COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 29

BY ERIC VAN HECK ANDPETER VERVEST

Realizing scenarios in which business is conductedthrough a rapidly formed network with anyone, anywhere,anytime regardless of different computer systems and business processes.

NETWORKWINS

I l l u s t r a t i o n b y KEN ORVIDAS

Once merely a dream, dig-ital networks are now arapidly maturing reality.These digital networkscan expose unexpected

behavioral properties of the individualactors. Combined as a swarm, networkedbusinesses are able to produce exceptionalor “smart” results they were not previouslycapable of generating. Companies makedifferent linkages, combine different capa-bilities from many different parties, aremore agile, and move positions faster.What are these “smart” business networksand why are they important? What shouldchief information officers and IT profes-sionals do to help their companies succeedin a networked world?

BEING SMART IN THE BUSINESS NETWORK

In less than 10 years, Amazon movedfrom electronic book retailing to become

HOW THE

SMART BUSINESS NETWORKS:

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 3130 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

the world’s leading “e-tailer.” Without stores andwith limited inventory, Amazon possibly has moreinformation on retail goods and their buyers andsellers than many other businesses. Amazon offersa business platform for the traditional retailer to“make a market” [9]. Within the Amazon businessnetworks, the retailer can:

• Facilitate search by buyers and sellers;• Discover and help set pricing and other transac-

tion conditions;• Manage and coordi-

nate the logisticalprocesses for transferof the physical or digi-tal goods;

• Settle payments andarrange fund transfer;and

• Authenticate the qualityof the goods sold andverify the credibility ofbuyers and sellers.

Thousands of elec-tronic retailers joinAmazon every monthfor all or some of thesefunctions. At the same time many leave, or arerebuked by the Amazon system. Amazon facilitatesproduct representation, regulatory compliance,risk management, and conflict resolution; it hasquickly established a reputation for trustworthytransactions. EBay, with over 222 million regis-tered buyers and sellers, has done the same for auc-tions. At the end of 2006, Skype had attracted 171million registered users. In less than four years,Skype has surpassed anything traditional telephoneservice providers have ever achieved.

These companies offer platforms on which userscan freely move and interact as long as the platformprovider allows them. These platforms show a strongnetwork effect: the more users, the more useful thenetwork becomes, the more difficult it becomes toswitch, and the less likely the user will move toanother network. Albert László Barabási [1] recog-nized this by analyzing Internet traffic. He demon-strated that the Net is not democratic, and thenumber of links per node follows a so-called powerlaw distribution. A few nodes have many links whilemany more other nodes have very few links. Thenode with many links attracts nodes with fewer linksfaster than the lesser-connected nodes. As the big getbigger, what options do the smaller actors have?

Consider the example of Kenny’s Bookshop andArt Gallery, a family-run small business in Galway,Ireland (www.kennys.ie). It sources valuable sec-ondhand books and sells them to interested collec-tors [5]. Established in 1940, Kenny’s beganfocusing on its bookshop customers and then, inthe 1970s, by mailing paper catalogs to librariesoverseas. In 2003, Kenny’s linked its library man-agement system to the Online Computer LibraryCenter (OCLC), a not-for-profit organization that

owns the largest databaseof bibliographic recordsin the world. Kenny’swas the first business tohave a commercialarrangement withOCLC, which allowed itto provide a full elec-tronic catalog on virtu-ally any secondhandbook highly efficientlyand faster than anyoneelse. Kenny’s reapedinstant financial rewardssince a cataloged second-

hand book is usually valued four times higher thanan uncataloged item. Through Bookrouter.com,Kenny’s published its rapidly growing stock ofnow-cataloged secondhand books on multipleInternet sites such as Amazon, Alibris, and Bib-liodirect, using their, or other logistics providers,physical delivery capabilities. In 2006, Kenny’sbookshop went completely online and their shopin Galway is now just hosting the art gallery.

Kenny’s could have remained a traditional book-shop with others—Amazon and eBay—capturingtheir business. They did the reverse. They becamesmart in their business network by capturing avaluable position and leveraging that positionacross as many links as they could.

Capturing and leveraging a position in a busi-ness network does not mean one must own, orcontrol, the platforms on which those networksrun. For example, TheBigWord company(www.thebigword.com) is able to serve the diversetranslation needs of large companies worldwide bysharing a “translation memory” across a network ofits clients and thousands of local “mother tongue”linguists [7]. It responds almost instantly to trans-lation needs—such as for publishing on Web sitesin many different languages—by posting the workto targeted groups of qualified translators, dividingand allocating the work, and managing the processin a way that is fully transparent to the client.

VanHeck table 1 (6/07)

Products and services

Value creation

Coordination and control

Information sharing

Infrastructure

Relative complex, bundled,and fast delivered products and services

Demand networks with quick connect and disconnect relationships

Network orchestration with distributed control and decision making

Information sharing over and with network partners

Network platform with networked business operating system

Relative simple, unbundled,and slowly delivered products and services

Supply chains with long term connected relationships

Hierarchical and central control and decision making

Information sharing with direct business partners

Actor platforms with information silos and systems

Characteristics New BusinessNetwork Approach

Traditional BusinessNetwork Approach

Table 1. Characteristics of traditional and newbusiness network approaches.

Table 1. Characteristics of traditional and new business

network approaches.

Operating in Spain, France, the U.K., and Portugal,Grupo Multiasistencia coordinates home insuranceclaims and repairs serving approximately nine million

customers. Founded in 1983, it manages small-trade profes-sionals and has evolved to become a service coordinator forover 100 large corporations including insurance companies,banks, department stores, andother retail chains. Europeanend customers and corporateclients are handled by itsMadrid-based control center.Its 375 Customer Service Repre-sentatives (CSR) receiverequests for home repairs byphone, email, or the Internetand deploy and control jobs toa network of 11,000 trade pro-fessionals who complete therepairs. The figure here illus-trates the main actors in thebusiness network.

After Multiasistencia’s suc-cess in managing collectivesand attracting corporate clients,it experienced in 2000 decliningquality resulting from the non-scalability of operating all theircommunication via the tele-phone. In that year the new CEOand CIO started to reengineerusing the newest informationand communication technolo-gies [1]. The result has been a Networked Business OperatingSystem (BOS) based on their call center, Internet, Web services,and mobile systems.

The network coordination is highly automated. On arepair order no one needs to intervene unless an exceptionoccurs. The BOS tracks more than 100 variables to assurequality standards and timely execution. With this automatedcoordination, productivity increased to 49.6% and the num-ber of errors dropped dramatically. The firm has automatedalmost all human communication except for incoming calls.Therefore the CSRs do not manage phone calls, they monitorthe processes to manage exceptions to ensure quality. Withthe intense use of ICT, particularly Web services, connectedand disconnected processes have been integrated and stan-dardized.

Rather than technological, a major challenge in implement-ing the BOS has been ensuring trust with the agents given thehigher levels of standardized processes and transparent con-trol. The BOS also has the capacity to adapt to environmentalchanges by defining behavioral limits as automatic responsesor human-driven actions for exception management. The BOSalso accommodates idiosyncratic corporate customer demandsas turnkey services. All services are audited through SLAs withall agents in the network. Multiasistencia has also boostedvalue and innovation by collaborating with its corporate clients[2, 3]. Approximately 80% of new software development hasbeen done in collaboration with its large customers. For exam-ple, in 2005, it created a new desktop telesurvey adjustmentservice. When the repair assessment exceeds a financial

threshold, loss adjusters can carry out a “desktop audit” on theInternet; digital photographs are taken on-site by trade profes-sionals, sent to the CSR, and reviewed for approval in realtime.

In 2006, Multiasistencia and a European insurer defined anew service to inform end customers about the process when-

ever a claim was pre-sented. The applicationsends Short MessageServices (SMS) messagesthrough cell phones tothe end customers. Asone of the insurer’s exec-utive explains, “Informa-tion management hasbeen the key to increas-ing loyalty by 20% withour customer base when-ever they use the ser-vice.” Trade professionalsalso benefit. They usedigitalized signatures andelectronic invoicing,thereby improving cashcollection and reducingadministrative costs by40%.

Besides the productiv-ity gains, this smart business network has also standardizedthe highly fragmented and heterogeneous household repairssector: applying predefined prices, transparent conditions, ser-vice performance and quality guarantees, ensuring timelyresponse (within 24 hours; three hours for emergencies) andwarranting the work for six months.

There are two critical challenges for Multiasistencia to sup-port the business model and the business relationships. Thefirst challenge is the interfacing of the BOS with human agents.The second challenge is negotiating and finding a balancebetween standardization and innovative environmentalresponses.

References1. Busquets, J. Multiasistencia on the Internet (A), (B), and (C).

Teaching Cases, DVD Multimedia and Teaching Note, EuropeanCase Clearing House, 2006; www.ecch.com.

