66 Oilfield Review

Managing Knowledge Management

PARTICIPANTS:

Erik Åbø StatoilStavanger, Norway

Lesley Chipperfield Shell International E&P Rijswijk, The Netherlands

Chris Mottershead BP Sunbury on Thames, England

John Old Texaco Houston, Texas, USA

Rodulfo Prieto Petróleos de Venezuela S.A. (PDVSA) Caracas, Venezuela

Jeff Stemke ChevronSan Ramon, California, USA

MODERATOR:

Reid Smith Houston, Texas

Oil and gas companies have varied approaches to sharing knowledge. Experts from

six E&P companies discuss experiences in establishing knowledge-management

infrastructures—what they’ve learned to date and what the future may hold.

Moderator, Reid Smith, Vice President, Knowl-edge Management, Schlumberger: Many companies develop knowledge manage-ment programs in response to key issues or aparticular event within the organization. Whatbusiness drivers are behind your knowledge-management efforts, and what are your compa-nies trying to achieve through knowledgemanagement?

Lesley Chipperfield, Manager of Organiza-tional Performance, Shell International E&P:Our start in knowledge management began withthe major reorganization of Shell in late 1995 andearly 1996 when we started transitioning from agroup of globally dispersed companies that werenot linked to a globally connected company.Before that, the companies didn’t talk directly toeach other, and if they did, it was through servicecompanies. Reorganization mandated that we dothings differently and required a cultural shift.

Many of the initiatives to start sharing knowl-edge were kicked off in 1996 and 1997.Management recognized that good things werehappening and, in 1998, my group was formed topull together these divergent, entrepreneurialknowledge-management efforts, which had actually fragmented us. Trying to leverage bestpractices helped establish our current direction.

Chris Mottershead, Technology Vice President,Lower Carbon Growth, Global Business Center, BP:BP started knowledge management in the drillingorganization in 1992 or 1993 with training andlearning. The average driller must make quick

decisions “on the go” and has considerably morepersonal accountability sooner than other disci-plines. As a result, drillers seem more willing toask for and accept help, and they are particularlyreceptive to new ideas. Like other companies,our organization was no longer centralized.People couldn’t perform well unless they couldengage their peers and get help that previouslyarrived with authority from the corporate office.They had to share information.

In 1994 and 1995, we started evaluating howto improve virtual teamworking (VT). A camera onevery desk let you see the person you were talk-ing and working with long distance. In the NorthSea, we established pilot VT programs withinbusiness units like the Miller field area. Initially,the objective was to improve onshore and off-shore communication. In the Andrew field busi-ness unit, we used VT to connect the variousdisparate activities of multiparty constructionprojects. Then, we connected the executive man-agers in our worldwide upstream businessthrough VT to avoid regional isolation. This wasremarkably successful, but once we got a globalorganization going, this type of hardware wasn’tneeded anymore.

Outside BP, knowledge management seemedto be grounded more in lessons-learneddatabases, which consisted of information thatno one really wanted, hidden away in computerfiles that very few people knew how to access.We created a knowledge-management team in1995 or 1996. Our chief executive thought knowl-edge management was important and still does,so it received much attention. For about two

For help in preparation of this article, thanks to DennyO’Brien and Justin Rounce, Sugar Land, Texas, USA.Lotus Domino, QuickPlace and SameTime are marks ofLotus Development Corporation. Microsoft Project andNetMeeting are marks of Microsoft Corporation. SiteScapeis a service mark of SiteScape, Inc.

Spring 2001 67

years, this team did groundbreaking work todefine knowledge management, look for bestpractices, particularly from the U.S. Army, andbuild relationships.

The strength and success of knowledge-management efforts depend on how broadly youbuild the network. In this case, it was quiteextensive and included people with academicand practical experience. The end result was fiveor six knowledge-management nuggets thatremained only with the 12 or so people on thatknowledge-management team. When wemerged with Amoco, differences between thetwo companies were so minor that it was almostsurreal. They had 12 people in knowledge man-agement; we had 13 or 14.

We thought there was a dangerous percep-tion that the only people doing knowledge

management were inside the knowledge-management team, and unless you had theirhelp, you weren’t doing knowledge management,which clearly wasn’t true. With a staff of 25,000technical people, most of the value comes fromknowledge they apply daily. Believing that 12 oreven 26 people were going to reach the entiretechnical staff was unrealistic, so we dispersedthis team back into the business units in 1999.That strategy was successful because values andactivities added by this team up to that pointwere transfused back into the organization. Agroup with two or three members from the origi-nal knowledge-management team was main-tained to support the exchange of knowledgebetween not only different geographies, butother business groups as well.

Chris Mottershead, BP: Like other companies, our organization wasno longer centralized. People couldn’t per-form well unless they could engage theirpeers and get help that previously arrivedwith authority from the corporate office. They had to share information.

Jeff Stemke, Knowledge Management/Collaboration Consultant, Chevron: We got into knowledge management in the early1990s as an outgrowth of the quality movement.The primary issue for us was cost reduction. Welooked outside the company and saw that ourcost per barrel was out of line. The idea of look-ing inside and outside the company for best prac-tices grew out of quality initiatives to reducecosts. We identified some breakthrough qualityprojects that touched broad processes likeenergy and project management, refining opera-tions and drilling and put teams together totackle those initiatives. They did some wonderfulwork pulling together best practices.

For example, we developed a world-class pro-ject-management process. Our refineries wentfrom six or eight regional fiefdoms to an inte-grated organization that understood the need forsharing operational knowledge. That workoccurred from 1992 to 1995. We made progressand had success, but the results didn’t stickbecause these efforts didn’t focus on looking forand sharing best practices as the standard wayof doing business. A good example was Year2000 (Y2K) preparation. We had to look atembedded systems and understand their impactfrom a Y2K perspective. In a knowledge-sharing

organization, you would think that various groupswould pool their knowledge, but that didn’t hap-pen right away. It took some time.

We reached a plateau because knowledgemanagement wasn’t embedded in our business.Our new CEO, Dave O’Reilly, has always appre-ciated knowledge management, but he puts it ina different context by talking about five strate-gic intents for the company. Many companieshave similar business strategies, such as oper-ating excellence, cost recovery, capital steward-ship and profitable growth, but he introduced afifth, organizational capability—our ability toexecute the other strategies. It includes learn-ing from each other, reusing what we know andworking effectively in teams. We’re still defin-ing the key elements of organizational capabil-ity, but they closely resemble the components ofknowledge management.

Rodulfo Prieto, Exploration Project Manager,Petróleos de Venezuela, S.A. (PDVSA):In the late 1980s, PDVSA started with quality andother issues that focus on business processes andsharing these processes, which was quite effec-tive because it defined the workflow for manyactivities. Most of the time, geologists operate inthe arena of mental creation. Mental creation is

important, especially at the beginning of projectsbefore commitments are made. When the threePDVSA affiliates merged in 1998, it was impor-tant to create an organization that combines tech-nological competencies and information throughwhat we were calling knowledge management.We like the term “knowledge” because it empha-sizes the mental-creation phase of every project.It implies working at the highest level of technol-ogy to share data and information, and close anygaps in our professional competencies.

We got into knowledge management becausewe had so many projects going on that it was diffi-cult to standardize them without limiting creativity.People know a lot, but apply only a small percent-age of what they know and share even less of whatthey apply. It was important for us to capture thisknowledge. Through knowledge management,leaders not only share experience and knowledge,but move ahead to create what I call “contamina-tion centers” where people infect each other withideas. This gives people visibility and an opportu-nity to say, this is good or this is not. In these cen-ters, people are motivated to improve businessactivities and return value to the organization.

We developed an internal system calledProject-Net, so that everyone can share bestpractices through our intranet. Everyone has

68 Oilfield Review

Rodulfo Prieto, PDVSA: We like the term “knowledge” because itemphasizes the mental-creation phase ofevery project. It implies working at the highest level of technology to share data and information, and close any gaps in ourprofessional competencies.

Jeff Stemke, Chevron: The idea of looking inside and outside the company for best practices grew out of quality initiatives to reduce costs.

