable isnon-trivial it requires work. Once this is accepted andunderstood, it becomes evident that, in the same way, if this typeof work is carried out, services may also be made tradable.

5.1. Transfers and transactions

Baldwin and Clark (2003) adopt an engineering systemsapproach to defining where transactions should be located in acomplex production system. The driver for defining transactionloci is the notion of mundane transaction costs, or the costs ofdefining what is to be transferred, counting, valuing and payingfor the transfers. At some points in the system, transfers aresimple, and therefore easy to standardize, easy to count and easyto value. At other places, transfers are complex and in some casesimpossible to standardize, impossible to count, and impossible tovalue.

At first sight, only products seem to be tradable accordingto Baldwin and Clarke's definition of transactions. In theiranalysis of a more complex transaction between twoenterprises, where specifications could not easily be translatedinto product characteristics, Baldwin and Clark (2003)contend that the two parties had to become locally andtemporally de-encapsulated with respect to the specificationissue. In their terms, buyer and supplier had to allow a wholeset of material, energy and information transfer to occur

800 L. Araujo, M. Spring / Industrial Maacross their boundaries, which were uncounted, unstandard-ized and uncompensated. The implication drawn from thisexample is that product-based transactions involve a wholeseries of service-based exchanges that resemble intra-firmtransfers.

Penrose (1959) seminal distinction between resources andthe productive services they render provides a good platform forthe following discussion. As Penrose (1959), p. 25, reminds us:

The important distinction between resources and services isnot their relative durability: rather it lies in the fact thatresources consist of a bundle of potential services and can,for the most part, be defined independently of their use,while services cannot be so defined, the very word serviceimplying a function, an activity.

As Penrose (1959), pp. 7475, remarks, it is the heteroge-neity rather than the homogeneity of human and materialproductive services that makes firms unique:

Productive services are not man-hours, or machine-hoursor bales of cotton or tons of coal, but the actual servicesrendered by the men, machines, cotton or coal in the productiveprocess. Although it is manifestly services that in this sense thatare the actual (physical) inputs in production, a less specific ormore indirect definition is usually required when services mustbe expressed as measurable homogeneous quantities, forexample, if it desired to measure the cost of certain productiveservices or to construct technological production functions forcertain outputs'

This is at least in part the problem identified by Baldwinand Clark (2003): to make a transfer into a transaction, there is aneed to standardize, count, and compensate. The crucial insightfor our present discussion is that, to a greater or lesser extent,effort is required, and costs are incurred in making transfers intotransactions, even for products.

The above discussion highlights some of the problemsassociated with understanding what makes products and servicestradable. Baldwin and Clark's (2003) distinction between trans-fers and transactions is a Coasean story of looking at institutionalstructures of production through aggregating activities into larg-er modules through transactional interfaces. When these inter-faces can be made to work, markets take over; when they do not,activities are aggregated into non-modular structures such asfirms (Langlois, 2002).

In short, tradability is associated with the ability to defineand attach property rights to tangible, material entities at pinchpoints in production systems where mundane transaction costsare low. But, as Callon and Muniesa (2005), p. 1233, remark, aservice can be made tradable by transforming it into a thing.Services can be traded by defining, delimiting and objectifyingtheir properties and thus become fit for the attachment andtransfer of property rights even if they have no physical,tangible properties. A car rental for example, defines a period ofusage of a car according to a well-defined script of rules andconditions. In this sense, a car rental is as tradable as the caritself.

The expansion of services is the result of increasing tradability

ing Management 35 (2006) 797805promoted by technological changes, namely advances in infor-mation and communication technologies (ICTs), as well as newconditions of access to socio-technical capacities (e.g. renting

rketadmission to a facility, leasing capital equipment). Advances inICTs allow information to be digitized, stored and transported inways that promote the fragmentation of services both institution-ally, through the emergence of new specialist providers, andspatially. ICTs often allow service provision to be broken down insmall components (e.g. back office functions) enabling firms toproduce and consume services in a variety of locations, in eithersynchronous (e.g. call centers) or asynchronous (e.g. collabora-tive software development) modes (WIR, 2004, p. 148). As far asnew forms of access to socio-technical capacities are concerned,new forms of non-ownership such as rental, hire and lease, areoften the springboards for new service offerings (Lovelock &Gummesson, 2004). The growth in the trade and tradability ofservices raises broader issues concerning what makes productsand services tradable.

