The New Intelligence: The Birth of the Knowledge Management Industry

The introduction of computers led to an unmanageable proliferation of data, which stimulated the birth of knowledge management (KM). To understand KM and all of its components (i.e., business intelligence, content management, etc.), it is necessary to first discuss the precursors to KM.

1. Buried in Information

Magicians, who are masters of optical illusions, convince breathless audiences that their tricks work by magic. But the tricks really result from precision sleight of hand, split-second timing, and careful planning. And Merlin's code prevents the brotherhood of magicians from ever whispering its secrets. Knowledge is all that separates the knowing from the unknowing, the magician from the audience. Knowledge is elusive (especially in large organizations), inconsistent, unripe, and buried in information.

Pick up a copy of The New York Times. There is more information contained between the front and back pages than we can possibly digest. Add to this the other 99 papers we will read this year, the 3000 notices or forms that we will read or complete, the 2463 hours of television we will watch, the 730 hours of radio we will listen to, and you have something called an information explosion.

According to Linda Costigan Lederman (1986), who devised these statistics for her piece "Communication in the Workplace," this does not even take into account the number of hours spent exchanging information in conversations. And should we not include information that is signaled by nonverbal means, such as the wink of an eye, a firm handshake, or a nod of the head.

Soothsayers predict that the amount of information that we are expected to absorb will double every four to five years. Even now, more information has been generated for mass distribution in the last three decades than in the previous five thousand years.

Maybe it is more than an information explosion; perhaps it is more like a glut. And with this glut comes the breakdown of our ability to mess with or even retrieve the information we so labor to possess. Akio Morita, former chairman of Sony Corporation, believes that our capacity to retrieve this information is declining. In fact, he believes that out of all the information that we absorb, we can retrieve from our memories only a paltry five percent.

Alvin Toffler, in his much-acclaimed book Future Shock (1984), paints an even bleaker picture. He writes of an actual breakdown of human performance under these extraordinary information loads and demonstrates its relationship to psychopathology. Work is increasingly being done in one's head rather than at the desk, as we try to cope with managing this massive information overdose. The information that we must assimilate has become more abstract as technological innovators find new, clever ways to present it. And we do not just process one stream of data bits at a time. Some researchers refer to our need to deal with more than one information flow at a time as polyphasic activity. Visualize Sam riding his convertible to his next meeting. He grips the steering wheel in his left hand, a cellular phone in his right, and on his lap rests his miniature tape recorder. And in the passenger seat is his UltraLite computer.

This increasingly large information flow is forcing people to adapt to mastering and making judgments about it in shorter periods of time. You can compare this burgeoning quantity of information to an algae-infested pond that no longer has enough oxygen to support fish. We may actually be pushing to extreme limits the physical ability of people to process information. When that begins to happen, great amounts of information just pass by, which people cannot evaluate.

Born of this glut is a new phrase that all agree has distinct meaning. Information anxiety, a nice turn of phrase coined by Richard Saul Wurman (2000) in his book of the same name, is that chasm between what we think we should understand and what we really do understand. As those papers and magazines and books pile up unread on our nightstands, we grow increasingly uneasy at our inability to keep pace. And, exhausted, we fall asleep with yet another unfinished book by our side. Information anxiety, according to Wurman, "is the hole between data and knowledge. It happens when information does not tell us what we want or need to know."

2. From Batch to Decision Support

In the 21st century, we might achieve technological nirvana. For some, this might be the fusion of human and machine. A great debate has been raging on the technological, political, and even moral battlegrounds over the development of what we rather prosaically call an artilect.

An artilect would be a machine that exhibits intelligence (artificial) to the degree of surpassing its creators one day. These ultraintelligent machines inspire fear, loathing, and a great deal of fascination in the scientific community as well as in the general public. Could we bear living in a world "peopled by creations massively smarter than we are?" But we jump ahead of ourselves. There are no artilects to loathe or love; there is, however, the possibility of creating computer systems that are just about as bright as we are in narrower spheres of expertise.

The key is to create computer systems that present less but more significant or germane data. Several methods can be used to present, to the end user, the best mix of the detail with the composite data. All of these methods recognize a relationship between the type of decision maker and the detail presented.