2. Nohria, N. and Ghoshal, S. The Differentiated Network. Jossey-Bass, 1997.

3. Von Hippel, E. Democratizing Innovation. MIT Press, Cambridge,MA, 2005.

Javier Busquets (xavier.busquets@esade.edu) is head of theDepartment of Information Systems at ESADE, University RamonLlull in Barcelona, Spain.Juan Rodon (joan.rodon@esade.edu) is an assistant professorof the Department of Information Systems at ESADE, UniversityRamon Llull in Barcelona, Spain.Jonathan Wareham (wareham@acm.org) is an associate professor in the Department of Information Systems at ESADE, University of Ramon Llull in Barcelona, Spain.

© 2007 ACM 0001-0782/07/0600 $5.00

c

MULTIASISTENCIA: THE NETWORK ORCHESTRATORBy Javier Busquets, Juan Rodón, and Jonathan Wareham

VanHeck sidebar (Busquets) fig. (6/07)

Sidebar (Busquets ) figure caption: Multiasistenciabusiness network actors.

1.a Request for arepair service

2. Transfer ofa request for arepair service

6. Invoicing

7. Invoicing,Information on

repairsand claims

8. Surveysatisfaction

4. provision of the service5. Invoicing on behalf of eRepairs(when repair not included under policy)

1.b Requestfor a repair

service

3. Repairassignmentand order

Corporate Clients

MultiasistenciaEnd Customers

TradeProfessionals

Multiasistencia business network actors.

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 3130 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

the world’s leading “e-tailer.” Without stores andwith limited inventory, Amazon possibly has moreinformation on retail goods and their buyers andsellers than many other businesses. Amazon offersa business platform for the traditional retailer to“make a market” [9]. Within the Amazon businessnetworks, the retailer can:

• Facilitate search by buyers and sellers;• Discover and help set pricing and other transac-

tion conditions;• Manage and coordi-

nate the logisticalprocesses for transferof the physical or digi-tal goods;

• Settle payments andarrange fund transfer;and

• Authenticate the qualityof the goods sold andverify the credibility ofbuyers and sellers.

Thousands of elec-tronic retailers joinAmazon every monthfor all or some of thesefunctions. At the same time many leave, or arerebuked by the Amazon system. Amazon facilitatesproduct representation, regulatory compliance,risk management, and conflict resolution; it hasquickly established a reputation for trustworthytransactions. EBay, with over 222 million regis-tered buyers and sellers, has done the same for auc-tions. At the end of 2006, Skype had attracted 171million registered users. In less than four years,Skype has surpassed anything traditional telephoneservice providers have ever achieved.

These companies offer platforms on which userscan freely move and interact as long as the platformprovider allows them. These platforms show a strongnetwork effect: the more users, the more useful thenetwork becomes, the more difficult it becomes toswitch, and the less likely the user will move toanother network. Albert László Barabási [1] recog-nized this by analyzing Internet traffic. He demon-strated that the Net is not democratic, and thenumber of links per node follows a so-called powerlaw distribution. A few nodes have many links whilemany more other nodes have very few links. Thenode with many links attracts nodes with fewer linksfaster than the lesser-connected nodes. As the big getbigger, what options do the smaller actors have?

Consider the example of Kenny’s Bookshop andArt Gallery, a family-run small business in Galway,Ireland (www.kennys.ie). It sources valuable sec-ondhand books and sells them to interested collec-tors [5]. Established in 1940, Kenny’s beganfocusing on its bookshop customers and then, inthe 1970s, by mailing paper catalogs to librariesoverseas. In 2003, Kenny’s linked its library man-agement system to the Online Computer LibraryCenter (OCLC), a not-for-profit organization that

owns the largest databaseof bibliographic recordsin the world. Kenny’swas the first business tohave a commercialarrangement withOCLC, which allowed itto provide a full elec-tronic catalog on virtu-ally any secondhandbook highly efficientlyand faster than anyoneelse. Kenny’s reapedinstant financial rewardssince a cataloged second-

hand book is usually valued four times higher thanan uncataloged item. Through Bookrouter.com,Kenny’s published its rapidly growing stock ofnow-cataloged secondhand books on multipleInternet sites such as Amazon, Alibris, and Bib-liodirect, using their, or other logistics providers,physical delivery capabilities. In 2006, Kenny’sbookshop went completely online and their shopin Galway is now just hosting the art gallery.

Kenny’s could have remained a traditional book-shop with others—Amazon and eBay—capturingtheir business. They did the reverse. They becamesmart in their business network by capturing avaluable position and leveraging that positionacross as many links as they could.

Capturing and leveraging a position in a busi-ness network does not mean one must own, orcontrol, the platforms on which those networksrun. For example, TheBigWord company(www.thebigword.com) is able to serve the diversetranslation needs of large companies worldwide bysharing a “translation memory” across a network ofits clients and thousands of local “mother tongue”linguists [7]. It responds almost instantly to trans-lation needs—such as for publishing on Web sitesin many different languages—by posting the workto targeted groups of qualified translators, dividingand allocating the work, and managing the processin a way that is fully transparent to the client.

VanHeck table 1 (6/07)

Products and services

Value creation

Coordination and control

Information sharing

Infrastructure

Relative complex, bundled,and fast delivered products and services

Demand networks with quick connect and disconnect relationships

Network orchestration with distributed control and decision making

Information sharing over and with network partners

Network platform with networked business operating system

Relative simple, unbundled,and slowly delivered products and services

Supply chains with long term connected relationships

Hierarchical and central control and decision making

Information sharing with direct business partners

Actor platforms with information silos and systems

Characteristics New BusinessNetwork Approach

Traditional BusinessNetwork Approach

Table 1. Characteristics of traditional and newbusiness network approaches.

Table 1. Characteristics of traditional and new business

network approaches.

Operating in Spain, France, the U.K., and Portugal,Grupo Multiasistencia coordinates home insuranceclaims and repairs serving approximately nine million

customers. Founded in 1983, it manages small-trade profes-sionals and has evolved to become a service coordinator forover 100 large corporations including insurance companies,banks, department stores, andother retail chains. Europeanend customers and corporateclients are handled by itsMadrid-based control center.Its 375 Customer Service Repre-sentatives (CSR) receiverequests for home repairs byphone, email, or the Internetand deploy and control jobs toa network of 11,000 trade pro-fessionals who complete therepairs. The figure here illus-trates the main actors in thebusiness network.

After Multiasistencia’s suc-cess in managing collectivesand attracting corporate clients,it experienced in 2000 decliningquality resulting from the non-scalability of operating all theircommunication via the tele-phone. In that year the new CEOand CIO started to reengineerusing the newest informationand communication technolo-gies [1]. The result has been a Networked Business OperatingSystem (BOS) based on their call center, Internet, Web services,and mobile systems.

The network coordination is highly automated. On arepair order no one needs to intervene unless an exceptionoccurs. The BOS tracks more than 100 variables to assurequality standards and timely execution. With this automatedcoordination, productivity increased to 49.6% and the num-ber of errors dropped dramatically. The firm has automatedalmost all human communication except for incoming calls.Therefore the CSRs do not manage phone calls, they monitorthe processes to manage exceptions to ensure quality. Withthe intense use of ICT, particularly Web services, connectedand disconnected processes have been integrated and stan-dardized.

Rather than technological, a major challenge in implement-ing the BOS has been ensuring trust with the agents given thehigher levels of standardized processes and transparent con-trol. The BOS also has the capacity to adapt to environmentalchanges by defining behavioral limits as automatic responsesor human-driven actions for exception management. The BOSalso accommodates idiosyncratic corporate customer demandsas turnkey services. All services are audited through SLAs withall agents in the network. Multiasistencia has also boostedvalue and innovation by collaborating with its corporate clients[2, 3]. Approximately 80% of new software development hasbeen done in collaboration with its large customers. For exam-ple, in 2005, it created a new desktop telesurvey adjustmentservice. When the repair assessment exceeds a financial

threshold, loss adjusters can carry out a “desktop audit” on theInternet; digital photographs are taken on-site by trade profes-sionals, sent to the CSR, and reviewed for approval in realtime.

In 2006, Multiasistencia and a European insurer defined anew service to inform end customers about the process when-

ever a claim was pre-sented. The applicationsends Short MessageServices (SMS) messagesthrough cell phones tothe end customers. Asone of the insurer’s exec-utive explains, “Informa-tion management hasbeen the key to increas-ing loyalty by 20% withour customer base when-ever they use the ser-vice.” Trade professionalsalso benefit. They usedigitalized signatures andelectronic invoicing,thereby improving cashcollection and reducingadministrative costs by40%.