5.2. Products/services and goods

Callon, Meadel, and Rabeharisoa (2002) sidestep thedistinction between products and services by focusing on thenotion of goods. A good implies a stabilization of characteristicsat the moment an entity, product or service is ready to be traded.A product is an economic good that can be seen from a varietyof perspectives: production, circulation and use. Thus a productcorresponds to a process, a trajectory in time, whereas a goodcorresponds to a state at a point in time.

Callon (1991), p. 136, defined products as programs ofaction aimed at coordinating a network of distributed roles. Thedefinition of what a product is implies a definition of the socio-technical context it will inhabit which should include the rolesplayed by other objects (e.g. power supplies) as well as differenttypes of human actors (e.g. users, after-sales service people).Products can thus be seen as texts embodying specific programsof action, inserting themselves in particular networks of skills(e.g. what skills do users need to make the product work, whatskills do repair people require to diagnose malfunctions). At thesame time, products are themselves accompanied by a variety oftexts as they move from production, circulation and use (e.g.logbooks, user manuals, warranties). In short, products do notjust circulate as independent intermediaries in pre-establishednetworks but define and describe network of actors connected totheir design, production and use (Callon et al., 2002).

A product's biography can only be traced within thesenetworks. A car, for example, moves from the designer's officeto the prototype stage andmanufacturer test tracks, to productionlines, to an item in amanufacturer's brochure, to transport depotsand dealership showrooms. Once in the hands of its firstregistered owner, it follows another trajectory, of clocking mileson roads, of regular service in maintenance workshops, to usedcar magazines and dealerships and finally, scrap and recyclingyards. The same product can thus have a rich and multifacetedbiography and constitute a different good at different stages of itslife-cycle (e.g. when it is sold new or second-hand). At eachpoint in this trajectory where it is traded (e.g. as a used car) it is

L. Araujo, M. Spring / Industrial Manecessary for the economic entities involved, particularly theowner, to do work to stabilize its properties so as to make ittradable. This may simply be the process of placing anadvertisement specifying the condition, age, mileage and soforth of the car. But for more complex products e.g. houses, thestabilization into a good involves complicated work bysurveyors, estate agents and conveyancing solicitors.

Service provision as Gadrey (2000) noted, involves themobilization and deployment of different forms of socio-technicalcapacities. Making services tradable, as Callon and Muniesa(2005) remark, is a process of definingboundaries around access tomaintained socio-technical capacities, of objectifying what is to beexchanged and of elaborating scripts that govern access to socio-technical capacities. Callon (2002) characterizes the stan-dardization of service provision and service interactions throughwhat he calls writing devices e.g. texts that objectify theelements of service provision through comprehensive descriptions.Using the example of boat trips in the Seine, Callon illustrates howinGadrey's logic of performance situations, the continuouswritingand rewriting of scripts define service content. Some of these textsare paradoxically described as product files, highlighting howthe process of objectification of a service in common languageimperceptibly slips into a product-based discourse.

None of the above arguments invalidates Hill's notion that onecannot attach property rights to services and make them circulateindependently of the relationship between provider and user. Butit highlights the spurious association between tangibility, propertyrights and tradability. As long as services can be bounded,objectified, clearly specified and scripted into socio-technicalcapacities they can be transacted just like products.

A number of implications follow from the above discussion.First, the institutional definition of services favored by Hillposes interesting issues related to how services can be madetradable. For example, in industrial markets customers can availthemselves of technical capacities before and after productexchanges that may be difficult to disentangle and valueindependently from product transactions. In other words, we arestill in the presence of services but services that have not beendisentangled from product transactions and objectified asbounded entities. Conversely, firms may adopt transfer pricingpolicies (e.g. based on activity based costing) to standardize,count and compensate the provision of services in internalcustomersupplier relationships. In both cases, we are in thepresence of services that are not the subject of market-basedtransactions even though in the latter case they can be accountedfor as intra-firm transactions.