Spring 2001 69

access to new applications, technology and,above all, information about how things arebeing done on exploration projects around thecorporation. We have also had a knowledge-management manager since 1998. This positionreports to the Board of Directors and is responsi-ble for ensuring proper application of technolo-gies and the sharing of best practices andlessons learned throughout the organization inorder to maintain a high level of competencies.

Erik Åbø, Chief Engineer, Well and ProductionTechnology, Statoil: We got into knowledge management for practi-cal reasons. Like other companies, we haddecentralized and needed a way to transfer expe-rience, share best practices and run networks.There also were an increasing number of projectswith limited staff, and we didn’t want peopleusing the majority of their time just to gatherdata. When we talk about knowledge manage-ment, it’s not just about communication and com-puter technology; it’s redesigning the workprocess to increase efficiency.

And third, subsurface information, especiallyseismic data, was increasing exponentiallybecause of advances in three-dimensional (3D)technology. We needed an easy, effective way to

access corporate data, but many of the existingsoftware tools didn’t work together or didn’twork in the same databases. To increase staffefficiency, we decided to establish a central datastore, use a portfolio of tools that could interfaceand redesign the work process. Most of the computer tools that manipulate subsurface datanow work on common platforms. For communica-tions in the decentralized organization, we cre-ated an extranet, which is something between anintranet and the Internet, and includes suppliersand partners.

Reid Smith, Schlumberger: Almost a decade of working to establish knowl-edge-management infrastructures has obviouslyprovided a wealth of experience and lessonslearned (see “Lessons and Nuggets,” page 72 ).What business problems have you addressed,and what approaches—communities of practice,networks, capturing and reusing best practices,lessons-learned databases, knowledge reposito-ries and portals—have you tried?

John Old, Focus Area Leader, InformationManagement, Texaco: Our strategy focuses on connecting people.We’ve done all those things to some extent, but

now Texaco emphasizes networks. One of thefactors for a successful network or community isa leader with energy to keep the group going. Wedeliberately created a network for knowledge-sharing activities. It is composed of people fromthe business units who have a passion for knowl-edge management. Twice a month, we gettogether, talk about what people are doing andwhat needs to be done to create and sustainknowledge-management energy.

Our general engineering group, whichfocuses largely on the downstream business, hasa database repository of best practices, or million-dollar stories, which are very good andactually entertaining to read. They probably getused more frequently than other databases.Knowledge-management tools like databaserepositories should be connected to people. Ifyou capture best practices, you should use themas a means of pointing out the right people totalk with when you face a problem.

Lesley Chipperfield, Shell International E&P: We’ve also used many approaches, but within E&P,we focus on people and people-to-people connec-tions. Perhaps this is a reflection of the E&P busi-ness where personal contacts are so important. Fora given topic, we can search our intranet and get

Erik Åbø, Statoil: There also were an increasing number ofprojects with limited staff, and we didn’twant people using the majority of their timejust to gather data. When we talk aboutknowledge management, it’s not just aboutcommunication and computer technology;it’s redesigning the work process toincrease efficiency.

John Old, Texaco: If you capture best practices, youshould use them as a means of pointingout the right people to talk with whenyou face a problem.

500 hits, but if someone recommends one, it hasmore value and credibility. The people you connectwith may direct you to a report, but that report nowhas a personal reference. We have a slogan,“Knowing who is as good as knowing how.” Ourprimary and most successful knowledge-sharingsolution for promoting people-to-people connec-tions is the networks that were initially establishedamong the technical communities.

Although we have moved away from con-cepts like central, formal best-practice reposito-ries, some communities use them if that suitstheir style of working. On a global basis, most ofour staff is connected, at least indirectly. You canpose a question on the intranet and get an aver-age of two or more replies. At least 75% of ourtechnical professionals belong to one or morenetworks. We call them global networks, butthey could be called communities of practice. It’sinteresting that in BP, knowledge managementbegan with drilling because in Shell the wellsgroup also started it first. We thought it wasbecause of a few energetic individuals, but itmay have more to do with the nature of drillingand completion engineering work.

Our three big technical communities arewells, subsurface and surface, but these globalnetworks were actually formed by merging 70 or

80 smaller discipline and subdiscipline groups.We looked at what the wells network had done,which was build a bigger community by drawingdisciplines together, and tried to replicate that inthe subsurface and surface networks. Many ofour business networks cut across all disciplinesand supplement the technical networks. Wealready have networks for knowledge-sharingand competitive-intelligence as well as anothercritical cross-discipline network for procurement,and networks for benchmarking, humanresources, safety and environment are just get-ting started. We’re trying to reduce the numberof networks that individuals must belong to inorder to access the expertise they need.

Chris Mottershead, BP: We draw a clear distinction between communi-ties of practice and communities of interest.Other than making sure they have resources, wedo not control communities of interest, which arecreated by people because they identify withthem. That’s the larger number of communities.For example, there’s a 3D geophysical modelingcommunity, but groups like this are not managedin any real sense. People join communities andparticipate because they have common interests.What we strive for is community ownership.

There is some degree of oversight and assur-ance to make sure that there are working net-works in areas where we must deliver againstspecific objectives, which are important to thecompany. Within the wells area, for example,there are five or six communities of practice orga-nized around themes like nonconventional wells,stuck-pipe prevention and deepwater drillingwhere, in a sense, there is self-regulation. If youhave a major role in these areas and you’re not ina network, someone may pointedly ask, in termsof accepted practice, why you don’t participate.

Connecting people was a factor in taking ourknowledge-management team apart and putting itback together later. We started knowledge man-agement by promoting networks, but in our orga-nization, the networks probably didn’t initially getcaptured within the formalities of knowledgemanagement. This is one reason we were uncom-fortable with a central knowledge-managementorganization because people started to rely onthat group to construct knowledge assets. As aresult, we pulled back and moved closer to whatShell and Texaco are doing—connecting people.

There were parallel efforts because we had alead person in knowledge management and onefor engineering codes and standards, and theywere on the same team. There was no one else on

70 Oilfield Review

Lesley Chipperfield, ShellInternational E&P: We have a slogan, “Knowing who is asgood as knowing how.” Our primary andmost successful knowledge-sharingsolution for promoting people-to-peopleconnections is the networks that wereinitially established among the technicalcommunities.

Chris Mottershead, BP: People join communities and participatebecause they have common interests. What we strive for is community ownership.

Spring 2001 71

the team; everyone was doing his or her regularjob. There may have been committees, but therewas no central engineering department to say thatsomething was right or wrong. Because we hadsimilar approaches in knowledge managementand explicit codes and standards, we tried to bringthose two things together by constructing a modelwith four types of organizational wisdom—everyday practices, shared practices, good prac-tices and recommended practices.

There was wisdom represented by practicesthat are used every day somewhere in the com-pany. There was wisdom from shared practicesthat result from connecting people. You may asksomeone for explicit help, which is similar toaccessing a report. You’re not as interested in the report itself as in the name on the report.Therefore, much of our networking involved waysof connecting people through tools likeSchlumberger Connect Oil & Gas. Then fromshared practices, you eventually make tacit knowl-edge into explicit knowledge, which we call goodpractices. The typical example is codes and stan-dards. They’re not right or wrong, but if you wanta turbine, here’s how you procure and install it. Anexpert user probably takes best practices and doessomething different, but you can just use a goodpractice if you don’t know anything else.

The fourth type of wisdom is from recom-mended practices, which are actually just expec-tations, not definitive answers. They arequestions you ask rather than answers you seek, which forces you to cascade back down thechain just by getting two people together. Thereal added value goes back to where it origi-nated in a shared practice, but the process isstructured in a way that connects disparatethings together.

Reid Smith, Schlumberger:Trust is extremely important. Using knowledge-management resources, one of our engineerswas able to find what he needed to get startedon an offshore extended-reach drilling project inWest Africa. He found a best practice that hewas extremely positive about because he knewthe person who submitted it. This person hadpreviously been his field service manager and hetrusted him.

How do you connect people to the things theyneed to know to do their jobs through problem-solving, knowledge-sharing and innovation? Doyou have different types of networks or commu-nities? How do the business units use them?How are they organized and what support dothey have?