2.1 Types of Decision Makers
There are three types of decision makers. At the bottom of the corporate hierarchy are the paper pushers. These are the people who do need to see all the data. They are the check processors, complaint takers, order takers, and customer service staff. These technical users need to input and review a wealth of data. Given the nature of the task, this is usually done in a rote fashion and not subject to complex decision making. At the other end of the spectrum are the organization's senior managers who use the information, gathered at the lower rungs, for strategic purposes.

A whole range of vendor-sold EIS (executive information systems) are now in vogue. These sport flashy colors, touch screens, and sparse information displays. Data displayed here would most likely be sales projections, profitability numbers, and key competitive data. In the middle is (you guessed it) the middle manager or tactical user. In organizational terms, these are the professionals who are right in the firing line and need the most careful balance of data. These are the poor unfortunates who sit buried under that great avalanche of information.

Companies have always collected data and stored it in diverse corporate databases. Automobile manufacturers keep massive amounts of data on file, concerning their suppliers and dealership locations. For a long time, no one attached any strategic value to this mass of detail, that is, until Chrysler fell into the black pit of insolvency. Lee Iacocca tweaked this very set of databases to save Chrysler from a catastrophe much worse than rust. By moving to a different level of information analysis, Iacocca was able to convince members of Congress to support the loans that injected life into the veins of his company.

2.2 Filtering
There is a host of filtering methodologies for serving up this sort of strategic knowledge on a silver platter. For the most part, they are classified in three different ways. The monitoring method serves up data to the user on an exception basis. This can be variance reporting, in which the system produces only exceptions based on programmatic review of the data, for example, to review credit card payments and display only those accounts in which the payment was not received or the payment is below the minimum payment.

The advent of fourth-generation languages (4GL), tools enabling the end user to access the corporate database with easy-to-use query syntax, has thrust the interrogative method of system tailoring to the popular forefront. This method takes into account the many occasions when the user cannot identify the set of information necessary to handle the day-to-day, ad hoc analyses required in complex decision-making environments. In these cases, all of the data elements need to be resident in an accessible database. And a tool needs to be in place to permit the user to easily and quickly develop queries and variations on these queries against this data.

When Banker's Trust, now part of Deutsche Bank, decided to get out of the retail business in the early 1980s, the data-processing effort to achieve this feat was enormous. One area that Banker's spun off rather quickly was the credit card subsidiary. The 4GL in use at that time was FOCUS. The users and the accounting staff used this tool to great advantage to ensure a smooth transition of accounts to the many final resting places. Some accounts were spun off to a bank in Chicago, some to Albany, whereas the high-rollers stayed behind in newly minted, privileged accounts.

A model-oriented approach comes in many flavors. Human resource or facilities departments are good candidates for descriptive models, which can be organization charts or floor plans. On the other hand, a normative representation of data is a good fit for budgeting when the goal is to provide the best answer to a given problem. Economic projections are good targets for modeling methodologies that have the ability to handle uncertainty. The operations management students (among us) gleefully apply game theory to those problems in which there is a requirement to find the best solution in spite of a profound lack of information. An example of a problem that would use this type of strategy would be a competitive marketing system, in which data about the competition is scant or unknown.

2.3 Transforming Information into Knowledge
Perhaps the industry with the largest share of traditional systems is the banking industry. No one personifies banking more, or has done more for the banking industry, than Citicorp's former chairman emeritus, Walter Wriston. In his 17 years as CEO, he revolutionized the international banking environment, in which the interface between humans and machines permits easy access to complex information. Wriston predicted that artificial intelligence would become the norm and not the exception. He looks forward to a day when he can walk up to an expert system in a bank lobby that will be able to answer complex questions about his account. "Can I invest in a tax-free fund? Can I do brokerage transactions?"

Quaker Oats in Chicago was one of the first consumer-goods marketers to realize the potential of strategic information; i.e., knowledge. Several decades ago, it innovated a computer system to analyze some two billion facts about different products and competitors. The use of this system permitted Quaker Oats to understand the data and draw insights from it. This led them to the number one spot in product categories such as Rice-A-Roni and the ever-popular Aunt Jemima Pancakes.