Besides the productiv-ity gains, this smart business network has also standardizedthe highly fragmented and heterogeneous household repairssector: applying predefined prices, transparent conditions, ser-vice performance and quality guarantees, ensuring timelyresponse (within 24 hours; three hours for emergencies) andwarranting the work for six months.

There are two critical challenges for Multiasistencia to sup-port the business model and the business relationships. Thefirst challenge is the interfacing of the BOS with human agents.The second challenge is negotiating and finding a balancebetween standardization and innovative environmentalresponses.

References1. Busquets, J. Multiasistencia on the Internet (A), (B), and (C).

Teaching Cases, DVD Multimedia and Teaching Note, EuropeanCase Clearing House, 2006; www.ecch.com.

2. Nohria, N. and Ghoshal, S. The Differentiated Network. Jossey-Bass, 1997.

3. Von Hippel, E. Democratizing Innovation. MIT Press, Cambridge,MA, 2005.

Javier Busquets (xavier.busquets@esade.edu) is head of theDepartment of Information Systems at ESADE, University RamonLlull in Barcelona, Spain.Juan Rodon (joan.rodon@esade.edu) is an assistant professorof the Department of Information Systems at ESADE, UniversityRamon Llull in Barcelona, Spain.Jonathan Wareham (wareham@acm.org) is an associate professor in the Department of Information Systems at ESADE, University of Ramon Llull in Barcelona, Spain.

© 2007 ACM 0001-0782/07/0600 $5.00

c

MULTIASISTENCIA: THE NETWORK ORCHESTRATORBy Javier Busquets, Juan Rodón, and Jonathan Wareham

VanHeck sidebar (Busquets) fig. (6/07)

Sidebar (Busquets ) figure caption: Multiasistenciabusiness network actors.

1.a Request for arepair service

2. Transfer ofa request for arepair service

6. Invoicing

7. Invoicing,Information on

repairsand claims

8. Surveysatisfaction

4. provision of the service5. Invoicing on behalf of eRepairs(when repair not included under policy)

1.b Requestfor a repair

service

3. Repairassignmentand order

Corporate Clients

MultiasistenciaEnd Customers

TradeProfessionals

Multiasistencia business network actors.

intelligence studies collective behavior in self-orga-nizing systems populated by simple individualsinteracting locally with one another and with theirenvironment without centralized control. However,in many cases, despite being unpredictable, suchswarms are able to exhibit impressive capabilitiesfor problem solving to, for example, seek food orrespond to an unforeseen problem.

While these studies provide indicators for net-work dynamics—formation, change, decay—andfor the ways in which the individual intelligence ofthe network actors is combined, the research insocial network analysis has made a significant con-tribution to a more profound understanding ofnetwork behavior. Social network researchers takeinto account the social relationships and ties ofindividuals and therefore the structure of the net-work. Building on social network analysis usingcomplex systems theory Dan Braha and YaneerBar-Yam [3] examined the statistical properties oflarge-scale product development information net-works for vehicle design. They find that such net-works have properties (sparseness, small world,scaling regimes) like those of other biological,social, and technological networks. They demon-strate that the distribution of incoming communi-cation links always has a cut-off—their numbersare restricted—while the distribution of outgoingcommunication links is considered scale-free,meaning some nodes act as highly connected hubs.This would be consistent with Herbert Simon’sbounded rationality argument that rational agentsexperience limits in formulating and solving com-plex problems and in processing (receiving, storing,retrieving, transmitting) information.

Braha and Bar-Yam found it seems easier to trans-mit information than to process information. Likeindividual human beings, a group of people—ornetwork of nodes—is limited by an inability todigest an intense input of data. It seems that smart-ness could be related to the capability to organize theinformation flows within the business network aswell as to the topological structure of the network.

Other researchers have shown the attractivenessand importance of certain positions in the network,that is, those nodes that are dominant and thosethat take subservient roles. For example, in 1992,Ronald Burt identified “bridging positions” wherethe network participants link through a focal actorwho holds the bridge [4]. This structure bringsinformation and control benefits (a central player)but also encourages the dependent actors to findalternative routes, for example, to disintermediatethe bridger.

As the analysis is applied to larger and more com-plex networks, more advanced ways are required toanalyze the structure of the network. In the sidebar“Network Horizon and Obtaining a Favorable Net-work Position,” van Liere and Koppius use socialnetwork analysis and simulation techniques toexplore the concept of the network horizon: thenumber of nodes an actor can “see” from a specificposition in the network [12]. With a larger networkhorizon a company can take a more advantageousnetwork position depending on the distribution ofthe network horizons across all actors and up to acertain saturation point. The results indicate theexpansion of the network horizon will soon be a crit-ical success factor for companies.

Most network scientists analyze the structureand dynamics of business networks independent ofthe technologies that enable the networks to per-form. Instead, researchers tend to concentrate onwhat makes the network effective, the linked rela-tionships between the actors, and how their intelli-gence is combined to reach the network’s goals.Digital technologies play a fundamental role intoday’s networks. They have facilitated improve-ments and fundamental changes in the ways inwhich organizations and individuals interact andcombine as well as revealing unexpected capabili-ties that create new markets and opportunities.One need only consider the rapid rise in digitalsocial networks and massive online computergames such as Second Life. These are exhibitingcapabilities that seem well beyond those of existingbusiness networks. The next critical step is todevelop a comprehensive understanding of theexpected smartness of the business network.

WHAT MAKES A SMART BUSINESS NETWORK?The key characteristics of a smart business networkare that it has the ability to “rapidly pick, plug, andplay” business processes to configure rapidly to meeta specific objective, for example, to react to a cus-tomer order or an unexpected situation (such asdealing with emergencies) [11]. One might regard asmart business network as an expectant web of par-ticipants ready to jump into action (pick) and com-bine rapidly (plug) to meet the requirements of aspecific situation (play). On completion, the partic-ipants are dispersed to “rest” while, perhaps, beingactive in other business networks or more traditionalsupply chains.

This combination of pick, plug, play, and dis-perse means the fundamental organizing capabilitiesfor a smart business network are: the ability toquickly connect and disconnect with an actor; the

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 33

TheBigWord rewards the translators by paying bythe number of words they translate and by provid-ing their administration and payment services.TheBigWord example demonstrates these businessnetworks can respond with much more agilitytogether than acting as an individual company. Butwhat are other characteristics and what are the rea-sons why these new forms are starting to be devel-oped and implemented now?

A NEW BUSINESS NETWORK APPROACH

Organizations are moving, or must move, fromtoday’s relatively stable and slow-moving businessnetworks to an open digital platform where busi-ness is conducted across a rapidly formed networkwith anyone, anywhere, anytime despite differentbusiness processes and computer systems. Table 1provides an overview of the characteristics of thetraditional and new business network approaches.The disadvantages and associated costs of the moretraditional approaches are caused by the inabilityto provide relatively complex, bundled, andquickly delivered products and services. Thepotential of the new business network approach isto create these types of products and services withthe help of combining business network insightswith telecommunication capabilities.

The business is no longer a self-contained orga-nization working together with closely coupledpartners: it is a participant in a number of net-works where it may lead or act together with oth-ers. The network includes additional layers ofmeaning—from the ICT infrastructures to theinteractions between businesses and individuals.Rather than viewing the business as a sequentialchain of events (a value chain), actors in a smartbusiness network seek linkages that are novel anddifferent to create remarkable, “better than usual”results. “Smart” has a connotation with fashionableand distinguished but can also be somewhat short-lived. What is smart today will be considered com-mon tomorrow. Smart is therefore a relative ratherthan an absolute term. Smartness means the net-work of cooperating businesses can create betterresults than other, less smart, business networks orother forms of business arrangement. To be smartin business is to be smarter than the competitorsjust as an athlete who is considered fast meansfaster than the other competitors.

Another way the new business networkapproach distinguishes itself is the way the networkis orchestrated. In the sidebar “Multiasistencia:The Network Orchestrator,” Busquets, Rodón, andWareham introduce the Spanish Grupo Multiasis-

tencia. They show how the smart business networkapproach with embedded business processes leadsto substantial business advantages, demonstratingthe importance of information sharing in the busi-ness network and the design and organizationaldynamics of the infrastructure.

The pivotal question of smart business net-works concerns the relationship between the strat-egy and structure of the business network on onehand and the underlying infrastructure on theother. As new technologies, such as RFID, allownetworks of organizations almost complete insightinto where their people, materials, suppliers andcustomers are at any point in time, the organiza-tions are able to organize differently. But if allother players in the network space have that sameinsight, the result of the interactions may not becompetitive. Therefore, a first critical step is todevelop a profound understanding about the func-tioning of the business network.1

UNDERSTANDING THE NETWORK

If two cars drive on the highway with enough dis-tance between them they have no relationship otherthan to share the same roadway. If, however, thesesame cars get very close, they start behaving differ-ently. If the first car brakes, the second car willbrake—but with a delay. If the first car accelerates,the second car will also speed up; again, with adelay. The drivers of these two cars may not noticemuch more than the distance between them. Butfrom a helicopter their behavior will appear as awave moving along the flow of traffic. Each partic-ipant does not see the behavior of the network butresponds to the local situation with his or her dri-ving logic. The impact of the individual driver’sactions in response to their specific situation andthe road rules they follow creates a collective net-work behavior not seen nor understood by the indi-viduals. Each driver acts on self-organizing drivinglogic according to the driving rules of the network.