A second issue concerns the difficulty of defining productsand services independently from forms of economic coordina-tion in particular produceruser interactions. As Baldwin andClark (2003) imply, market-based transactions conform neatlyto the thinnest of interfaces between buyer and seller. Transfersthat score low on standardization, low on ease of counting andvaluing belong to the realm of hierarchies and relationalcontracting. By implication, services should be seen mainly astransfers and thus fall either within relational contracting orhierarchy. But this is a difficult argument to sustain. Transactionpoints do not emerge naturally where products cross thin

801ing Management 35 (2006) 797805buyersupplier interfaces. Many transactions are embedded incomplex buyersupplier relationships with multiple productand service exchanges. The task of making services objectified

rketand bounded entities that can be transacted is not necessarily aneasy one, but is one that is routinely achieved. Baldwin andClarke's implicit suggestion that services fall by default into thecategory of uncounted, unstandardized and uncompensatedtransfers should be tempered by Callon et al.'s (2002) notionthat specialized work is required to make products as well asservices tradable. Both cases require the objectification andstabilization of what it is that is being transacted.

6. Manufacturing moves into services

It was noted above that relationships between firms and,indeed, between firms and users, rather than being eitherproduct-based or service-based, typically involve complexmixtures of product and service elements. In this section weexamine two approaches to this theme: one, springing from theinteraction approach literature; and a second, more recenttheme, which suggests that there is a substantial shift amongstmanufacturing firms to move into service provision.

In an early example of studies of services in industrial markets,Cunningham and Roberts (1974) remarked that a service is eithercentral to a transaction or it is supplied in conjunction with aproduct. When a service is bundled with a product, it is eitheraddressed at reducing the workload of the buyer in making thepurchase, to reduce the uncertainty associated with the purchaseor to increase the usefulness and availability of the product.Hkansson, Johanson, and Wootz (1976) and Hkansson (1980)characterized buyersupplier relationships in terms of problem-solving and transfer, to denote the range of services involved infinding a solution to an identified need and to transfer the solutionto the customer's setting. The emphasis is on the interactionbetween two parties that co-participate in the development of thesolution and its transfer. Mattsson (1973) and Page andSiemplenski (1983) used the term systems selling and marketingto refer to integrated solutions of hardware and software, productsand services that can in principle be marketed separately.

The upshot of these early examples is that services are alwaysinvolved in complex buyersupplier relationships,whether they areaccounted for as a separate part of a transaction or they are bundledin with the sale of products. More recently, in a number ofmanufacturing sectors erstwhile manufacturers have been integrat-ing forward into systems assembly and supply and the provision ofvalue-added services. Prominent industrial firms such as IBM andGeneral Electric currently derive a significant share of theirrevenues and profits from services and their improved performancesince the mid 1990s is often attributed to a shift from a product to aservice focus.Manufacturing has increasingly been seen as an ever-narrower stage of a value-added chain where most profitopportunities reside upstream and downstream of actual manufac-turing or what Pavitt (2003), p. 88, called the skilled activitiesthat manufacturing firms undertake except manufacturing itself.

If it is the case that manufacturing firms are becoming moreinvolved in service activities, then Vandermerwe and Rada(1988) and Chase and Garvin (1989) were amongst the first to

802 L. Araujo, M. Spring / Industrial Maextol the virtues of such a move. Wise and Baumgartner (1999)note that in many industries the installed base of products hasbeen expanding steadily due to the accumulation of past sales aswell as longer product life spans. The proportion of new salesversus the installed base has pushed manufacturers downstreamaway from manufacturing new products and towards servicesrequired to operate and maintain the installed base. The strategicimplication of this shift is clear: manufacturers that ignoreopportunities for forward integration will see their profits erode,as more value flow downstream and away from manufacturing.

Davies (2003) argues that rather than just adding services toexisting products, firms are changing their strategies, occupyingnew positions in value chains and developing systemsintegration capabilities. The argument is that the trend towardsoutsourcing and vertical disintegration has led to the emergenceof a new type of specialist firm, the systems integrator. Systemsintegrators focus on the specialist capabilities required todevelop and design complex product architectures as well as thecapabilities required to interact with and coordinate a networkof external suppliers.

Contrary to Wise and Baumgartner's view that opportunitieslie downstream from manufacturing, Davies argues that manyservices are offered before and during the making of the product.Within these services, Davies (2003), p. 340, includes pre-bidnegotiations, the bid-to-contract phase, and the project imple-mentation phase including the detailed design, integration andtesting as well as handover to the customer. In the literature onsustainability, productservice systems are seen as a way tomove away from sales of material products to dematerializedservice provision often associated with changes in ownershipstructures e.g. renting or leasing rather than owningequipment (Mont, 2002; Roy, 2000).

