FORECASTING AND

MANAGEMENT OF
TECHNOLOGY
ALAN L. PORTER
Georgia Institute of Technology, Atlanta, Georgia
A. THOMAS ROPER and THOMAS W. MASON
Rose-Hulman Institute of Technology, Terre Haute, Indiana
FREDERICK A. ROSSINI
George Mason University, Fairfax, Virginia
JERRY BANKS
Georgia Institute of Technology, Atlanta, Georgia with software and documentation by
BRADLEY J. WIEDERHOLT
Institute for Software Innovation, Alpharetta, Georgia

A WILEY-INTERSCIENCE PUBLICATION
JOHN WILEY & SONS, INC.
NEW YORK CHICHESTER BRISBANE TORONTO SINGAPORE
 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

8
MONITORING
OVERVIEW

Monitoring uses diverse sources of information to set the stage for forecasting, to feed information to the
forecasting and planning processes, and to update and improve the projections upon which management
decisions are made. This chapter presents the assumptions underlying monitoring, the various types of
monitoring, and examples of the use of monitoring in industry. Special emphasis is placed on monitoring the
context of technological changes as well as the changes themselves. Techniques to develop, interpret, and
communicate monitoring results are presented, and suggestions are offered on its integration within an
organization.

8.1 INTRODUCTION

Monitoring is to watch, observe, check, and keep up with developments, usually in a well-defined area of
interest for a very specific purpose (Coates et al., 1986. p. 31).

This description captures the essence of this simple technique. Monitoring means scanning the appropriate
environment for pertinent information. That information may pertain to a particular technology
technological monitoring in which case one may want historical information on the technology's
development, current information on the state of the art today, and/or information pointing directly to future
prospects. Alternatively, monitoring may consider the context in hich technology developscontextual
monitoring, or issues management1. This approach derives from the assumption that technological change is
foreshadowed by changes in other technologies and/or in the socioeconomic environment. Thus it should be
possible to monitor signals in these environments, analyze them, and forecast technological development.
For instance, concerns about carbon dioxide emissions promoting the greenhouse effect may presage the
return to favor of nuclear power.
1
issues Management is defined by Coates et al. {1986) as "the organized activity of identifying emerging trends, concerns, or
issues likely to affect an organization in the next few years and developing a wider and more positive range of organizational
responses toward that future" (p. ix). Issues management can help companies devise positive, anticipatory (rather than merely
reactive) responses to new technologies, possible governmental constraints, or potential confrontations. Issues management em-
phasizes contextual monitoring (law-making, social controversies, etc.). but it can include technological monitoring, too.

Forecasting and Management of Technology 114

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

The last part of the monitoring definition emphasizes "specific purpose." This emphasis is vital to avoid
accumulating vast quantities of ill-focused information with no organizing principle. According to Coates et
al. (1986, p. 31), possible objectives in monitoring include:
Detecting scientific/technical or socioeconomic events important to your company
Defining potential threats for the organization implied by those events
Seeking opportunities for the organization implied by changes in the environment
Alerting management to trends that are converging, diverging, speeding up, slowing down, or interacting.

These generic objectiveslooking for events and trends that imply threat or opportunitycan be broken
down further: identifying competing or supporting relationships, flagging critical needs, and ascertaining
potential breakthroughs and their effect on the organization. Whatever the objective, careful definition of
purpose and focus are essential to worthwhile monitoring.

Successful forecasting based on monitoring involves more than merely gathering data. The forecaster must
sift the information for meaningful signals and envision their implications. This is best done through a
systematic monitoring procedure that directs information search and interpretation. Finally, results of
monitoring must be synthesized and communicated effectively to generate appropriate action. Indeed,
substantive work may be dwarfed by the effort to orchestrate, package, and sell the information to achieve
the necessary action (Coates et al., 1986).

Monitoring is one of the most useful techniques in forecasting. Applications depend on what the user needs.
For instance, an Air Force research organization built a technological forecast for artificial intelligence
largely on an expert/literature monitoring effort to systematize research funding plans (Reitman et al..
1985). Search Technology Inc. has used a monitoring system to help design a new technology called the
Designer's Associate (DA)a computer-based system to assist cockpit designerswhich was prepared for
the Air Force. This required an assessment of what technological alternatives might be available to
accomplish given DA functions in 1995. Monitoring contributed in the following ways (Porter, 1988):

It ascertained technological capabilities as of 1988 to help "configure" the DA.

It alerted management to possible new technological capabilities for 1995.
It assessed the risk in designing a system based on those potential capabilities for 1995.
Monitoring also can serve those on the lookout for new products or new processes. Vanston
(1985) has sampled such American corporate applications. For example:
ALCOA watches for new market areas and examines threats to present markets.
Kraft analyzes changes in home eating patterns to determine effects on new food products.
Owens-Corning Fiberglas seeks new products and processes.
Johnson Controls predicts competitors' entrance into new technical areas.

Exhibit 8.1 provides a case illustration of the implications of monitoring.

Exhibit 8.1 Whirlpool Beats the Competition (Based on Davis, 1973).