The study of networked behavior beyond thefamiliar territories of business and ICT networks—those of social interactions, ant colonies, bees, andother biological systems—reveals attributes andcharacteristics that can be applied to the design ofsmart business networks. The behavior of the indi-vidual drivers, as described in the preceding exam-ple, demonstrates swarm intelligence: theemergence of seemingly intelligent or, perhaps,smart, behavior from many individuals [2]. Swarm

32 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

1See “Decision Making in Very Large Networks,” by Peter J. Denning and Rick Hayes-

Roth in the November 2006 issue of Communications for more detailed information.

intelligence studies collective behavior in self-orga-nizing systems populated by simple individualsinteracting locally with one another and with theirenvironment without centralized control. However,in many cases, despite being unpredictable, suchswarms are able to exhibit impressive capabilitiesfor problem solving to, for example, seek food orrespond to an unforeseen problem.

While these studies provide indicators for net-work dynamics—formation, change, decay—andfor the ways in which the individual intelligence ofthe network actors is combined, the research insocial network analysis has made a significant con-tribution to a more profound understanding ofnetwork behavior. Social network researchers takeinto account the social relationships and ties ofindividuals and therefore the structure of the net-work. Building on social network analysis usingcomplex systems theory Dan Braha and YaneerBar-Yam [3] examined the statistical properties oflarge-scale product development information net-works for vehicle design. They find that such net-works have properties (sparseness, small world,scaling regimes) like those of other biological,social, and technological networks. They demon-strate that the distribution of incoming communi-cation links always has a cut-off—their numbersare restricted—while the distribution of outgoingcommunication links is considered scale-free,meaning some nodes act as highly connected hubs.This would be consistent with Herbert Simon’sbounded rationality argument that rational agentsexperience limits in formulating and solving com-plex problems and in processing (receiving, storing,retrieving, transmitting) information.

Braha and Bar-Yam found it seems easier to trans-mit information than to process information. Likeindividual human beings, a group of people—ornetwork of nodes—is limited by an inability todigest an intense input of data. It seems that smart-ness could be related to the capability to organize theinformation flows within the business network aswell as to the topological structure of the network.

Other researchers have shown the attractivenessand importance of certain positions in the network,that is, those nodes that are dominant and thosethat take subservient roles. For example, in 1992,Ronald Burt identified “bridging positions” wherethe network participants link through a focal actorwho holds the bridge [4]. This structure bringsinformation and control benefits (a central player)but also encourages the dependent actors to findalternative routes, for example, to disintermediatethe bridger.

As the analysis is applied to larger and more com-plex networks, more advanced ways are required toanalyze the structure of the network. In the sidebar“Network Horizon and Obtaining a Favorable Net-work Position,” van Liere and Koppius use socialnetwork analysis and simulation techniques toexplore the concept of the network horizon: thenumber of nodes an actor can “see” from a specificposition in the network [12]. With a larger networkhorizon a company can take a more advantageousnetwork position depending on the distribution ofthe network horizons across all actors and up to acertain saturation point. The results indicate theexpansion of the network horizon will soon be a crit-ical success factor for companies.

Most network scientists analyze the structureand dynamics of business networks independent ofthe technologies that enable the networks to per-form. Instead, researchers tend to concentrate onwhat makes the network effective, the linked rela-tionships between the actors, and how their intelli-gence is combined to reach the network’s goals.Digital technologies play a fundamental role intoday’s networks. They have facilitated improve-ments and fundamental changes in the ways inwhich organizations and individuals interact andcombine as well as revealing unexpected capabili-ties that create new markets and opportunities.One need only consider the rapid rise in digitalsocial networks and massive online computergames such as Second Life. These are exhibitingcapabilities that seem well beyond those of existingbusiness networks. The next critical step is todevelop a comprehensive understanding of theexpected smartness of the business network.

WHAT MAKES A SMART BUSINESS NETWORK?The key characteristics of a smart business networkare that it has the ability to “rapidly pick, plug, andplay” business processes to configure rapidly to meeta specific objective, for example, to react to a cus-tomer order or an unexpected situation (such asdealing with emergencies) [11]. One might regard asmart business network as an expectant web of par-ticipants ready to jump into action (pick) and com-bine rapidly (plug) to meet the requirements of aspecific situation (play). On completion, the partic-ipants are dispersed to “rest” while, perhaps, beingactive in other business networks or more traditionalsupply chains.

This combination of pick, plug, play, and dis-perse means the fundamental organizing capabilitiesfor a smart business network are: the ability toquickly connect and disconnect with an actor; the

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 33

TheBigWord rewards the translators by paying bythe number of words they translate and by provid-ing their administration and payment services.TheBigWord example demonstrates these businessnetworks can respond with much more agilitytogether than acting as an individual company. Butwhat are other characteristics and what are the rea-sons why these new forms are starting to be devel-oped and implemented now?

A NEW BUSINESS NETWORK APPROACH

Organizations are moving, or must move, fromtoday’s relatively stable and slow-moving businessnetworks to an open digital platform where busi-ness is conducted across a rapidly formed networkwith anyone, anywhere, anytime despite differentbusiness processes and computer systems. Table 1provides an overview of the characteristics of thetraditional and new business network approaches.The disadvantages and associated costs of the moretraditional approaches are caused by the inabilityto provide relatively complex, bundled, andquickly delivered products and services. Thepotential of the new business network approach isto create these types of products and services withthe help of combining business network insightswith telecommunication capabilities.

The business is no longer a self-contained orga-nization working together with closely coupledpartners: it is a participant in a number of net-works where it may lead or act together with oth-ers. The network includes additional layers ofmeaning—from the ICT infrastructures to theinteractions between businesses and individuals.Rather than viewing the business as a sequentialchain of events (a value chain), actors in a smartbusiness network seek linkages that are novel anddifferent to create remarkable, “better than usual”results. “Smart” has a connotation with fashionableand distinguished but can also be somewhat short-lived. What is smart today will be considered com-mon tomorrow. Smart is therefore a relative ratherthan an absolute term. Smartness means the net-work of cooperating businesses can create betterresults than other, less smart, business networks orother forms of business arrangement. To be smartin business is to be smarter than the competitorsjust as an athlete who is considered fast meansfaster than the other competitors.

Another way the new business networkapproach distinguishes itself is the way the networkis orchestrated. In the sidebar “Multiasistencia:The Network Orchestrator,” Busquets, Rodón, andWareham introduce the Spanish Grupo Multiasis-

tencia. They show how the smart business networkapproach with embedded business processes leadsto substantial business advantages, demonstratingthe importance of information sharing in the busi-ness network and the design and organizationaldynamics of the infrastructure.

The pivotal question of smart business net-works concerns the relationship between the strat-egy and structure of the business network on onehand and the underlying infrastructure on theother. As new technologies, such as RFID, allownetworks of organizations almost complete insightinto where their people, materials, suppliers andcustomers are at any point in time, the organiza-tions are able to organize differently. But if allother players in the network space have that sameinsight, the result of the interactions may not becompetitive. Therefore, a first critical step is todevelop a profound understanding about the func-tioning of the business network.1

UNDERSTANDING THE NETWORK

If two cars drive on the highway with enough dis-tance between them they have no relationship otherthan to share the same roadway. If, however, thesesame cars get very close, they start behaving differ-ently. If the first car brakes, the second car willbrake—but with a delay. If the first car accelerates,the second car will also speed up; again, with adelay. The drivers of these two cars may not noticemuch more than the distance between them. Butfrom a helicopter their behavior will appear as awave moving along the flow of traffic. Each partic-ipant does not see the behavior of the network butresponds to the local situation with his or her dri-ving logic. The impact of the individual driver’sactions in response to their specific situation andthe road rules they follow creates a collective net-work behavior not seen nor understood by the indi-viduals. Each driver acts on self-organizing drivinglogic according to the driving rules of the network.

The study of networked behavior beyond thefamiliar territories of business and ICT networks—those of social interactions, ant colonies, bees, andother biological systems—reveals attributes andcharacteristics that can be applied to the design ofsmart business networks. The behavior of the indi-vidual drivers, as described in the preceding exam-ple, demonstrates swarm intelligence: theemergence of seemingly intelligent or, perhaps,smart, behavior from many individuals [2]. Swarm

32 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

1See “Decision Making in Very Large Networks,” by Peter J. Denning and Rick Hayes-

Roth in the November 2006 issue of Communications for more detailed information.