Rodulfo Prieto, PDVSA:Our production and exploration communitiesdeveloped in different ways. In the productionbusiness unit, communities act as centers ofexcellence. They validate, certify and provide sup-port for decision-making. Most production projectswill have members from the community consult onthe project when an important benchmark is beingdiscussed. This process is mandatory.

In the exploration business unit, we have acommunity of interest where people participate bychoice in their area of specialization. There is dualcitizenship in the sense that you’re part of a pro-ject, but you also belong to the exploration com-munity at the same time. The project has theoption to consult with the community if necessary.

We don’t have a unique structure for commu-nities. For example, there are communities ofinterest that developed on their own like the geo-physical group, which organized an informalweekly forum within the company to share knowl-edge. We support these communities and thistype of communication between members as well.Communities ask for budgets when they want todo something. In exploration, communities arefunded through various projects or technologicalgroups under the knowledge-management depart-ment. I’m interested in how funding and budgetingare done in other organizations

Reid Smith, Schlumberger: Trust is extremely important. Usingknowledge-management resources, oneof our engineers was able to find what he needed to get started on an offshoreextended-reach drilling project in WestAfrica. He found a best practice that hewas extremely positive about because he knew the person who submitted it.

Rodulfo Prieto, PDVSA: In the exploration business unit, we have acommunity of interest where people partic-ipate by choice in their area of specializa-tion. There is dual citizenship in the sensethat you’re part of a project, but you alsobelong to the exploration community at the same time.

72

Management push is necessary to get throughthe valley of pain on the way to the summit ofgain—Erik Åbø, Statoil.

The greatest challenge in creating a knowledge-sharing culture is coming to the realization that sharing knowledge saves rather than consumes time—Lesley Chipperfield, ShellInternational E&P.

Successful knowledge management must engageeverybody in the organization. It is not a centralactivity that collates and validates a corporateencyclopedia of subject-specific knowledge orknowledge-management processes. Having acentral team means that you may rely on it toomuch. Knowledge management is not an ency-clopedia, but rather a recipe book to start people talking. Knowledge management is aconsolidated tool for the disparate learningthroughout the organization, but it is not adefinitive answer. It is a creative tool that people use to invent their own answers,

informed by and armed with the experience of others. Fundamentally, this is knowledgemanagement of knowledge management—Chris Mottershead, BP.

When members of a group, team, network orcommunity understand the individual communi-cation preferences of each person, and theirresponsibility to communicate in ways thatmatch those preferences, the effectiveness ofthe whole group improves—John Old, Texaco.

Our biggest challenge is to have a corporationwith organizations, teams and individuals thatapply what they know and recognize what needsto be learned or unlearned to improve cost and risk assessment, and reduce project cycletime through integration of all available information—Rodulfo Prieto, PDVSA.

The biggest challenge is to create and nurture a knowledge-sharing culture in which peopleshare knowledge and learn from others as amatter of course. They see it as simply the rightthing to do—Reid Smith, Schlumberger.

We must provide connections between individu-als who seek and supply problem-solving ideasand experience, and those in a community whoshare experience and reuse practices for per-sonal professional development, corporatelearning and innovation. We also need to con-nect people with explicit knowledge by makingit easier to tap into the growing volume of docu-mented practices and lessons learned.

As chief knowledge bee, I travel around cross-pollinating the organization by picking up ideasand practices in one location and spreadingthem to others. Occasionally, I help a teamdevelop or use a knowledge-sharing tool or pro-cess and leave some “honey” behind. I realizethat it takes time to redesign the “hive,” so I use honey rather than a sting to induce partic-ipation—Jeff Stemke, Chevron.

Oilfield Review

Lessons and Nuggets

Lesley Chipperfield, Shell International E&P:One of the roles of my group is to help networksget funding, but value propositions come fromthe networks. We can help, but networks have tomake a case for their existence. However,because networks are now such an integral partof the company infrastructure, we’ve moved pastthe point where groups must start from zero. Askif we could do away with networks, and peoplereact in horror because they wouldn’t have suffi-cient resources. We rely on picking up practicesfrom others rather than finding them ourselves.The network organization is completely informal.

We have a combination of common-interestnetworks and communities of practice. There area few exceptions, but people are not obliged to

belong to these networks and communities. Theyget involved because they want to contribute. Inthe early stages, people would ask, “How do Ifind time for networks?” I haven’t heard that forquite a while. The first year, I got calls to budgetfor the hours people spent networking. Thatproblem has also gone away. People see net-works as a valuable resource because they sim-ply make sense. Now we budget primarily aroundpeople who actually manage knowledge and networks, mostly out in the business groups. Inmy group, the minority of our knowledge-management budget is spent to facilitate andsupport best practices.

People were beginning to ask, “If you have aline role, why aren’t you active in a network?”

We, therefore, started appointing global consul-tants—individuals who are experts in a specificarea, but may work anywhere in the company.They are listed in an expert directory. It’s an eliteclub, so we don’t call it Yellow Pages becausenot everyone can join. You have to be nominatedand approved. These individuals are expected tobe active in their networks, but they also areexpected to contribute to other networks on anexchange basis and can do work for other com-panies if invited to do so.

John Old, Texaco:That’s similar to the Texaco Fellows program. Ithas separate funding and members are expectedto contribute globally.

Spring 2001 73

Erik Åbø, Statoil: To handle funding, we establish discipline advi-sors whose job descriptions include running net-works. This position is based on the need for anetwork in a particular area. These individualsare used in more or less the same way as globalconsultants. Those who contribute to networksare also contributing to other projects, butbecause of advances in communications andcomputers, the time they actually spend runningnetworks is minor.

Lesley Chipperfield, Shell International E&P:A number of things can be done, depending onthe roles of people and networks. Our wells com-munity has regular teleconferences and face-to-face meetings between coordinators around theworld to discuss issues. We don’t have disciplineadvisors anymore, so networks grew to fill thegap that was left when we reorganized. Theoperating units are more asset-focused, so wedon’t have strong disciplines anymore. Becauseof the way the organization evolved, peoplemaintain their discipline affinity primarily throughparticipation in networks.

Chris Mottershead, BP: We started rebuilding and repopulating theknowledge-management group under technologyvice presidents rather than discipline leaders.We felt that after five or six years of lettingknowledge management develop independentlyin areas such as drilling or project management,we were starting to lose value and needed tochange again. Now, we have moved back to asystem with discipline leaders. The question isalways, “What’s the next step that’s needed toaddress the gap you’re trying to fill?” But onceyou fill that gap, you’ll probably want to do things differently.

John Old, Texaco:That touches on the fact that there’s growingrecognition in all types of industries and compa-nies that the only sustainable advantage a com-pany has is how people work together. That’ssomething no one can copy, unless you take overthe other company and maintain complete control, which is not likely to happen. Thinkingabout how we get people to work together inuniquely different ways is the direction manycompanies are taking. This rise of communities isjust the beginning.

Jeff Stemke, Chevron:The most successful communities, like our best-practices refining networks, have defined busi-ness goals, clear sponsorship from seniormanagement and a dedicated coordinator. In ourcase, a person, called a master, has full-timeresponsibility to collect knowledge throughoutthe refining organization for a particular process.That’s also the model we’re applying in theupstream area.

At the other extreme are informal com-munities where there’s no leader, just a group of people who get together. They may haveteleconferences or meetings occasionally, butthere is no formal process for sharing knowledge.These groups are valuable only if you happen toknow the community. If you aren’t connected toit, you don’t know that it’s out there and have nolegacy to tap into and get started.

We now recognize that networks need a coor-dinator. This position is funded or we recommendhighly that it be funded to the extent of 10 or 20%of a person’s job, depending on the communitysize and activity. We have not been totally suc-cessful in making the communities vital. Theredefinitely needs to be some executive sponsorship

John Old, Texaco: … there’s growing recognition in all types of industries and compa-nies that the only sustainableadvantage a company has is howpeople work together.