Filtering is a mainstay in the marketing arena. For years, marketers have used smart software to filter relevancies out of the information glut. It all started with the first totally computerized census back in 1970. The Census Bureau recorded demographic data on computer tapes. This provided a plethora of information right down to the city block. By the time of the 1980 census, these stats had ballooned into 300,000 pages of statistics. And a whopping ten times that amount sat patiently on computer tapes. Today, this information is available on a desktop.

Many industries quickly followed suit. An investment service gathered data on some 5000 companies and offered this data along with smart filtering software. Individual.com started by sifting through full-text articles and pinpointing items of interest to its subscribers. Their take on their business is that they are operating an information refinery that takes a broad stream of raw data and turns it into actionable knowledge. Dean LeBaron agrees with this approach. He was very much in the avant-garde in the mid-1970s. That is when he preached the use of computers to improve the quality of investing. Batterymarch Financial Management is one of Boston's leading money management firms with a portfolio of over $11 billion. LeBaron runs Batterymarch as one large expert system. It is designed to operate the way an intelligent institutional investor would operate, if put on silicon substrate.

One of the most interesting expert-system success stories involved Campbell Soup. A senior technician was going to retire. Unfortunately, his expertise was going to retire as well. This fellow had decades worth of experience in determining where the problems were in the vast cookers that Campbell Soup used to make chicken soup. A few bubbles on the top meant one thing. A certain smell meant something else. Before he retired, Campbell Soup invested time and money in knowledge-engineering his expertise so that it could be stored in a knowledge base and used by his less-experienced successors.

4. Data, Information, Knowledge, and Wisdom, and Why We Need to Do This

The goal of KM is to turn raw data into knowledge, if not wisdom, as shown in Figure 1.6. That it is important to turn all of this raw data into knowledge is without question. Measuring the value of intellectual assets to ascertain the true value of an organization's future earning potential is almost turning into a field of its own.

Bruce P. Mehlman (2002), assistant secretary for technology policy for the U.S. Department of Commerce, gave a speech on intellectual property in the age of innovation at the Licensing Executives Society Spring Meeting in Washington, D.C. Mehlman stressed that the wealth of nations is indeed changing. The impact of innovation and technology on our society is already profound and unmistakable. Just look at the outsized impacts of information industries. The information technology sector accounts for just 7 percent of all businesses in our economy. Yet, between 1996 and 2000, it drove 28 percent of the overall U.S. real economic growth and created jobs at twice the pace of other sectors, jobs that paid twice as much on average.

The growing importance of knowledge and innovation presents both good and bad news for the United States, relative to its global competitors. On the one hand, by almost any measure, America is the most innovative nation on earth.

• The United States generates the most patents per capita.
• The United States conducts more research and development than any other nation. The United States finances 44 percent of the total worldwide investment in R&D, which is equal to the combined total of Japan, the United Kingdom, Canada, France, Germany, and Italy.
• The U.S. workforce is more research intensive than other regions. Researchers represent only 5.3 percent of the overall workforce in Europe, as compared to 8.1 percent in the United States.
• As measured by scientific publications, American scientific output exceeds those of the European Union and Japan (708 to 613 and 498, respectively) per million people.
• U.S. labs and universities remain a more attractive destination for the best and brightest young minds in the world. Of the PhDs who come here from China, 85 percent remain in the United States because it is a better place to do business. By contrast, many EU nations remain challenged when trying to attract top scientists and students.
• And perhaps most significantly, Americans have enjoyed the most rational, predictable, and consistent framework for intellectual property rights in the world, encouraging investment and rewarding innovation.

Notwithstanding the advantages and current leadership, the rest of the world is not blind to the importance of innovative capacity in the 21st century, and they are not standing still. America's global competitiveness faces pressure on multiple fronts including:

• Education: American students at the K-12 level continue to fall behind their international counterparts in math and science learning. U.S. eighth graders ranked 19th out of 38 nations in math and 18th in science in the 1999 Third International Math and Science Study.
• Purchasing power parity: When the CEO of an American multinational corporation was asked why they were moving so many R&D operations off shore, he replied that it cost 90 percent less to develop a PhD in Russia than in the United States. Just as manufacturing jobs have moved steadily abroad, innovation work may continue to globalize as highly skilled foreign labor proves cheaper.
• Global R&D trends: Although the United States accounts for 44 percent of worldwide R&D today, we accounted for 70 percent in 1970 (Alliance for Science & Technology Research in America). The European Union is racing to match our investments in nanotechnology, whereas Asian nations have collectively pulled ahead.