The much heralded move of manufacturing into upstream anddownstream services may thus have little to do with new activitiesin value systems. Instead, we should be focusing on threephenomena that may explain this trend. First, the increasingdisaggregation of value systems into more complex patterns ofrelationships amongst specialized producers and service providersbrings about a growth in service provision often shifting internaltransfers into transactions in Baldwin and Clark's (2003) language.Secondly, the task of designing new services in the shape of newforms of access to maintained technical capacities provides firmswith opportunities to broaden their offering and increase revenuestreams often from the same customer base. For example, waterutilities use the technical capacities to maintain and repair theirinfrastructure to design plumbing insurance services for house-holds. Finally, the stream of product innovations often createstemporary disparities between producer and user capabilities thatcan be exploited by service providers aiming to bridge that gap.

As far as the first trend is concerned, as Pavitt (2003) remindsus, processes of technological convergence and vertical dis-integration have been at work for a while in a number of indus-tries. The end result is not just the disaggregation of corporatevalue chains but also the emergence of specialized firms providingproducts and services across a broad range of industries. Thedrivers for this process must be understood with reference to thedivision of labor and increasing returns. A recent example is

ing Management 35 (2006) 797805provided by the growth of contract manufacturing in electronics(Sturgeon, 2002). In this field, pressures of market volatility andincreased international competition coupled with growing

rketoutsourcing have led to a distinctive model of industrial organi-zation which Sturgeon (2002), p. 455, calls the modular produc-tion network because distinct breaks in the value chain tend toform at points where information regarding product specificationscan be highly formalized. [] The locus of these value chain breakpoints appear to be largely determined by technical factors,especially the open and de facto standards that determine theprotocol for the hand-off of codified specifications.

For Pavitt (2003) both modularity and advances in ICTs havecontributed to the institutional and spatial fragmentation of productdesign and manufacturing activities in a number of industries. Theend result is the fragmentation and migration of manufacturing tofar flung places but the retention of highly skilled services in highwage countries. These services comprise such diverse activities ashigh-tech machines processing information rather than materials,knowledge underlying manufacturing activities, capabilities fordesigning, integrating and supporting complex physical systemsand logistics operations. As Pavitt (2003), p. 88, remarks: The factthat some of these activities are defined as services oftenconfuses rather than clarifies.

The second major trend has to do with the increasing numberof firms that are disaggregating and reorganizing their structures,multiplying the number of services they provide through accessto new forms of socio-technical capacities (Vandermerwe,1990). Zenger and Hesterly (1997) provide an argument as tohow internal initiatives and innovation in performance mea-surement enhance a firm's capacity to circumscribe and isolateinternal activities. Recent innovations in performance measure-ment both financial and non-financial such as benchmarking,quality measures, and activity based costing provide firms withmore accurate, fine-grainedmeasures of subunit performance. Inturn, the increased accuracy of cost measurement at the subunitlevel, aids the internal disaggregation of firms and the generationof new sources of external services. For example, with accuratecost drivers in place, activity centers can be benchmarkedagainst external suppliers and highly efficient units may beencouraged or required to provide external services. In summary,the disaggregation of large firms and the infusion of market-based incentives into hierarchies are improving the ability offirms to isolate specific socio-technical capacities which can beused to provide internal as well as external services.

This example poses a further challenge to the modeladvanced by Baldwin and Clark (2003) regarding clear-cutdistinctions between transactions and transfers. In effect, thedisaggregation of large firms and the injection of market-basedincentives into hierarchies blur the distinctions betweentransactions and transfers as Zenger and Hesterly (1997),p. 216, note for many inter-unit exchanges, the formal legaldistinction has become the only difference between market andhierarchical governance.

Finally the proliferation of new products, as Langlois andCosgel (1999) argue, often provide entrepreneurial, service-based opportunities to provide a bridge between the capabilitiesof producers and users. Consultants may step in to help users

L. Araujo, M. Spring / Industrial Maspecify and seek solutions to newly-found needs as well asintegrate those solutions into their operations. Producers maybundle services with products in situations where the connec-tion of the capabilities embodied in the product and the contextof usage of the product are not immediately apparent to users.

In summary, the growth in services associated with manufac-turing should not only be associated with a simple logic ofsubstitution but also connected to the reorganization of theinstitutional structures of production. As we have argued, trendstowards the disaggregation of value chains and finer partitioningof activities, the fragmentation of corporate structures and theproliferation of innovations all provide entrepreneurial opportu-nities for new service provision.