Erik Åbø, Statoil: Those who contribute to networks arealso contributing to other projects, butbecause of advances in communicationsand computers, the time they actuallyspend running networks is minor.

and specific deliverables or metrics that the com-munity strives to achieve and that people canmeasure. In this way, communities know they’reon track and others can see what they haveachieved. This hybrid model is summarized in therecommendations for successful communitiesfrom a best-practice study report, titled “Buildingand Sustaining Communities of Practice,” by theAmerican Productivity & Quality Center (APQC).1

Reid Smith, Schlumberger:It’s important to understand a community’s respon-sibility. Every community does not have to havequantitative goals, but if it doesn’t have somethingto strive for—a road map, a vision, some goals—as suggested earlier, the community may break upover time. I also see evidence of community mem-bers being unhappy when they don’t understandwhy we are doing knowledge management. Chris,I understand that BP technology vice presidents arenow responsible for marketing, licensing and sell-ing communities to make sure that results are uti-lized in other parts of the company.

Chris Mottershead, BP:Yes, that’s correct. During the 1990s, our refiningnetworks were similar to what we’re talkingabout. For example, there was a maintenancenetwork, or task team, but it in no way managedthe collective wisdom of people involved in that

activity. Instead, it dealt with how to make main-tenance less expensive next year than this year.These teams were called networks, and theirfocus was explicit, but they weren’t building aknowledge repository. The next level is networksof practice, where technology vice presidents aregiven accountability to make sure disciplines arehealthy and people can do their jobs. These net-works not only deliver today’s results, but alsobuild for the future. In some sense, communitiesneed to be accountable.

We gave communities more visibility andrationalized them in the same way as Shellbecause there were too many and we couldn’tmaintain good control and oversight. We neededto reduce the number of communities and givethem clear deliverables. We decided which net-works or communities were needed to help trans-fer knowledge. These were based largely onpeople knowing each other, which is similar to theway people validate information by the name ona report and then start to recognize it as reliable.

Sometimes, to believe that the message is apiece of wisdom rather than an individual prefer-ence or prejudice, users need to look the messenger in the eye. This emphasizes theimportance of a spectrum of knowledge-sharingactivities, from explicit knowledge in repositoriesto getting people together in one room every yearor two. You need to do all of these things.

John Old, Texaco:About a year ago, we wanted to get the leadersfrom various communities together. At that time, ifwe talked about networks or communities, it didn’t get much attention. But lessons learned wasa term everyone was using and wanting to hearmore about, so we held a lessons-learned summit,and got the community leaders together to talkabout what made networks work and what didn’t.

The same things that came out of the APQCbest-practice study, like an engaged leader and aclear business purpose, were at the top of thelist. Knowledge-sharing groups need to have aclear business purpose or measurable objectivessuch as improving the reliability of rotatingmachinery, which is one of the most successfulTexaco networks. What’s in it for the individualhas a strong influence on network success. If thepeople themselves don’t personally get a lot outof a network, it’s almost always unsuccessful.

Lesley Chipperfield, Shell International E&P: A clear business purpose is important, but itchanges over time, sometimes quite rapidly.Networks have to continuously and effectivelyreview their core purpose. We have many strandsrunning in our networks. For example, you mighthave improving reliability as a specific theme, butwith other themes running in parallel. We have acontinuous theme of networks being the place to

74 Oilfield Review

Jeff Stemke, Chevron: The most successful communities, like ourbest-practices refining networks, have definedbusiness goals, clear sponsorship from seniormanagement and a dedicated coordinator.

John Old, Texaco: If the people themselves don’t personally get a lot out of a network, it’s almost alwaysunsuccessful.

1. American Productivity & Quality Center (APQC), Houston,Texas, USA. [http://www.apqc.org]

Spring 2001 75

seek advice, which is tremendously valued by theorganization. In some ways, our networks aremodeled around “good-old-boy” networks. Ioften describe it as letting the “new guys” intoexisting networks quickly.

What we do now is register people for ahome network when they attend training by hav-ing a short module or an opportunity to explorethe networks. We’re trying to address the timerequired for people coming into an organizationto climb the steep learning curve of knowing whocan help. That’s a powerful thing.

Measuring time to competency as a metric issomething we’re working on. You can’t necessar-ily bring everyone to a training center, andexperts may not be located where learning is tak-ing place. In the future, we will use networks asa direct tool to support training programs, and asa tremendous resource for learning.

Reid Smith, Schlumberger:We try to apply everywhere what we learn any-where. The Schlumberger InTouch system helpsensure that field engineers have access to thebest available knowledge. There are about 165full-time InTouch engineers staffing 75 helpdesks. If you can’t find the answer on a knowl-edge platform, you go to this help desk. Whetheryou send e-mail, connect through the Web or call

on the telephone, it’s their job to find the answer.Sometimes they go directly to the people in acommunity who are responsible for validatinginformation.

We learned a great deal about validation fromChevron. Validation of information is an extremelyimportant issue and one of the things that differ-entiates the style of network. How do your organi-zations validate information? Do most communitiesmake decisions, or do they just advise?

Rodulfo Prieto, PDVSA:When communities make decisions, they may“gold plate” things to protect themselves. Goldplating is sometimes common in seismic inter-pretation and drilling. In 3D seismic surveys,interpreters often want to do high-resolutionsequence stratigraphy and evaluate every line ofdata, which may add minimal value comparedwith interpreting every other line, but in terms ofpeople and time, costs can double. In drilling,however, it may be important to evaluate severaldifferent scenarios to determine the safest andlowest cost option.

There are different types of communities:some advise; others make decisions. In ourexploration organization, project members con-sult with communities on an as-needed basis forspecific situations. However, the added value of

various solutions presented by the communitiesmust be taken into account, so project leadersand team members discuss the recommenda-tions and decide which approach is best.

Erik Åbø, Statoil: I have an example that illustrates business objec-tives and addresses concerns about gold plating.When our well-intervention network getstogether, they see the Key Performance Indicator(KPI) relating to their network. A KPI that we usein well construction and well intervention is costper foot or meter. Coiled tubing is often expen-sive and risky to use. A well tractor is expensiveenough, but costs less than coiled tubing.Therefore, engineers try to use well tractorsbecause they want to see the KPI going in theright direction.

This is a good approach because the managerdoesn’t decide whether to use coiled tubing or awell tractor. The program engineer decides. He’sthe one who goes to network meetings, sees theimprovement and gets ownership of the KPI. It’simportant to link key performance indicators totechnical networks. As an engineer, I like gold-plated solutions, but I also like to see improve-ment in the KPI because we are all competitive.

Lesley Chipperfield, Shell International E&P: In some ways, our networks are modeledaround “good-old-boy” networks. I oftendescribe it as letting the “new guys” into existing networks quickly.

Reid Smith, Schlumberger: We try to apply everywhere what we learnanywhere. The Schlumberger InTouch systemtries to ensure that field engineers haveaccess to the best available knowledge.

Jeff Stemke, Chevron:We have good ideas, good practices, local bestpractices and industry best practices. There weredefinitions associated with this validation con-cept, but there wasn’t a clear process. If I putsomething into a best-practice site because I didit and it worked for me, that’s the first level ofqualification. But how do I validate information ina community context? We don’t have people ded-icated to validating concepts like those in theSchlumberger InTouch system, so the idea is touse other members of the community for valida-tion because they can say, “I tried that and, in mycircumstance, it worked.” This does two things.First, it validates the idea, and second, it estab-lishes behaviors around not just sharing, butreusing, knowledge. I’d like to know if anyoneelse is using this type of validation concept.

Reid Smith, Schlumberger:That’s exactly what the engineers staffing theInTouch help desks do. These engineers comedirectly out of the business segments. This community of people finds and validates informa-tion or solutions. It’s part of their job. They are the network.

John Old, Texaco:It also sounds like what other companies,Electronic Data Systems (EDS) for example, do.The first level is “I’ve tried something and itworked for me in my local situation.” The nextlevel is something that’s been applied in a num-ber of business units or different situations, soit’s worth looking at to see if it’s applicable inother areas or across the company. Then, if thesethings apply everywhere, don’t ask questions,just go do it. It’s the same as in the SchlumbergerInTouch system.