Our society, therefore, is facing threats from outside (global competition) as well as inside (failure of the educational system from middle school on up through college). Industry has no choice but to invest in systems that make us smarter.

5. Embedding Knowledge Management in the Organization

According to Pollard (2005) the expectations for KM are that it will be able to improve the following:

1. Growth and innovation
2. Productivity and efficiency reflected in cost savings
3. Customer relationships
4. Employee learning, satisfaction, and retention
5. Management decision making

KM can meet these goals if it is embedded within the organization using a bottom-up approach, rather than a top-down approach. Top-down approaches are usually forced upon employees and, hence, resisted or at least isolated. The bottom-up approach is somewhat akin to viral marketing, where you get one person enthusiastic about a product or service, who tells someone, who in turn tells someone else. By providing the tools, methodologies, training, and support on a unit or departmental level, you encourage employees to capture, share, and archive their knowledge for the good of the organization.

Of course, KM needs to have a focus. Pollard found that KM's safe haven seems to be the information technology department. This is quite a natural fit given KM's newly refocused definition, which includes topics such as organizational learning, technology transfer, competitive intelligence, data warehousing and business intelligence, document management (Davenport and Prusak, 2003), and its dependence on information technology resources.

Pollard suggests a number of techniques such as the following, to disseminate KM practices:

1. Do not force people to adapt. They must be self-motivated.
2. Change the job of knowledge professionals. In other words, get rid of those KM departments, and enable others to carry on these tasks; e.g., analysts, database administrators, and librarians can all go out and assist others to more effectively manage their information warehouses.
3. Stop collecting data centrally in a massive knowledge base. There is no reason why employees cannot store their domain of knowledge in their own private databases. Pollard makes a good point about respecting the privacy and confidentiality of personal information.

   This was quite a problem during the initial entry of artificial intelligence into the business world. People do not like to share what gives them their own personal competitive edge. For example, many financial investment houses never did get cooperation from their highly paid traders to build expert systems. After all, why should a trader give away what made him or her valuable to the firm? Of course, my own take on this is that a central KM repository needs to continue to exist, with the personal KM repositories used as inputs as well as outputs to the central store, where applicable.

4. Help people connect to experts inside and outside the organization.

The current craze over balanced scorecard (Kaplan and Norton, 1996) and performance management and measurement might also be used as a lever to further embed KM within an organization. Balanced scorecard has four perspectives, as shown in Figure 1.7. Each one of these perspectives defines a set of objectives, measures, targets, and initiatives to achieve the goals of that perspective. Although the "learning and growth" perspective is a natural fit for KM, the remaining perspectives should also be considered. Balanced scorecard is best utilized when, similar to KM itself, it is cascaded through the organization. In other words, each unit or division should create its own scorecard, although it should tie in with the goals and objectives of the organizational scorecard. Adding goals, metrics, etc., for KM activities is a sure way to get these departments to at least consider usage within the department.

Finally, the role of the CEO should not be underestimated. Leaders challenge the process, inspire a shared vision, enable others to act, model the way, and encourage the heart (Lynch, 2003).

Sunassee and Sewry (2003) go a bit further and propose a framework for organizational KM. The proposed framework consists of three main interlinked components: knowledge management of the organization, knowledge management of the people, and knowledge management of the infrastructure and processes. They indicate that the organization needs to achieve a balance between these three subsystems to achieve a successful KM effort.

They stress that it is critical that the KM of the organization be carefully aligned to the overall business strategy of the organization. A close second in importance is an effort to make people feel as if they are part of the change process when implementing KM, including an emphasis on individual learning and innovative thinking.

The model also proposes a set of critical success factors such as the following that will serve to increase the chances of a successful implementation:

1. Align KM strategy with business strategy
2. Receive top management support
3. Create and manage knowledge culture
4. Use of pilot project
5. Create and manage organizational learning
6. Manage people
7. Choose the right technology
8. Include double-loop

6. Where Do We Go Next?

Now that we understand how KM got its start and why it is important, the next chapter delves into KM strategies.

About the Author

From Knowledge Management, Business Intelligence, and Content Management: The IT Practitioner's Guide by Jessica Keyes. Auerbach Publications, 2006.