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 35

selection and execution of busi-ness processes across the net-work; and establishing thedecision rules and the embed-ded logic within the businessnetwork.

QUICK CONNECT AND DISCONNECT

Quick connect requires that, asa result of an event the smartbusiness network must seek andselect those members who,together, can fulfill the requiredgoal. This means the networklogic acts on a menu of poten-tial fulfillment partners to selectthose who can combine to pro-duce the desired results. Oncethe appropriate participants arefound and the connection hasbeen established, the process of“play”—performing the busi-ness transaction—can begin.Goldman, Nagel, and Preiss [6]described this in their discus-sion of virtual organizations.The capability of quickly con-nected plug-compatibilityenables a superior responsespeed and greater component variety particularlyfor dealing with new requirements.

While the ability to quickly connect has receivedattention, the capability to quickly disconnectrequires more. Members will join the business net-work and participate on the basis of risk andreward. While this can be clear while they areactive, rules should be agreed upon for when theactors are disconnected (having completed a spe-cific customer order or while they will no longer bea member of the network). Decision rules and logicwith regard to connection and disconnection willbe a crucial component for the success of the busi-ness network.

PLUG AND PLAY: BRINGING THE NETWORK TOGETHER

When selected, the network participants must beable to interoperate. They must be pluggedtogether to enable the required network outcome.This means they must act with modularity: thedecomposition of a system by grouping elementsinto a smaller number of subsystems with rulesgoverning the architecture for mixing and match-ing these components.

The concept of modularity has a long history in

product design and manufacturing enabling prod-uct construction of tailored products from stan-dardized components: the combination ofLego-like modules that are combined in a specificway. Modularity brings the benefits of versatility(the diverse set of products that an organizationcan produce) and agility (the ability to respondquickly to fulfill an unpredictable customer order)while, at the same time, delivering within theboundaries of allowed value chain total costs andlead times. Martijn Hoogeweegen et al. developeda method to design modular business networks andto optimize the allocation of tasks in a businessnetwork based on modularity principles [8]. Thenetwork nodes will be considered black boxes pro-viding the functionality required by the businessnetwork and are “played” according to the networkrules. However, modular design requires muchmore coordination than non-modular: the greaterthe number of components the higher the organiz-ing cost. A crucial decision is the degree of modu-larity or granularity of a system, or businessnetwork, and that is determined by the balancebetween coordination costs and the complexity ofthe network.

EMBEDDED BUSINESS LOGIC

Each business network participant has specificcapabilities captured in its business processes (ownbusiness logic) that it executes according to this

34 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

VanHeck table 2 (6/07)

Table 2. Critical trade-offs and questions with regard tosmart business networks.

Issues Trade-Offs Critical Questions

Network outcomes

Network execution

Network governance

Network design

Enabling technologies

Preferred versus expected outcomes

Network versus individual outcomes and performance

Individual versus network execution

Simple versus complex transactions

Global versus local execution

Integrated versus modular products/services/processes

Central governed versus adaptive, self organizing

Governance to innovate versus governance to operate efficiently

End-to-end control versus individual control

Networked business operating system versus actor business operating systems

Global versus local adaptation

Dynamic versus static business network

Open technologies versus closed technologies

Individual versus network technology adoption

What do customers want in terms of products and services (quality, complexity, bundling, delivery time, price, offline/online)? How to value customer demand?

What can we offer to customers in terms of products and services (quality, complexity, bundling, delivery time, price, offline/online)? What actors will work together to fulfill the customer order? How?

What level of modularity is required?

Who is responsible for the customer order?

How are decision rights delegated?

How is the relational governance evaluated?

How to keep end-to-end control in the network?

How to design and test the networkked business operating system?

How to deal with the environmental and internal dynamics of the business network?

What technologies will be disruptive for the business network?

How to speed up technology adoption in the business network?

Table 2. Critical trade-offs and

questions regardingsmart business

networks.

In the networked world, decision makers must ensure theircompanies understand their position in the business net-work. They should see their company as a node in much

larger and extensive networks of interdependent firms that areconstantly being reconfigured. This perspective implies that deci-sion makers must be acutely aware of their companies’ uniqueposition in the network; not only the direct (supplier or client)contacts but also the contacts’ contacts, their contacts, and soforth. The number of nodes that an actor can “see” from a spe-cific position in the network is called the network horizon. Thefigure here illustrates how thenetwork horizon of Firm A isexpanded using the opportuni-ties of the bridging position ina business network. Our stud-ies have indicated the networkhorizon is an important deter-minant of both individual firmperformance and the overallnetwork dynamics and perfor-mance.

Research was carried outby developing a managementexercise called the BusinessNetworking Engine [2] that canbe used to test the effects ofnew, networked organizationsbased on the concepts of mod-ularity and loosely coupled busi-ness processes. A series ofexperiments with executivesfrom the insurance industryshowed that firms with a higher network horizon achieved higherperformance. These firms were better able to identify favorablenetwork positions where they could act as a bridge between dif-ferent parts of the network: a position rich in structural holes [1].Firms with a lower network horizon rarely identified such oppor-tunities and hence had lower performance.

As the network adapts to its members’ responses tochanging market conditions and competitor actions, the bridg-ing position advantage may be temporary. So when does anetwork position confer a sustainable competitive advantage?Extensive simulation revealed that a key factor is the way thenetwork horizon is distributed across the different firms (inother words, the network horizon heterogeneity). The resultsare striking. In the traditional supply chain where all firmshave a low network horizon (that is, the firms know theirdirect upstream and downstream partners but are aware oflittle else), a competitive advantage can be sustained for sometime.

However, when the network horizon becomes more heterogeneous across firms, for instance because some firmsstart to think in network terms instead of supply chain terms,the situation changes significantly. In heterogeneous networks,a low network horizon confers little advantage but having evena modest network horizon can sustain a competitive advantagefor some time. However, once all firms have a high networkhorizon, the network becomes homogenous again. Any opportunity can be spotted by many firms and any competitiveadvantage is therefore short-lived.

These findings might explain why network orchestrators ina wide variety of industries are so successful. Companies suchas Li & Fung in the apparel industry, Solectron and Flextronicsin semiconductors, and Grupo Multiasistencia in home insur-ance claims and repairs were among the first to see them-selves as nodes in a larger business network. In ahomogenous, low-horizon network, their extensive networkhorizon enabled them to position themselves as networkorchestrators connecting otherwise disconnected parts of thenetwork and become significant companies in their respective

industries while othersstruggled to make the tran-sition to network thinking.

We foresee two chal-lenges for organizationaldecision makers seekingthe networked approach.The first challenge is tostart to think about a net-work strategy, rather thanseeking competitive advan-tage based upon internalcapabilities. By choosingand maneuvering to thecorrect network position afirm can reap even morebenefits from its currentcapability-based competi-

tive advantage. The second challenge is to identify which net-works are relevant for their business and where they canexpand their network horizon. As business networks continueto grow, the entire network cannot be scanned. Some parts ofthe network are more relevant than other parts. As competitiveadvantage is increasingly network-based, the expansion of thenetwork horizon will be a crucial success factor in the nearfuture.

References1. Burt, R.S. Structural Holes: The Social Structure of Competition.

Harvard University Press, Cambridge, MA, 1992.2. Hoogeweegen, M.R., van Liere, D.W., Vervest, P.H.M., Hagdorn

van der Meijden, L., and de Lepper, I. Strategizing for mass cus-tomization by playing the business networking game. DecisionSupport Systems 42, 3 (Mar. 2006), 1402–1412.

More information about the Business Network Engine is available at:bne.rsm.nl.

Diederik van Liere (dliere@rsm.nl) is a Ph.D. candidate atthe Department of Decision and Information Sciences of RSMErasmus University in Rotterdam, The Netherlands. Otto Koppius (okoppius@rsm.nl) is an assistant professor atthe Department of Decision and Information Sciences of RSMErasmus University in Rotterdam, The Netherlands.

2007 ACM 0001-0782/07/0600 $5.00

c

By Diederik van Liere and Otto Koppius

VanHeck sidebar (Van Liere) fig. (6/07)

Sidebar (van Liere) figure caption: Bridging position andnetwork horizon in a network.

EF

G

B

A

DC

B

A

CD

F E

G

Firm A occupies a bridging positionbecause it can broker information andresources between firms B, C and Dwhile these are dependent on firm A.

Suppose firm A has a network horizonthat includes firms E, F and G andrecognizes the opportunity tostrengthen its bridging position.Therefore it establishes a relationshipwith firm F.

Bridging position and network horizon.

NETWORK HORIZON AND OBTAINING A FAVORABLE NETWORK POSITION

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 35

selection and execution of busi-ness processes across the net-work; and establishing thedecision rules and the embed-ded logic within the businessnetwork.