Lesley Chipperfield, Shell International E&P:As far as validating best practices, in general, ourcommunities validate. We have a number of so-called best-practice databases on the Web sites.Some exist within networks and others are indocument-management systems, so they’repretty diverse. Practices that are reused the mostexist in some way within the networks. Thesepractices are generally flushed out becausesomebody asks a question. Usually, there’salready a demand before the practice is cap-tured. Best practices don’t necessarily originatebecause someone’s proud of something they’vedone. They originate because someone asks aquestion and someone else offers the practice asan answer.

Then you get other contributions from thecommunity such as, “There’s a better idea or,under those circumstances, that won’t work; thismight work better.” One of the roles of ourfunded network positions is responsibility forclosing out those conversations and capturingthat knowledge nugget. Those nuggets can thenbe filed where they are easy to access. This pro-cess works better in some areas than others.Realistically, it does require a truly knowledge-able network moderator to understand the con-versation and the quality of the proposal, andwhat should happen next.

Reid Smith, Schlumberger: If I’m not mistaken, Shell has dual axes. One axisis the formal project or organizational hierarchyand the other is more discipline- or community-related. The horizontal axis is a cross section ofthe company that some say more closely repre-sents people’s careers as they move through a variety of jobs. Do you see networks or communities playing a role in career develop-ment as well as in meeting knowledge-manage-ment objectives and providing the other functionswe’ve discussed?

Lesley Chipperfield, Shell International E&P:In the mid-1990s, we established an internalopen resourcing system. Shell was emphasizingthat it was the individual’s career. There is nocentral staff looking after individual careers, butcareer advice is available through SkillpoolManagers. Each of the technical networksaddresses skill issues—the Learning &Development staff uses them to identify newofferings. Network participation is instrumentalin building technical reputations and, therefore,indirectly in career development.

John Old, Texaco: We don’t have explicit discipline networks,unless they arise around some clear businesspurpose. The competence of the discipline is onlyindirectly identified as a separate business pur-pose. What we have in the upstream business isa network of resource advisers, a group of peoplewho are responsible for ensuring that people areaware of job openings where their expertise isneeded and that there is equal opportunity to getjobs throughout the company. This is a higherlevel network that helps people with assign-ments and technical competence.

Rodulfo Prieto, PDVSA:We also have two axes to address technicaldevelopment. In addition to the previously men-tioned communities of interest, we have a tech-nical resource department that is responsible fordeveloping people and assigning them to differ-ent projects. The leaders of each technical lineusually participate in one or more communitiesand promote forums on specific issues within the technical line. They also are responsible forclosing competency gaps as well as the develop-ment of all personnel. In exploration, we haveseven technical lines—geochemistry, seismic

76 Oilfield Review

Rodulfo Prieto, PDVSA:There are different types of communities:some advise; others make decisions.

Spring 2001 77

Jeff Stemke, Chevron: We want to leverage available technology, so it isimportant to have an established profile, or visi-bility, with those you are trying to help, especiallywhen vendors present new technologies that will“solve all their problems.” In large companies, ateam or community often hears about a new tooland wants to implement it. A better approach isto look at what a group is trying to accomplishand match technologies with those needs. If agroup knows you, or you know the members, youcan ensure that what's happening makes sense.

In our experience, many of the knowledge-management technologies are collaborative,Web-based tools like Lotus Domino orQuickPlace for discussions and document-shar-ing, and real-time application-sharing and video-conferencing tools like NetMeeting orSameTime. Expertise directories are a kind oftechnology-supported, Web-based solution. Indrilling, we worked with a consortium of compa-nies to produce Get Smart, which is a MicrosoftProject-based template for drilling. In terms ofrelative investment and participation in the pro-cess, I agree that the cost of tools should bemuch less than the “people cost.” It's moreimportant to focus on the business problem andthen match up technologies after a team decideswhat they are trying to do and how informationneeds to be transferred.

Lesley Chipperfield, Shell International E&P: It used to be that if we wanted global linking orthe potential for it, our greatest challenge wasavoiding fragmentation from local selection oftechnology. We ask groups to think about wherethey are going next and where they want to be inthe longer term to avoid isolation from othercommunities. We prefer the simplest tools withthe lowest possible entry level in terms of peopleusing them. The more features these tools have,the more difficult they are to use. We opt for sim-ple tools and standard products like SiteScapeand NetMeeting to keep collaboration as simpleas possible.

One of the interesting problems about technol-ogy is that the network moderators from our tech-nical disciplines have backgrounds in the businessunits rather than information technology (IT), but they often end up talking about knowledge-management tools and technology when we havemeetings about where the networks should gonext. There is always a huge effort to maintain abalance between putting our energy into lookingfor new tools and developing better features ormaking better use of what we have, and I think itwill always be that way. It has something to dowith being in a technical company where peoplelike to try out new technologies. My group seesminimizing fragmentation and increasing potentialfor future connectivity, even if it isn't obvioustoday, as key roles we can play.

interpretation, stratigraphy, integration, struc-tural geology, petrophysics and reservoir charac-terization—that come together because ofcommon interests. We probably need to inte-grate those communities into one.

Erik Åbø, Statoil:We’ve discussed the efforts of technical disci-plines, but we also have what we call processnetworks. Discipline advisors run technical net-works, but vice presidents run process networks.For example, there are several networks likeexploration and reservoir development, drillingand well technology, development concepts andprojects, project management, purchasing, andfinally, operations and maintenance. Historically,vice presidents had a central staff developmentfunction, but we now have totally free access tojobs. The networks do not have a formal deci-sion-making capacity, but facilitate sharing ofbest practices and make sure that the best peo-ple with the right competency levels for thefuture are in staff positions. Senior managers runthese networks.

Reid Smith, Schlumberger:Conventional wisdom is that you shouldn’t spendmore than a third of your knowledge-manage-ment budget on technology, but technology oftenseems to be the focus. Companies use a varietyof technologies such as e-mail, people finders,Yellow Pages, discussion forums, portal softwareand other collaboration tools. What are yourexperiences with knowledge-management toolsand technology and what has worked for you?

Lesley Chipperfield, Shell International E&P:Our experience is that we frequently have to per-suade new communities to address other issuesinstead of jumping into technologies before theneeds are clearly defined. Managers or peopleinvolved in networks often have a technology theywant to use. People hear of a new collaborationtool and want to implement it before the objec-tives are defined. Our strategy is to focus on peo-ple and processes first, enabling them with a smallrange of tools from e-mail to more elaborate col-laboration projects, tools and Web-site space.

> Communicating best practices: the Shell EP Newsletter. [Courtesy of Shell International E&P]

Jeff Stemke, Chevron: A few years ago, one of the things we imple-mented was a standard global desktop. Everyonenow has the same software and basically thesame hardware. This was primarily a cost-cuttinginitiative. We were supporting so many differentkinds of software and hardware that standardiz-ing saved about $40 to 50 million a year. We nowuse the same versions of word processing andspreadsheet software, so many of the file-sharing barriers have disappeared. An intangiblebenefit is that collaboration is now easier.

In the early 1990s, we had group softwarewars with one team using tool A and anotherteam using tool B. Now, we are developing thementality that common is good, but that attitudehasn’t completely taken hold. There isn’t a cor-porate collaboration standard yet, but we’reworking on one. To achieve truly global commu-nications, we need to make it easy to connectwith others inside or outside the company who

are using the same processes. Then, when a newtool comes along, you can compare it with theexisting infrastructure to look for added valuethat justifies its use.

Erik Åbø, Statoil: We switched to a common platform a couple ofyears ago. It was pain before gain, but when wesay common platform, it doesn't mean the samesupplier for all tools. A good supplier is able tosupply tools that link to any platform. Open stan-dards are important for the future. There is com-petition between service companies for theirplatform to be the industry standard, but evenlarger service suppliers should be willing to coop-erate with other companies to make sure that asa business we are operating on an open standard,regardless of subsurface tool constraints.

One of the things we did was to tell peoplethat this transition was going to hurt. We made

them aware that a difficult period was coming.Because they were unfamiliar with the newtools, some people thought the old tools werebetter and could do more. We told everyone thatswitching to one platform required tools that fitthat platform. After a time, geologists, geophysi-cists and engineers have become comfortablewith the new tools. We now are past the painand into the gain period.