7. Conclusions

This paper has revisited the nature of productservicedistinctions and attempted to shed new light on the growth ofservices associated with manufacturing. We concur withGadrey's (2000), p. 386, remark that it is unlikely we will everarrive at a definitive list of factors distinguishing products fromservices. Indeed, we go further and suggest that such quest isboth unproductive and distracting. Instead, we propose thatproducts and services should be regarded as different types ofintermediaries requiring stabilization and objectification to betransacted as Callon et al. (2002) have argued. The objectifi-cation of an entity does not require that it has to be inscribed intotangible materials even though many services rely on tangibleresources as Levitt (1972) pointed out. In summary, makingservices tradable requires the regulation of access to maintainedsocio-technical capacities which may involve a variety of modesof interaction between service providers and users. In somecases, relational and spatial proximity are required whereas inmany others access can take place without either.

If it is not straightforward to see how intangible services can bemade tradable, looking at products as anything other thanautonomous entities is equally counter-intuitive. Our suggestion,following Callon (1991), is that products constitute programs ofaction inscribed in physical, tangible materials. But it would bewrong to see products as simple embodiments of knowledge andvehicles for disembedding services from the vagaries ofproduceruser interactions. The production, circulation and useof products define particular networks and allocate roles toparticipants in those networks (Callon et al., 2002). Hill's (1977)suggestion that products should be regarded as entities whoseexistence is independent of their qualities or the relationships thatbrought them into being needs amending. The production,circulation and use of products cannot be separated from therange of services associated with each of those activities.

The managerial implications of a new perspective on productsand services can be seen at three levels. First, rather than attemptto focus on the management of product or service-based firms, weshould focus on products and services can be both complementsand substitutes for each other. The suggestion to focus onofferings of productservice combinations employed by Ford,Gadde, Hkansson, and Snehota (2003) is a helpful start toovercome the serviceproduct duality. The challenge for most

803ing Management 35 (2006) 797805firms is how to generate a variety of revenue streams from bothproduct and service transactions, as the recent literature on theexpanding the role of manufacturing suggests. In addition, we

rketsuggest that moves of manufacturing firms into services oftenentail significant investments in creating new forms of access tomaintained socio-technical capacities and new organizationalarrangements to support these services.

Secondly, the focus on the product or service-based firms hasdetracted from a closer examination of the different categories ofservices and the impact ofmigration into service-intensive offeringson firms' strategies and structures. In addition, we need a betterunderstanding of how to categorize services from a businessperspective. Gadrey's (2000) three service logics based on demandrationales and Axelsson and Wynstra's (2002) proposal forcategorizing business services provide helpful starting points fororganizing and managing firms with complex productserviceofferings. The transition from product to a combined productservice or solutions provider is also poorly understood (Miller,Hope, Eisenstat, Foote, & Galbraith, 2002). It often involves newforms of organization, new value propositions to customers, newways of making services tradable, novel pricing strategies andbusiness models. The challenge is not just to reorganizecorporations but also to find new ways to connect sets ofcapabilities within and across suppliers and customers' boundaries.

Lastly, the institutional approach outlined in this paperprovides a good opportunity to assess the notion of manufac-turing as the provision of a service throughout a product's lifecycle. For example, the quest for replacing the sale of productswith new modes of ownership and the provision of services overthe life-cycle of a product often requires a reassessment of whatneeds to be performed in-house or what should be outsourced.The improvement in the tradability of many services enablesfirms to employ fragmentation and offshoring strategies inservices that may have wider implications for the location andcoordination of activities than the preceding fragmentation ofmanufacturing (WIR, 2004). Although neither fragmentationnor outsourcing is a novel phenomenon, these developmentsmay lead in time to new patterns of location and coordination ofgeographically dispersed activities.

Acknowledgements

Earlier versions of this paper were presented at seminars inLancaster and the StockholmSchool of Economics, an IMPMarketStudies seminar organized by Damien McLoughlin (UniversityCollege Dublin) in April 2005, the EUROMA conference(Budapest) in June 2005 and the 21st IMPConference (Rotterdam)in September 2005. Comments and suggestions by participants atthese events, two anonymous referees and Keith Blois (Oxford) aregratefully acknowledged. The usual disclaimers apply.

References

Araujo, L., Dubois, A., & Gadde, L. -E. (1999). Managing interfaces withsuppliers. Industrial Marketing Management, 28(5), 497506.

Axelsson, B., & Wynstra, F. (2002). Buying business services. Chichester: JohnWiley.