QUICK CONNECT AND DISCONNECT

Quick connect requires that, asa result of an event the smartbusiness network must seek andselect those members who,together, can fulfill the requiredgoal. This means the networklogic acts on a menu of poten-tial fulfillment partners to selectthose who can combine to pro-duce the desired results. Oncethe appropriate participants arefound and the connection hasbeen established, the process of“play”—performing the busi-ness transaction—can begin.Goldman, Nagel, and Preiss [6]described this in their discus-sion of virtual organizations.The capability of quickly con-nected plug-compatibilityenables a superior responsespeed and greater component variety particularlyfor dealing with new requirements.

While the ability to quickly connect has receivedattention, the capability to quickly disconnectrequires more. Members will join the business net-work and participate on the basis of risk andreward. While this can be clear while they areactive, rules should be agreed upon for when theactors are disconnected (having completed a spe-cific customer order or while they will no longer bea member of the network). Decision rules and logicwith regard to connection and disconnection willbe a crucial component for the success of the busi-ness network.

PLUG AND PLAY: BRINGING THE NETWORK TOGETHER

When selected, the network participants must beable to interoperate. They must be pluggedtogether to enable the required network outcome.This means they must act with modularity: thedecomposition of a system by grouping elementsinto a smaller number of subsystems with rulesgoverning the architecture for mixing and match-ing these components.

The concept of modularity has a long history in

product design and manufacturing enabling prod-uct construction of tailored products from stan-dardized components: the combination ofLego-like modules that are combined in a specificway. Modularity brings the benefits of versatility(the diverse set of products that an organizationcan produce) and agility (the ability to respondquickly to fulfill an unpredictable customer order)while, at the same time, delivering within theboundaries of allowed value chain total costs andlead times. Martijn Hoogeweegen et al. developeda method to design modular business networks andto optimize the allocation of tasks in a businessnetwork based on modularity principles [8]. Thenetwork nodes will be considered black boxes pro-viding the functionality required by the businessnetwork and are “played” according to the networkrules. However, modular design requires muchmore coordination than non-modular: the greaterthe number of components the higher the organiz-ing cost. A crucial decision is the degree of modu-larity or granularity of a system, or businessnetwork, and that is determined by the balancebetween coordination costs and the complexity ofthe network.

EMBEDDED BUSINESS LOGIC

Each business network participant has specificcapabilities captured in its business processes (ownbusiness logic) that it executes according to this

34 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

VanHeck table 2 (6/07)

Table 2. Critical trade-offs and questions with regard tosmart business networks.

Issues Trade-Offs Critical Questions

Network outcomes

Network execution

Network governance

Network design

Enabling technologies

Preferred versus expected outcomes

Network versus individual outcomes and performance

Individual versus network execution

Simple versus complex transactions

Global versus local execution

Integrated versus modular products/services/processes

Central governed versus adaptive, self organizing

Governance to innovate versus governance to operate efficiently

End-to-end control versus individual control

Networked business operating system versus actor business operating systems

Global versus local adaptation

Dynamic versus static business network

Open technologies versus closed technologies

Individual versus network technology adoption

What do customers want in terms of products and services (quality, complexity, bundling, delivery time, price, offline/online)? How to value customer demand?

What can we offer to customers in terms of products and services (quality, complexity, bundling, delivery time, price, offline/online)? What actors will work together to fulfill the customer order? How?

What level of modularity is required?

Who is responsible for the customer order?

How are decision rights delegated?

How is the relational governance evaluated?

How to keep end-to-end control in the network?

How to design and test the networkked business operating system?

How to deal with the environmental and internal dynamics of the business network?

What technologies will be disruptive for the business network?

How to speed up technology adoption in the business network?

Table 2. Critical trade-offs and

questions regardingsmart business

networks.

In the networked world, decision makers must ensure theircompanies understand their position in the business net-work. They should see their company as a node in much

larger and extensive networks of interdependent firms that areconstantly being reconfigured. This perspective implies that deci-sion makers must be acutely aware of their companies’ uniqueposition in the network; not only the direct (supplier or client)contacts but also the contacts’ contacts, their contacts, and soforth. The number of nodes that an actor can “see” from a spe-cific position in the network is called the network horizon. Thefigure here illustrates how thenetwork horizon of Firm A isexpanded using the opportuni-ties of the bridging position ina business network. Our stud-ies have indicated the networkhorizon is an important deter-minant of both individual firmperformance and the overallnetwork dynamics and perfor-mance.

Research was carried outby developing a managementexercise called the BusinessNetworking Engine [2] that canbe used to test the effects ofnew, networked organizationsbased on the concepts of mod-ularity and loosely coupled busi-ness processes. A series ofexperiments with executivesfrom the insurance industryshowed that firms with a higher network horizon achieved higherperformance. These firms were better able to identify favorablenetwork positions where they could act as a bridge between dif-ferent parts of the network: a position rich in structural holes [1].Firms with a lower network horizon rarely identified such oppor-tunities and hence had lower performance.

As the network adapts to its members’ responses tochanging market conditions and competitor actions, the bridg-ing position advantage may be temporary. So when does anetwork position confer a sustainable competitive advantage?Extensive simulation revealed that a key factor is the way thenetwork horizon is distributed across the different firms (inother words, the network horizon heterogeneity). The resultsare striking. In the traditional supply chain where all firmshave a low network horizon (that is, the firms know theirdirect upstream and downstream partners but are aware oflittle else), a competitive advantage can be sustained for sometime.

However, when the network horizon becomes more heterogeneous across firms, for instance because some firmsstart to think in network terms instead of supply chain terms,the situation changes significantly. In heterogeneous networks,a low network horizon confers little advantage but having evena modest network horizon can sustain a competitive advantagefor some time. However, once all firms have a high networkhorizon, the network becomes homogenous again. Any opportunity can be spotted by many firms and any competitiveadvantage is therefore short-lived.

These findings might explain why network orchestrators ina wide variety of industries are so successful. Companies suchas Li & Fung in the apparel industry, Solectron and Flextronicsin semiconductors, and Grupo Multiasistencia in home insur-ance claims and repairs were among the first to see them-selves as nodes in a larger business network. In ahomogenous, low-horizon network, their extensive networkhorizon enabled them to position themselves as networkorchestrators connecting otherwise disconnected parts of thenetwork and become significant companies in their respective

industries while othersstruggled to make the tran-sition to network thinking.

We foresee two chal-lenges for organizationaldecision makers seekingthe networked approach.The first challenge is tostart to think about a net-work strategy, rather thanseeking competitive advan-tage based upon internalcapabilities. By choosingand maneuvering to thecorrect network position afirm can reap even morebenefits from its currentcapability-based competi-

tive advantage. The second challenge is to identify which net-works are relevant for their business and where they canexpand their network horizon. As business networks continueto grow, the entire network cannot be scanned. Some parts ofthe network are more relevant than other parts. As competitiveadvantage is increasingly network-based, the expansion of thenetwork horizon will be a crucial success factor in the nearfuture.

References1. Burt, R.S. Structural Holes: The Social Structure of Competition.

Harvard University Press, Cambridge, MA, 1992.2. Hoogeweegen, M.R., van Liere, D.W., Vervest, P.H.M., Hagdorn

van der Meijden, L., and de Lepper, I. Strategizing for mass cus-tomization by playing the business networking game. DecisionSupport Systems 42, 3 (Mar. 2006), 1402–1412.

More information about the Business Network Engine is available at:bne.rsm.nl.

Diederik van Liere (dliere@rsm.nl) is a Ph.D. candidate atthe Department of Decision and Information Sciences of RSMErasmus University in Rotterdam, The Netherlands. Otto Koppius (okoppius@rsm.nl) is an assistant professor atthe Department of Decision and Information Sciences of RSMErasmus University in Rotterdam, The Netherlands.

2007 ACM 0001-0782/07/0600 $5.00

c

By Diederik van Liere and Otto Koppius

VanHeck sidebar (Van Liere) fig. (6/07)

Sidebar (van Liere) figure caption: Bridging position andnetwork horizon in a network.

EF

G

B

A

DC

B

A

CD

F E

G

Firm A occupies a bridging positionbecause it can broker information andresources between firms B, C and Dwhile these are dependent on firm A.

Suppose firm A has a network horizonthat includes firms E, F and G andrecognizes the opportunity tostrengthen its bridging position.Therefore it establishes a relationshipwith firm F.

Bridging position and network horizon.

NETWORK HORIZON AND OBTAINING A FAVORABLE NETWORK POSITION

• Continual Improvement: The capabilities andprocesses of joining and leaving the network overtime, of network renewal and sustainability.

• Fault Tolerance: To malfunction of a node, forexample a business malfunction or bankruptcy.