John Old, Texaco: We spent about five years trying to get businessunits to agree on common E&P computing stan-dards. Finally, a senior upstream vice presidentsaid, “We’re just going to do it.” The other thingthat helped was getting some key geoscientistsin the company to testify that it really doesn’tmake much difference what tool you use. Theymay be different, and you have to learn them, butonce you do, it's easy to switch. Once that mes-sage started coming from credible people in theorganization, most of the resistance ended.

Reid Smith, Schlumberger: In a broad context, knowledge-management pro-jects seem to turn into electronic-, or e-, businessprojects over time. Everyone starts with procure-ment and commerce, but I'm including internal aswell as external e-Transactions with employees,partners and suppliers. There are many reasonsfor this. One is that over time, we realize that theknowledge we use comes from an extendedenterprise that includes not just one’s own com-pany, but also partners, suppliers and academiccollaborators at universities. The second reasonis that much of what we learn is of value to ourcustomers and suppliers as well as our ownemployees. The third reason, which is probablythe most practical, is that it is expensive to con-struct knowledge assets in communities and net-works. By making these assets multipurpose, werealize a better return on the investments madeto construct them.

Some of your companies, Chevron for example, are very advanced in e-Business. Areyour e-Business, e-Commerce and knowledge-management efforts linked? If not, should they be?

Jeff Stemke, Chevron:Our CIO would not differentiate e-Business fromknowledge management because he says that e-Business is the only business. Instead, we talkabout being Internet ready. Basically, our busi-nesses need to work globally to effectivelyenable us to connect with partners, suppliers andcustomers. A part of that effort encompasses tra-ditional e-Commerce and e-Business; another

78 Oilfield Review

…creating a corporate culture where ourpeople become the main driver, so that infor-mation is captured, documented, shared,applied and even renewed in order for theright person to use it at the right time—Rodulfo Prieto, PDVSA.

…creating a new working environmentwhere knowledge and experience can easilybe shared, where individuals apply collectiveknowledge to make optimal decisions in realtime—Reid Smith, Schlumberger.

…a knowledge architecture, or processstructure. The key elements are processes,technology and behaviors that deliver theright content to the right people at the righttime in the right context, so they canquickly solve problems, exploit businessopportunities, accelerate competency andinnovation, and make the best decisions—Jeff Stemke, Chevron.

…creating an environment for sharing infor-mation and practices to attain corporate busi-ness goals—Erik Åbø, Statoil.

…connecting people so they can share exper-tise globally—making the best resourcesaccessible to each business opportunity atthe right time, regardless of organizational orgeographic location—Lesley Chipperfield,Shell International E&P.

…about keeping track of those who know therecipe, and nurturing the culture and thetechnology that will get them talking. It’s notabout creating an encyclopedia that captureseverything that anyone ever knew (ArianWard, Work Frontiers International)—ChrisMottershead, BP.

…helping people effortlessly connect to theindividuals and artifacts they need to achievetheir personal and business objectives—John Old, Texaco.

Knowledge Management Is…

Spring 2001 79

part includes collaboration through knowledgemanagement. In that sense, making the companyeffective using the Web for all the things thatneed to be done, at least where practical, is thedirection we’re going.

There’s definitely aspects of e-Business andknowledge management to the Chevron BusinessElectronic Support Tool (C-BEST), which com-bines our on-line lubricant advisory recommenda-tion system with a mechanism for customers tomonitor financial transactions. This tool helpscustomers select a product or recommend a com-petitive alternative to what they might already beusing in their operation.

The Chevron Retailers Alliance (CRA) is a net-work of 8000 service stations, which generallyare independently owned. Because many haveconvenience stores, we had the idea of connect-ing them as a cooperative buying group to reducemerchandise cost. The knowledge-managementaspect is actually a help desk that allows service-station owners to call in and raise issues or askquestions. The CRA group knows where to go foranswers and makes sure things get done. Theydon’t call it knowledge management, but it fitsthe definition.

Lesley Chipperfield, Shell International E&P:You could say that e-Business is the only businessbecause almost everything is done electronically.Our e-Business organization is separate from theIT organization and my group. But because of itsstate of maturity, that’s actually quite healthy. It’sanybody’s call at the moment, but we do havelinks. In fact, a member of my group is working onthe e-Business team, so we’re using knowledge-sharing as an e-Business driver.

We will use our experience with knowledgemanagement and best practices to work with thee-Business team on a people-to-people basisrather than trying to link the organizations, which is always a struggle that leads to the soft-ware-related wars that we talked about. That’sthe direction we’re taking, and I fully expectknowledge management to be a key part of anye-Initiatives.

Rodulfo Prieto, PDVSA:Like many modern corporations, PDVSA uses e-Business extensively throughout the organiza-tion. The Bariven affiliate of PDVSA uses e-Commerce to integrate business processes andtechnology to eliminate internal and external bar-riers, and create added value by taking advan-tage of market opportunities. Our internationalmarketing group also applies e-Business intrades by substituting Internet transactions for

telephone calls and faxes. For e-Learning, thePDVSA Educational Center conducts 20% of alltraining on the Internet. We used e-Business toshare data with private investors during theopening up of the Venezuelan oil industry.

The e-Transaction is also important internally.We are using e-Business with software applica-tions that are controlled directly by anotherapplication to establish user times and chargesin real time. We also have contracts that peoplesign to run projects, which are documented by e-Transactions. If you assign a person to a projector you’re going to be doing something specific fora project, you execute an e-Transaction and signit electronically, which is a different level of e-Business that we use.

Erik Åbø, Statoil: We have a project under way on e-Business or e-Commerce, e-Collaboration and e-Learning.First, you have e-Commerce, the contractual partand daily business of buying or selling. Thenthere’s collaboration through our extranet wheredifferent suppliers and partners are involved. It’simportant for our suppliers to participate in inter-nal discussions on the extranet where there areprivate chat rooms for different organizations toaccess. If we have a contract or technical projectwith Schlumberger, Halliburton people can’t gointo that chat room. There’s a high grade of secu-rity and strict regulations in terms of firewallprinciples. The other part is e-Learning, which isbasically internal.

In addition, about three years ago, we had aninternal e-Learning process called the IT step.Everyone in the company got a personal com-puter at home. General training was done in theirspare time using CD-ROMs, which saved timeand probably offset the cost of the computers.We’re going to take this concept further becausee-Learning will be very important in the next ITstep within Statoil.

Chris Mottershead, BP:Last summer, BP and Shell collaborated on e-Learning. What Shell is trying to achieve withe-Learning is impressive. It is not based on reduc-ing training costs and shared practices of themoment, but rather on building a competentworkforce for the future when that workforce isincreasingly global, with little likelihood of everbeing centralized again. Shell is addressing thisin terms of how to produce a skilled organizationfor the next decade. By contrast, much of whatwe’ve talked about in knowledge management ishow to respond to immediate needs.

It struck me that we have high-level, collabo-rative tools, but they’re all about driving effi-ciency. What you actually want to do is improvethe quality of person-to-person relationships. Weset targets of having 99% of procurement trans-actions automated by the end of 2000 and aredisappointed that we only got to 97%. We arevictims because we missed by two percent.Shell’s approach shows a great deal of insightabout the things that should be managed.

Reid Smith, Schlumberger: What successes have you had? What were the stumbling blocks and barriers? What are themetrics that you use to measure knowledge-management progress?

Jeff Stemke, Chevron:The metric that made the most impact is bottom-line savings. Our goal was to reduce operatingcosts. In the past eight years, we have reducedoperating costs by over $2.5 billion a year, in part through the successes of some of our earlybreakthrough best-practice-sharing knowledge-management projects. Knowledge managementplayed a role in putting better processes in placeto operate refineries, manage energy and run cap-ital projects. The connection between knowledgemanagement and cost savings is tenuous. Whileit is difficult to make a direct connection, thesekinds of results helped validate knowledge man-agement to the point that perhaps detailed costjustification isn’t as important for new initiatives.