Baldwin, C., & Clark, K. (2003), Where do transactions come from? A perspective

804 L. Araujo, M. Spring / Industrial Mafrom engineering design,Working PaperNo. 03-031,HarvardBusiness School.Bowen, J., & Ford, R. C. (2002). Managing service organizations: Does having

a thing make a difference? Journal of Management, 28(3), 447.Boyt, T., & Harvey, M. (1997). Classification of industrial services: A modelwith strategic implications. Industrial Marketing Management, 26(4),291300.

Callon, M. (1991). Techno-economic networks and irreversibility. In John Law(Ed.), A sociology of monsters: Essays on power, technology and dominationLondon: Routledge.

Callon, M. (2002). Writing and (re)writing devices as tools for managingcomplexity. In John Law, & Annemarie Mol (Eds.), Complexities. Socialstudies of knowledge practices. Durham, NC.: Duke University Press.

Callon, M., Meadel, C., & Rabeharisoa, V. (2002). The economy of qualities.Economy and Society, 2(31), 194217.

Callon, M., & Muniesa, F. (2005). Economic markets as calculative collectivedevices. Organization Studies, 26(8), 12291250.

Chase, R. B., & Garvin, D. A. (1989). The service factory. Harvard BusinessReview, 67(4), 6170.

Coase, R. H. (1992). The institutional structure of production. AmericanEconomic Review, 82(4), 713719.

Cunningham, M. T., & Roberts, D. A. (1974). The role of customer service inindustrial marketing. European Journal of Marketing, 8(1), 1529.

Davies, A. (2003). Integrated solutions: The changing business of systemsintegration. In Andrea Prencipe, Andrew Davies, & Mike Hobday(Eds.), The business of systems integration. Oxford: Oxford UniversityPress.

Delaunay, J. -C., & Gadrey, J. (1987). Les Enjeux de la Socit de Service. Paris:Presses de la Fondation Nationale des Sciences Politiques.

Demsetz, H. (1993). The theory of the firm revisited. In Oliver E. Williamson, &Sydney G. Winter (Eds.), The nature of the firm. Origins, evolution anddevelopment. New York: Oxford University Press.

Fisk, R. P., Brown, S. W., & Bittner, M. J. (1993). Tracking the evolution of theservices marketing literature. Journal of Retailing, 69(1), 61103.

Ford, D., Gadde, L. E., Hkansson, H., & Snehota, I. (2003). Managingbusiness relationships. Chichester: John Wiley.

Gadrey, J. (2000). The characterization of goods and services: An alternativeapproach. Review of Income and Wealth, 46(3), 369387.

Grabher, G. (2004). Learning in projects, remembering in networks?Communality, sociality, and connectivity in project ecologies. EuropeanUrban and Regional Studies, 11(2), 103123.

Grnroos, C. (1979). An applied theory for marketing industrial services. In-dustrial Marketing Management, 8(1), 4550.

Grnroos, C. (1998). Marketing services: The case of a missing product.Journal of Business and Industrial Marketing, 13(4/5), 322338.

Gummesson, E. (1978). Toward a theory of professional service marketing.Industrial Marketing Management, 7(2), 8995.

Hkansson, H. (1980). Marketing strategies in industrial markets: A frameworkapplied to a steel producer. European Journal of Marketing, 14(5/6), 365378.

Hkansson, H., Johanson, J., &Wootz, B. (1976). Influence tactics in buyersellerprocesses. Industrial Marketing Management, 5(6), 319332.

Hill, P. (1977). On goods and services. Review of Income and Wealth, 23(4),315338.

Hill, P. (1999). Tangibles, intangibles and services: A new taxonomy for theclassification of output. Canadian Journal of Economics, 32(2), 426447.

Hodgson, G. (1988). Economics and institutions: A manifesto for a moderninstitutional economics. Cambridge: Polity Press.

Johnston, R. (1999). Service operations management: Return to roots. Interna-tional Journal of Operations and Production Management, 19(2), 104124.

Kotler, P. (1980). Marketing management: Analysis, planning and control, 4thedition. Englewood Cliffs, NJ: Prentice-Hall.

Kotler, P. (2003). Marketing management, 11th edition. Upper Saddle River,NJ: Prentice-Hall.

Langlois, R. N. (2002). Modularity in technology and organization. Journal ofEconomic Behavior and Organization, 49(1), 1937.