The concept of a BOS is analogous to that of thecomputer operating system. The invention of anoperating system for the PC in the 1970s allowedapplication software to be run on different com-puters as long as they conform to the rules of theoperating system. Implementation of a BOSenables the portability of business processes andfacilitates the end-to-end management of processesrunning across many different organizations inmany different forms. It coordinates the processesamong the networked businesses and its logic isembedded in the systems used by these businesses.

QUESTIONS FOR THE CIOFor those recognizing the promise, or necessity, ofcapturing the capabilities of smart business net-works to be proactive and effective in this largelyunmapped territory of “being networked” the fun-damental questions include:

• What are the characteristics of smart businessnetworks and how is the current business net-work functioning?

• What are the key enablers and what is theexpected smartness of the business network?

• How to capture the business logic and what arethe potential functionalities of the NetworkedBOS and how to create this business operatinglayer?

• What are the important trade-offs that must beanalyzed and decided upon?

Concerning the last question, we have defined aset of important trade-offs and related specificquestions that must be answered with regard to thebusiness network outcomes, execution, governance,design, and enabling technologies. Table 2 providesthe trade-offs and critical questions.

THE CHALLENGE AHEAD

We may believe that we are familiar with the net-worked world. However, in recent years our under-standing has been disturbed by the rapidemergence of digital social networks—a seemingswarm of consumers who are interacting in waysbeyond that of most organizations. The pressuresof global competition and the need for effectivenessand agility demand new ways to organize. Compa-

nies must develop and act “smart” in rapidly chang-ing and expanding business networks enabled bytoday’s pervasive communications technologies.Chief information officers and professionals mustspan the boundaries between their own organiza-tion and the growing networks in which their orga-nizations operate. They must understand newvocabularies that are not necessarily technically orbusiness oriented. Decision making in very largenetworks is fundamentally different from what weare used to. Understanding how federated activitiesemerge and operate is not a monolithic disciplineand requires the CIO to adapt and adjust to chang-ing conditions.

References1. Barabási, A.L. Linked—How Everything Is Connected to Everything

Else and What It Means for Business, Science, and Everyday Life. Pen-guin Group, New York, 2003.

2. Bonabeau, E. and Meyer, C. Swarm intelligence: A whole new way tothink about business. Harvard Business Review (May 2001), 106–114.

3. Braha, D. and Bar-Yam, Y. Information flow structure in large-scaleproduct development organizational networks. Journal of InformationTechnology 19, 4 (Dec. 2004), 244–253.

4. Burt, R.S. Structural Holes—The Social Structure of Competition. Har-vard University Press, Cambridge, MA, 1992.

5. Golden, W., Hughes, M., and Burke, H. Node to network: Partner-ships in the secondhand book trade. In P. Vervest et al., Smart Busi-ness Networks, Springer, Heidelberg-New York, 2005.

6. Goldman, S.L., Nagel, R.N., and Preiss, K. Agile Competitors andVirtual Organizations: Strategies for Enriching the Customer. Van Nos-trand Reinhold, New York, 1995.

7. Holland, C., Shaw, D.R., Westwood, J.B., and Harris, I. Marketingtranslation services internationally: Exploiting IT to achieve a smartnetwork. Journal of Information Technology 19, 4 (Dec. 2004),254–260.

8. Hoogeweegen, M., Teunissen, W.J., Vervest, P.H.M., and Wagenaar,R. Modular network design: Using information and communicationstechnology to allocate production tasks in a virtual organization.Decision Sciences 30, 4 (Fall 1999), 1073–1103.

9. Kambil, A. and van Heck, E. Making Markets: How Firms Can Designand Profit from Online Auctions and Exchanges. Harvard BusinessSchool Press, Boston, MA, 2002.

10. Konsynski, B. and Tiwani, A. The improvisation-efficiency paradox ininter-firm electronic networks: Governance and architecture consider-ations. Journal of Information Technology 19, 4 (Dec. 2004), 234–243.

11. van Heck, E., Preiss, K., and Pau, L-F. Smart Business Networks.Springer, Heidelberg-New York, 2005.

12. van Liere, D.W. Dynamics of Network Positions. RSM Erasmus Uni-versity, ERIM Ph.D series, May 2007.

Eric van Heck (evanheck@rsm.nl) is professor of informationmanagement and markets at the Department of Decision and Information Sciences of RSM Erasmus University and director ofdoctoral education at Erasmus Research Institute of Management(ERIM) in Rotterdam. Peter Vervest (vervest@d-age.com) is professor of business networking at the Department of Decision and Information Sciences of RSM Erasmus University and partner at D-Age.

This article is based on our work as discussed in the Smart Business Network Initia-tive (www.sbniweb.org).

© 2007 ACM 0001-0782/07/0600 $5.00

c

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 37

logic. Traditionally,when such participantscombine they createinterfaces between capa-bilities: translating fromone business logic toanother and executingaccordingly. This can beseen in the outsourcingphenomenon: carve outthe total function of aparticular business oper-ation and hand it over toanother party. As indicated earlier, traditional busi-ness network approaches lack the ability to rapidlypick, plug, and play to configure rapidly to meet aspecific objective, for example, to react to a cus-tomer order or an unexpected event. Figure 1 pre-sents part of a global business network. Its focus ison the actors and relationships from manufacturersvia multi-modal transportation (road, train, sea-ship) to retailers. Inmost current businessnetworks, companiesare developing capabili-ties at the logistics layerand the transactionlayer. As discussed inTable 1, these actorsfocus on their directpartners and are notable to have the end-to-end management ofprocesses runningacross many differentorganizations in manydifferent forms. Theactor platforms aredominated by information silos residing either indifferent places within an organization as “islands,”or in two or more different organizations. Individ-ual actors are orchestrating processes in their partof the supply chain.

Figure 2 presents an example situation illustratingthe use of the new business network approach. Thecentral idea is that linking partners is on the basis oflinking processes but allowing individual executionaccording to those processes: they act individuallyaccording to the joint rules of the network.

As shown in Figure 2, each of the smart businessnetwork participants becomes a “smart insect” in agoal-seeking swarm. The network separates processfrom execution. It shares the processes required toachieve its goals (the shared business logic) allow-

ing each participant toexecute in its own wayaccording to this logic.This means that, to be amember of the network,an organization must beable to absorb theshared logic and executeaccordingly. This is the“own business logic” ofthe network that can be

enabled by a Networked Business Operating Sys-tem (BOS). Based on the service-oriented architec-ture it resolves the problem of information silos byloose coupling of underlying systems, which areconnected together in a business operating layer.This layer allows process execution and manage-ment “from a distance” from the underlying appli-cation systems. The enabling interorganizationaltechnology architecture must reflect this loose cou-pling. Loose coupling is not synonymous with

decentralized processes.It is quite the opposite,where the processes aremore tightly coordinatedbecause the rigidity of theIT architecture is nolonger a constraint [10].The business operatinglayer can become rathercomplex due to the factthat business logic isdeveloped related to suchissues as:

• Membership Selection: The capabilities todecide which business entities can act as nodesof the network;

• Linking: The positioning and connecting ofnodes to the other parts of the network. Thelinking processes can include the directories(search and select) and routing (path finding)through the network as well as typical commu-nications tasks such as handshake, authentica-tion, and trust establishment;

• Goal Setting: The coordination mechanismsthat determine goals in the business networkand the tasks and responsibilities assigned toeach member node;

• Risk and Reward Management: The division ofmaterial results (profit and loss in a monetarybut also in a fairly loose and generic sense) andthe perceived value by each of the participatingbusiness entities of its share;

36 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

VanHeck fig 1 (6/07)-19.5 picas

Figure 1. The traditional business network approach.

TransactionLayer

LogisticsLayer

BANK

PRODUCER FORWARDER AGENTSHIPPING LINE RETAIL

END CONSUMERINSURER

SEASHIP

INLAND SHIPPING

TERMINALDISTRIBUTIONROAD TRANSPORT

PRODUCER

RFIDdata Scanning

data

Scanning

RFIDdata

Figure 1. The traditional businessnetwork approach.

VanHeck fig 2 (6/07)-19.5 picas

Figure 2. The new business network approach.

TransactionLayer

BusinessOperating Layer

LogisticsLayer

BANK

PRODUCER

Standards

Networked Business Operating System

FORWARDER AGENT SHIPPING LINE RETAIL

END CONSUMERINSURER

SEASHIP

INLAND SHIPPING

TERMINALDISTRIBUTIONROAD TRANSPORT

PRODUCER

RFIDdata Scanning

dataScanning

RFIDdata

EPCGlobal

Figure 2. The new business network approach.

• Continual Improvement: The capabilities andprocesses of joining and leaving the network overtime, of network renewal and sustainability.

• Fault Tolerance: To malfunction of a node, forexample a business malfunction or bankruptcy.