On the other hand, what I see in our commu-nities is that without some metric that connectsto the business, you may not get the results youwant. You can, for example, use metrics for anumber of different things. In a problem-solvingcommunity of practice, the metric could be thenumber of problems resolved or the number ofresponses to questions. If your goal is to developand deploy best practices, what is the evidencethat you’re reusing them? We may struggle withthe metrics, but it’s clear that you’ve got to havesomething that connects closely to the businessobjectives you’re trying to achieve.

Lesley Chipperfield, Shell International E&P:To some extent, we’ve been able to move awayfrom hard-dollar measures for knowledge-management activities over the last couple ofyears. Initially, we had to provide quantitativeproof of value, but now we’ve moved more towardvalue propositions, or proposals, which can havequalitative elements. Last year, we reviewed allthe so-called new ways of working as part of ourpursuit of excellence program. Along with the

consultants in our global networks, we did a for-mal value review through interviews with peoplewho collect data. Interestingly, knowledge man-agement saved more than $100 million a year.

That feedback gave our leadership a feelingof comfort, and it also gave our group a greatdeal of confidence. However, what we foundmost useful were the best practices. We devel-oped these success stories called “Stories fromthe Edge” for distribution with our internal tech-nical magazine the EP Newsletter, which is circulated to 12,000 staff members worldwide toraise awareness of what is going on in the busi-ness. That was enormously helpful and we gottremendously positive feedback.

That publication was distributed almost ayear ago, and now the second installment subti-tled “Our New Ways of Working” is ready. Itincludes new stories and corollaries to previousbest practices. We found this to be extremelyeffective in terms of persuading doubters. Ourmanagers are among the best advocates of whatwe’re doing in knowledge management and usemany of these stories in internal and externalpresentations. These stories also are used inother publications and presentations, which illus-trates that our commitment is paying off and ishelping to legitimize knowledge managementwithin the company.

Chris Mottershead, BP: If you want to be successful, then the way yousay it has to be authentic and true to what yousay. If it’s not, people become uncomfortable. Itseems to me that, in a sense, what we’re doingis sharing knowledge about sharing knowledge,but we’re doing it in a way that is perfectly con-sistent with the knowledge that’s already in peo-ple’s heads. Therefore, what we’re actually doingis managing knowledge management, engagingsomeone to consider and understand what’s pos-sible, rather than simply giving them data.They’re defining what knowledge managementmeans by being examples themselves.

That doesn’t negate the need for KPIs, but KPIsmay not be the justification that people think theyare, as much as something that brings alignmentwith activities and objectives. You don’t get align-ment from case studies. What you get is knowl-edge and understanding, so you need both. I agreethat networks need KPIs because that’s the metriceveryone understands right now. For example, ifwe want to reduce lifting costs by 5% this year,that’s the objective. And I now know whether I’vegot a contribution to make in that area. But justsaying we’ve got a target to shoot for doesn’t helpeveryone engage and collaborate about what canbe done to meet the objective. Best practices andsuccess stories are needed for that.

Rodulfo Prieto, PDVSA:We have included intangibles like people’s atti-tudes in our scorecard by developing a way tomeasure what we call "Attitudinal Base forChange" (ABC) and establishing metrics for 12 to14 items related to intangibles such as how people feel about being connected and workingwith each other. This effort to gather informationon knowledge-sharing and knowledge-teachingwas started about three years ago. We submittedthe data for an independent evaluation, and sofar the results have been helpful and encourag-ing. Our knowledge-sharing results have shownimprovement during the past two years.Leadership, accountability, sharing knowledgeand orientation to excellence are some of theaspects that are being evaluated.

Reid Smith, Schlumberger: This year, we made knowledge-sharing part ofour annual performance reviews for the firsttime. We thought about adding it three yearsago, but decided against having another head-quarters initiative. We knew it would ultimatelybe necessary to recognize knowledge-sharing inthe performance-management process, butwaited until our field engineers, particularlythose associated with the InTouch system,

suggested it. They wanted knowledge-sharing onthe appraisal form along with the other indica-tors that they’re measured against. They definedit and wrote up the description, which made itmuch easier to initiate. As a result of this rolloutin Oilfield Services, most Schlumberger fieldpersonnel have objectives related to best prac-tices, lessons learned and other aspects ofknowledge management.

Intellectual capital has implications related tohow our companies go forward, working in pairsor in industry groups. Is intellectual capital val-ued inside and outside of your companies? Whatpossibilities exist for knowledge-sharing acrosscompanies and across the industry?

Lesley Chipperfield, Shell International E&P: If you go back five years or even 18 months, wedidn’t speak with competitors on a meaningfullevel. Then ironically, with the recognition thatintellectual capital is tremendously valuable, westarted being more open. We used to treat every-thing like the “crown jewels,” but now, we’rebeginning to differentiate in terms of what wecan learn through collaboration with other com-panies and assign value only to the truly impor-tant factors in performance. If, for example, wewant to share information about implementingan accounting system, we break it down to abase level and compare experience betweencompanies. No one is concerned about this typeof knowledge interaction anymore.

John Old, Texaco:Establishing metrics to justify intangible intellec-tual capital and knowledge assets provides inter-nal value for a company. There’s also value instarting to think of people as an opportunity toinvest in intellectual capital, rather than as a cost.

Chris Mottershead, BP: We need to do what you’re describing, but weneed to do it because we value people and wantto see how we’re adding value to our people.Considering the staff to be intellectual assetsand investing in them is at the core of Shell’s e-Learning process, which I think is a correct

80 Oilfield Review

Reid Smith, Schlumberger: This year, we made knowledge-sharing part of our annual performance reviews for the first time.

Spring 2001 81

approach. You end up with KPIs from the oldparadigm, and you do need KPIs, but they need torespect what we’re investing in. Where peopleare concerned, you hear proposals about measur-ing things like number of patents created, butthis type of metric would be a spurious measureif you were actually trying to build the compe-tence of your organization for the future. Thedanger is that if the stock price doesn’t move,everyone stops emphasizing the value of intellec-tual capital. What we need to do is to continue toinvest in marketing knowledge management.There needs to be an external imperative in orderto get companies to innovate freely.

Lesley Chipperfield, Shell International E&P:There’s an ongoing challenge in picking the rightperformance indicators and the right mix ofdevelopment costs, production costs and otherindicators that are more difficult to measure.Companies are developing training metrics andtrying to measure the investments they make inpeople. For the first time, we have an element onour corporate scorecard that has to do with people and includes things like personal develop-ment plans as well as learning opportunities. Atthis stage, there are a number of considerations,but they’re primarily soft, not hard, issues. We’renot asking about the training budget, and we’renot looking for organizational or mechanicalthings. Therefore, these measurements must bedone through structured discussions to collectinformation that’s fair and will actually influenceoverall business performance.

Jeff Stemke, Chevron:We also should consider quantifying intellectualassets as motivation to move more toward man-aging our companies in a way that stresses theimportance of people. Even though we’re decen-tralized, there are opportunities in big companiesfor integrating knowledge-management pro-cesses across the enterprise, but business-unitbarriers sometimes make this difficult.

John Old, Texaco:Companies tend to have systems that mandatedoing X or Y within a certain time frame and bud-get. We don’t give people freedom to go out andlook for innovative business approaches. Whensomebody does create something unique or inno-vative, it doesn’t flow throughout the organiza-tion. Systematic knowledge management, doneproperly, enables knowledge to emerge and flowto the right people within an organization at theright time, so that they can act more efficientlyand effectively.

Reid Smith, Schlumberger: Looking ahead, I’m interested in where you think knowledge management stands today andwhere you think it is going (see “KnowledgeManagement Is …,” page 78 ). What challengesand opportunities do you see for the future?

Lesley Chipperfield, Shell International E&P:Of the elements mentioned, the biggest is inno-vation and creativity through diversity, bringing inpeople who aren’t traditionally at the table togenerate new ideas. This is something we’rebeginning to see, but still within the traditionallinks between disciplines. When you bring non-E&P people with fresh business ideas into themix, it generates new opportunities. You see thisin some of the e-Initiatives. The speed at whichwe roll out new technologies will continue toincrease. That was one of the factors that got usinto knowledge management in the first place,and it will never disappear entirely.