Langlois, R. N., & Cosgel, M. M. (1999). The organization of consumption. InMarina Bianchi (Ed.), The active consumer. Novelty and surprise inconsumer choice. London: Routledge.

Levitt, T. (1972). Production line approach to service.Harvard Business Review,

ing Management 35 (2006) 79780550(5), 4153.Levitt, T. (1981). Marketing intangible products and product intangibles. Har-

vard Business Review, 59(3), 94103.

Loasby, B. J. (1998). The concept of capabilities. In Nicolai J. Foss, & Brian F.Loasby (Eds.), Economic organization, capabilities and co-ordination.Essays in honour of G.B. Richardson. London: Routledge.

Lovelock, C. (1983). Classifying services to gain strategic marketing insights.Journal of Marketing, 47(2), 920.

Lovelock, C., & Gummesson, E. (2004). Whither services marketing? In search of anew paradigm and fresh perspectives. Journal of Service Research, 7(1), 2041.

Mattsson, L. -G. (1973). Systems selling as a strategy in industrial markets.Industrial Marketing Management, 3(2), 107120.

Miller, D., Hope, Q., Eisenstat, R., Foote, N., &Galbraith, J. (2002). The problem ofsolutions: Balancing clients and capabilities. Business Horizons, 45(2), 312.

Mont, O. K. (2002). Clarifying the concept of productservice system. Journalof Cleaner Production, 10(3), 237245.

Page, A. L., & Siemplenski, M. (1983). Product systems marketing. IndustrialMarketing Management, 12(2), 8999.

Pavitt, K. (2003). Specialization and systems integration: Where manufactureand services still meet. In Andrea Prencipe, AndrewDavis, & Mike Hobday(Eds.), The business of systems integration.Oxford: Oxford University Press.

Penrose, E. (1959). The theory of the growth of the firm.Oxford: Basil Blackwell.Roy, R. (2000). Sustainable productservice systems. Futures, 32(34), 289299.Spring, M., & Dalrymple, J. F. (2000). Product customization and manufacturing

strategy. International Journal of Operations and ProductionManagement, 20(34), 441467.

Sturgeon, T. J. (2002). Modular production networks: A new American model ofindustrial organization. Industrial and Corporate Change, 11(1), 451496.

Sundbo, J. (1994). Modulization of service production and a thesis ofconvergence between service and manufacturing organizations. Scandina-vian Journal of Management, 10(3), 245266.

Vandermerwe, S. (1990). The market power is in the services: Because the valueis in the results. European Management Journal, 8(4), 464473.

Vandermerwe, S., & Rada, J. (1988). Servitization of business: Adding value byadding services. European Management Journal, 6(4), 314324.

Vargo, S. L., & Lusch, R. F. (2004a). The four service marketing myths:Remnants of a goods-based, manufacturing model. Journal of ServiceResearch, 6(4), 324335.

Vargo, S. L., & Lusch, R. F. (2004b). Evolving to a new dominant logic formarketing. Journal of Marketing, 68(1), 117.

Wise, R., & Baumgartner, P. (1999). Go downstream. The new profit imperativein manufacturing. Harvard Business Review, 77(5), 133141.

World Investment Report (2004). The shift towards services. New York andGeneva: United Nations Conference on Trade and Development (UNCTAD).

Zeithaml, V., Parasuraman, A., & Berry, L. L. (1985). Problems and strategies inservices marketing. Journal of Marketing, 49(1), 3346.

Zenger, T. R., & Hesterly, W. S. (1997). The disaggregation of corporations:Selective intervention, high-powered incentives, and molecular units. Or-ganization Science, 8(3), 209222.

Luis Araujo is Professor of Industrial Marketing at Lancaster UniversityManagement School. His research interests and publications lie in the field ofBusiness Marketing and Industrial Networks.

Martin Spring is Senior Lecturer in Operations Management at LancasterUniversity Management School. He is a former co-editor of the InternationalJournal of Operations and Production Management and his research interestslie in the areas of power, trust and service operations management in businessmarkets.

805L. Araujo, M. Spring / Industrial Marketing Management 35 (2006) 797805

Services, products, and the institutional structure of productionIntroductionThe serviceproduct distinctionAn institutional perspective on service definitionThree demand rationales for servicesThe tradability of servicesTransfers and transactionsProducts/services and goods

Manufacturing moves into servicesConclusionsAcknowledgementsReferences