The concept of a BOS is analogous to that of thecomputer operating system. The invention of anoperating system for the PC in the 1970s allowedapplication software to be run on different com-puters as long as they conform to the rules of theoperating system. Implementation of a BOSenables the portability of business processes andfacilitates the end-to-end management of processesrunning across many different organizations inmany different forms. It coordinates the processesamong the networked businesses and its logic isembedded in the systems used by these businesses.

QUESTIONS FOR THE CIOFor those recognizing the promise, or necessity, ofcapturing the capabilities of smart business net-works to be proactive and effective in this largelyunmapped territory of “being networked” the fun-damental questions include:

• What are the characteristics of smart businessnetworks and how is the current business net-work functioning?

• What are the key enablers and what is theexpected smartness of the business network?

• How to capture the business logic and what arethe potential functionalities of the NetworkedBOS and how to create this business operatinglayer?

• What are the important trade-offs that must beanalyzed and decided upon?

Concerning the last question, we have defined aset of important trade-offs and related specificquestions that must be answered with regard to thebusiness network outcomes, execution, governance,design, and enabling technologies. Table 2 providesthe trade-offs and critical questions.

THE CHALLENGE AHEAD

We may believe that we are familiar with the net-worked world. However, in recent years our under-standing has been disturbed by the rapidemergence of digital social networks—a seemingswarm of consumers who are interacting in waysbeyond that of most organizations. The pressuresof global competition and the need for effectivenessand agility demand new ways to organize. Compa-

nies must develop and act “smart” in rapidly chang-ing and expanding business networks enabled bytoday’s pervasive communications technologies.Chief information officers and professionals mustspan the boundaries between their own organiza-tion and the growing networks in which their orga-nizations operate. They must understand newvocabularies that are not necessarily technically orbusiness oriented. Decision making in very largenetworks is fundamentally different from what weare used to. Understanding how federated activitiesemerge and operate is not a monolithic disciplineand requires the CIO to adapt and adjust to chang-ing conditions.

References1. Barabási, A.L. Linked—How Everything Is Connected to Everything

Else and What It Means for Business, Science, and Everyday Life. Pen-guin Group, New York, 2003.

2. Bonabeau, E. and Meyer, C. Swarm intelligence: A whole new way tothink about business. Harvard Business Review (May 2001), 106–114.

3. Braha, D. and Bar-Yam, Y. Information flow structure in large-scaleproduct development organizational networks. Journal of InformationTechnology 19, 4 (Dec. 2004), 244–253.

4. Burt, R.S. Structural Holes—The Social Structure of Competition. Har-vard University Press, Cambridge, MA, 1992.

5. Golden, W., Hughes, M., and Burke, H. Node to network: Partner-ships in the secondhand book trade. In P. Vervest et al., Smart Busi-ness Networks, Springer, Heidelberg-New York, 2005.

6. Goldman, S.L., Nagel, R.N., and Preiss, K. Agile Competitors andVirtual Organizations: Strategies for Enriching the Customer. Van Nos-trand Reinhold, New York, 1995.

7. Holland, C., Shaw, D.R., Westwood, J.B., and Harris, I. Marketingtranslation services internationally: Exploiting IT to achieve a smartnetwork. Journal of Information Technology 19, 4 (Dec. 2004),254–260.

8. Hoogeweegen, M., Teunissen, W.J., Vervest, P.H.M., and Wagenaar,R. Modular network design: Using information and communicationstechnology to allocate production tasks in a virtual organization.Decision Sciences 30, 4 (Fall 1999), 1073–1103.

9. Kambil, A. and van Heck, E. Making Markets: How Firms Can Designand Profit from Online Auctions and Exchanges. Harvard BusinessSchool Press, Boston, MA, 2002.

10. Konsynski, B. and Tiwani, A. The improvisation-efficiency paradox ininter-firm electronic networks: Governance and architecture consider-ations. Journal of Information Technology 19, 4 (Dec. 2004), 234–243.

11. van Heck, E., Preiss, K., and Pau, L-F. Smart Business Networks.Springer, Heidelberg-New York, 2005.

12. van Liere, D.W. Dynamics of Network Positions. RSM Erasmus Uni-versity, ERIM Ph.D series, May 2007.

Eric van Heck (evanheck@rsm.nl) is professor of informationmanagement and markets at the Department of Decision and Information Sciences of RSM Erasmus University and director ofdoctoral education at Erasmus Research Institute of Management(ERIM) in Rotterdam. Peter Vervest (vervest@d-age.com) is professor of business networking at the Department of Decision and Information Sciences of RSM Erasmus University and partner at D-Age.

This article is based on our work as discussed in the Smart Business Network Initia-tive (www.sbniweb.org).

© 2007 ACM 0001-0782/07/0600 $5.00

c

COMMUNICATIONS OF THE ACM June 2007/Vol. 50, No. 6 37

logic. Traditionally,when such participantscombine they createinterfaces between capa-bilities: translating fromone business logic toanother and executingaccordingly. This can beseen in the outsourcingphenomenon: carve outthe total function of aparticular business oper-ation and hand it over toanother party. As indicated earlier, traditional busi-ness network approaches lack the ability to rapidlypick, plug, and play to configure rapidly to meet aspecific objective, for example, to react to a cus-tomer order or an unexpected event. Figure 1 pre-sents part of a global business network. Its focus ison the actors and relationships from manufacturersvia multi-modal transportation (road, train, sea-ship) to retailers. Inmost current businessnetworks, companiesare developing capabili-ties at the logistics layerand the transactionlayer. As discussed inTable 1, these actorsfocus on their directpartners and are notable to have the end-to-end management ofprocesses runningacross many differentorganizations in manydifferent forms. Theactor platforms aredominated by information silos residing either indifferent places within an organization as “islands,”or in two or more different organizations. Individ-ual actors are orchestrating processes in their partof the supply chain.

Figure 2 presents an example situation illustratingthe use of the new business network approach. Thecentral idea is that linking partners is on the basis oflinking processes but allowing individual executionaccording to those processes: they act individuallyaccording to the joint rules of the network.

As shown in Figure 2, each of the smart businessnetwork participants becomes a “smart insect” in agoal-seeking swarm. The network separates processfrom execution. It shares the processes required toachieve its goals (the shared business logic) allow-

ing each participant toexecute in its own wayaccording to this logic.This means that, to be amember of the network,an organization must beable to absorb theshared logic and executeaccordingly. This is the“own business logic” ofthe network that can be

enabled by a Networked Business Operating Sys-tem (BOS). Based on the service-oriented architec-ture it resolves the problem of information silos byloose coupling of underlying systems, which areconnected together in a business operating layer.This layer allows process execution and manage-ment “from a distance” from the underlying appli-cation systems. The enabling interorganizationaltechnology architecture must reflect this loose cou-pling. Loose coupling is not synonymous with

decentralized processes.It is quite the opposite,where the processes aremore tightly coordinatedbecause the rigidity of theIT architecture is nolonger a constraint [10].The business operatinglayer can become rathercomplex due to the factthat business logic isdeveloped related to suchissues as:

• Membership Selection: The capabilities todecide which business entities can act as nodesof the network;

• Linking: The positioning and connecting ofnodes to the other parts of the network. Thelinking processes can include the directories(search and select) and routing (path finding)through the network as well as typical commu-nications tasks such as handshake, authentica-tion, and trust establishment;

• Goal Setting: The coordination mechanismsthat determine goals in the business networkand the tasks and responsibilities assigned toeach member node;

• Risk and Reward Management: The division ofmaterial results (profit and loss in a monetarybut also in a fairly loose and generic sense) andthe perceived value by each of the participatingbusiness entities of its share;

36 June 2007/Vol. 50, No. 6 COMMUNICATIONS OF THE ACM

VanHeck fig 1 (6/07)-19.5 picas

Figure 1. The traditional business network approach.

TransactionLayer

LogisticsLayer

BANK

PRODUCER FORWARDER AGENTSHIPPING LINE RETAIL

END CONSUMERINSURER

SEASHIP

INLAND SHIPPING

TERMINALDISTRIBUTIONROAD TRANSPORT

PRODUCER

RFIDdata Scanning

data

Scanning

RFIDdata

Figure 1. The traditional businessnetwork approach.

VanHeck fig 2 (6/07)-19.5 picas

Figure 2. The new business network approach.

TransactionLayer

BusinessOperating Layer

LogisticsLayer

BANK

PRODUCER

Standards

Networked Business Operating System

FORWARDER AGENT SHIPPING LINE RETAIL

END CONSUMERINSURER

SEASHIP

INLAND SHIPPING

TERMINALDISTRIBUTIONROAD TRANSPORT

PRODUCER

RFIDdata Scanning

dataScanning

RFIDdata

EPCGlobal

Figure 2. The new business network approach.