In terms of attracting people to the company,another factor is individual expectations. Newemployees come in with different expectationsand ways of working. When we talk about attrac-tive careers and what individuals expect, it is nolonger enough to invite them to join Shell and seethe world until they retire. Many people don’twant that. They want something different. Tobring in new people, we have to respond by giving them satisfying jobs and letting themdevelop and flourish during their careers.

The question of affiliation, or where peoplefind their identities within a company, is also animportant one. And in this new knowledge-management world, we must come to grips withthat. It’s less important for new hires, but theaffiliation of people who’ve been in the companyfor many years tends to be with a local companyand discipline. However, as people move fromone asset team to another, you start to see thataffiliation has more to do with the project orasset that they’re on at the time.

This is a cultural shift that we have to man-age. New employees seem to handle this betterthan those who’ve been around longer and feelcomfortable with their established roots. Newpeople feel freer within the organization. Peopleneed to be able to join in where their expertise isbest deployed. You really start getting to the bot-tom line when people are working where theycan uniquely add value.

Lesley Chipperfield, ShellInternational E&P: If you go back five years or even 18 months, we didn’t speak withcompetitors on a meaningful level.Then ironically, with the recognitionthat intellectual capital is tremen-dously valuable, we started beingmore open.

Jeff Stemke, Chevron:Accelerated deployment of innovative ideas is animportant business driver, and technology com-panies need to do it more effectively. The key tofuture success is for the members of technicaland process networks who understand businessproblems and technological solutions to makequicker connections. We all have great ideas, sothe sooner we deploy them, the more benefitsand savings we achieve. This is an area whereknowledge management plays a big role. Weunderstand mechanically how we make connec-tions and can get that to happen, which is moreof a here-and-now aspect. The future is abouthow to improve knowledge management throughcreativity and innovation. In other words, how toachieve broader coverage. Experts discover newthings and turn them into products, but how canwe increase the number of people who are effec-tively involved in that creative process?

Looking ahead, I also see knowledge man-agement in terms of five components: people,processes, behavior, technology and content.People are going to change during the nextdecade as we move toward more virtual teamswhere people in different companies worktogether. But we’ll probably go beyond that tohave independent agents linked together for pro-jects. These people may work for different com-panies or be self-employed knowledge agents.How we connect them is one challenge. There

are more critical aspects that I don’t haveanswers for. How do we manage this collectionof individuals, whether they’re in different com-panies or acting as individual agents? How do weretain the allegiances that we now have on inter-nal projects? And how do we motivate peopleand achieve results when they aren’t all workingwithin the company?

John Old, Texaco: The strength of knowledge management lies notonly in how broad the networks are, but also intheir diversity. We have a deep-rooted mechani-cal view of organizations that inhibits us. We talkabout change, but commitments to rigid perfor-mance-management systems, expense budgetsand capital-expenditure programs keep compa-nies from changing. Although these systems aredesigned to ensure stability, companies are goingto take a hard look at these systems. With a moreorganic approach, you get a clear view of a com-pany’s purpose and principles. For example, theopen job market is an organic approach.

When you let people migrate to where theywant to work, it is incumbent on companies andprojects to be as attractive as possible so peoplewant to work there. It’s attractive for newrecruits to see a place where they can potentiallyachieve anything they want. There’s a perceptionthat big companies are not innovative and even-

tually must turn to alliances with smaller compa-nies for innovation. But in large companies, likethe ones represented here, there’s a lot ofuntapped potential because our mechanicalstructures put people in boxes and keep themthere. In the near future, companies will recog-nize the power in their employees’ minds andallow them more freedom to innovate.

No one is going to get out of bed in the morn-ing and charge into work to improve the return oncapital employed, but they will charge into workto deliver light, comfort and mobility to people.That’s not overly prescriptive and gives peoplefreedom to let their imagination roam, which can take companies in new and unpredicteddirections. That’s the type of concept that some-one will latch onto in the future.

Erik Åbø, Statoil: In the future, key performance indicators will stillbe a driving force, but reducing finding costs bydecreasing risk is also important, as is reducinglifting costs. By addressing production and effi-ciency through proper knowledge management,we will focus on KPIs and create an environmentthat puts people as well as competence in per-spective. Everyone should be able to use theoverall knowledge of his or her company. Themost difficult task will continue to be managingthe volume of information. However, it shouldn’t

82 Oilfield Review

Jeff Stemke, Chevron: The key to future success is for themembers of technical and process net-works who understand business problemsand technological solutions to makequicker connections.

Erik Åbø, Statoil: By addressing production and efficiencythrough proper knowledge management, we will focus on key performance indicatorsand create an environment that puts peopleas well as competence in perspective.

Spring 2001 83

be necessary to find and access all the availabledata just to do your job efficiently, improve per-formance and contribute in key areas. Knowledgemanagement is how we deal with that problemand make sure that people have access to theright information.

Lesley Chipperfield, Shell International E&P: The idea of everyone having some stake in theKPI is extremely powerful. It starts to enable thescenarios you’re painting. If you don’t have a KPI,or metric, then you have potential for chaosbecause people are pursuing what interests themrather than what’s in the company’s best interest.Our networks, for example, now have a score-card, which is effectively the KPI. They’re not per-fect yet, but they’re evolving to truly meaningfulmeasurements. However, this concept of a busi-ness objective that may not necessarily have adirect link with that network would be anextremely powerful way of driving the behaviorsthat you need and getting everyone movingtoward the same goal, maybe at different speedsand from different directions.

Chris Mottershead, BP: The industry is radically different from what any-one would have predicted ten years ago. In fact,our conversation about e-Learning represents anopen forum that you couldn’t have imagined even12 months ago. Relationships are no longeradversarial, which seems to be a signpost for the

future. However, it’s unwise to extrapolate ormake linear projections from this point in time.

Future communities of practice will knockdown old boundaries. Companies with communi-ties that can come together to satisfy the differ-ent needs of an organization will be successful.We don’t know what 2011 will be like, but it willprobably have communities of practice that areradically different and more open. What we’vetalked about are the basic foundations and build-ing blocks of a new world where winning compa-nies will recognize the need for diversity andrealize that they must integrate all of theseknowledge-management components. But, thenhow do you measure performance against clearobjectives with metrics?

We’ve spent 10 years building an industryaround key performance indicators. If lifting costis $3 per barrel, but it needs to be $2.90 per bar-rel next year, that’s the KPI. Therefore, much ofthe current knowledge-management technologygets deployed to meet KPI objectives. This willcontinue because it’s a necessary part of deliver-ing performance. However, companies will say,“We have all the necessary components, so wemust be open, knock down firewalls and inte-grate knowledge management in a way thatbrings clarity, focus and direction, but doesn’toverprescribe.” We will have to continue whatwe’re already doing inside our companies and, attimes, across the industry. That will push the limits of knowledge management.

Rodulfo Prieto, PDVSA:The speed at which we make adaptations isgoing to be very important. We mentioned fasterdeployment, but faster input also is going to be akey factor. How quickly we put the piecestogether and adapt them into our companies willmake a difference in the future. That’s where I believe knowledge management is headed. Ienvision a world with no firewalls. Competitionwill not be over data, but over ways to use datafaster and develop solutions more quickly.

Reid Smith, Schlumberger: Some people say we have to focus on the future;we have to know what the standards will be.Others say knowledge management is changingso fast that we shouldn’t try to look ahead andthink about the future. Knowledge managementis changing quickly, and while predictions may bewrong, it’s important to project and plan ahead tobe better prepared to react to whatever comes up.“Too fast to follow” is a tag line that Shell uses.

Lesley Chipperfield, Shell International E&P:Yes, but we have to go beyond reacting. If knowl-edge management is too fast to follow, we mustspot opportunities rather than react to whatevertechnological solution or answer has recentlycome along. It’s important to seize knowledge-management opportunities. —MET

Chris Mottershead, BP: What we’ve talked about are the basic foun-dations and building blocks of a new worldwhere winning companies will recognize theneed for diversity and realize that they mustintegrate all of these knowledge-managementcomponents.