Whirlpool Corporation tracked developments in the chemical and textile industries (not their own business
domains) to identify permanent press fabrics prior to their commercialization. They acted quickly on this
information to generate the first washer and dryer permanent press cycles, beating their competition to
market by about a year. This resulted in a substantial increase in market share. The major events involving
permanent press products and their impact on Whirlpool are as follows:

Forecasting and Management of Technology 115

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

January 1964 Permanent press cycles on Whirlpool washers and dryers; first
by appliance industry
March 1965 Research project for a "new concept ironer" dropped: monies diverted to other uses
______________________________________________________________________________________

In a quite different context, through its National Center for Science and Technology for Development, the
Chinese State Science and Technology Commission maintains a significant technology forecasting effort.
Monitoring new technologies available from developed nations, such as the United States and Japan, serves
its immediate industrial needs. Monitoring for emerging technologies helps to guide China's own
technological development efforts by avoiding dead ends and seeking realistic niches.

Policymakers also can be served by the monitoring of emerging trends and possible breakthroughs in
technology as well as by contextual monitoring. The Congressional Clearinghouse for the Future provides
this service for the U.S. Congress.

Monitoring also provides a base for technology forecasting using other techniques. It provides data and
background for choosing and using forecasting tools. Section 8.2 considers the principles for sound
monitoring; Section 8.3 offers a systematic monitoring process; Section 8.4 addresses implementation
issues.

8.2 PRINCIPLES

This section describes the basic assumptions that support monitoring and the monitoring methods that are
commonly applied.

8.2.1 Assumptions

Monitoring is an opportunistic technique. If patterned technological development is expected to continue,

monitoring tracks this historical development as a basis for projection. If patterns are not as well-defined,
monitoring relies on current information on the state-of-the-art and expert assessments of future advances.
Such advances may be incremental or breakthroughs monitoring can tackle either. In some cases,
monitoring contributes more to current awareness than to true forecasts, but it is an extremely robust
technique as long as the information gathered is cautiously interpreted.

Both technological and contextual monitoring must be tailored to fit the developmental stage of the
technology, as well as to fit user needs. Table 8.1 presents a version of the "suspect" linear model of change
(recall the innovation process models discussed in Chapter 2)-not to argue for linearity, but to suggest what
information may be most salient to monitoring at various stages of a technology's development.

Some have tried to develop a monitoring framework based on technological characteristics. In particular,
there have been attempts to distinguish functional and structural measures. For instance, Knight (1985)
differentiates main structural units of the computer (memory, computation, control, input-output) and key
functional attributes (computing power, cost, and reliability) He then cumulates qualitative and quantitative
measures on each of these over extended time periods His own assessment is that the approach does a good
job of depicting the evolution of the computing industry, but is not detailed enough to isolate specific
technological changes.

Forecasting and Management of Technology 116

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

This approach is appealing, but there are concerns about its application For instance, over the long term,
structural frameworks may change (for example, with the advent of personal computers or neural
computing) and functional parameters may evolve (for instance, increases in computing power may result in
qualitative changes in criteria or new applications such as artificial intelligence may rely on
nonmathematical processing power) It makes the most sense to consider function and structure, or other
organizing principles, in light of the scope, duration, and purposes of the monitoring effort.

8.2.2 Types of Monitoring

Table 8 2 poses a series of overarching choices that must be made about any monitoring process The basic
focus, whether one is monitoring technology or contextual factors, has been discussed in Section 8 1

The first choice is the time frame. In some cases, a technology or a particular context is being studied in an
exploratory or pre implementation mode (for instance, to help decide if the company wants to pursue that
technology enter a certain market, etc ).

Forecasting and Management of Technology 117

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

In this instance, monitoring is likely to have limited resources and uncertain management commitment, it is
also likely to require strong selling to convey an effective punch line In contrast imminent decision
monitoring suggests an attentive and anxious audience whose specific concerns will be directly addressed in
the monitoring report The postdecion time frame implies evaluative monitoring (that is, determining what
happened by focusing on actual events directly involving the organization) This may be required by outside
authorities (as a way to ensure that negative effects are mitigated) or may be undertaken to help the
organization manage a specific technology In any event, this postdecision monitoring clearly differs from
preimplementation monitoring.

The second choice is whether to undertake a one-time study or to set up an ongoing monitoring system The
former must ensure that it does not narrow the monitoring focus prematurely and miss significant influences
The latter has the advantage of compiling contextual information over a long period that can be used to form
more precise inquiries (see Exhibit 8 2) The monitoring system must have a clear focus and precise
objectives to avoid becoming irrelevant to the organization.

Technological monitoring approaches will differ depending on the level of focus Macro technology
monitoring could encompass a whole spectrum (such as the information technologies as per Exhibit 8 2)
Aternately, micromonitonng might focus on one technology (the computer) or a particular type of that
technology (personal computers) Another choice would be to focus on systems (office automation),
individual devices (personal computers) or components (microprocessor chips) Whatever level is
appropriate for the study, it is incumbent upon the analyst to consider how changes at other levels might
impact the focal technology. For example, the person monitoring office automation cannot ignore
improvements in computer storage that might substitute for networking arrangements.

Forecasting and Management of Technology 118

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

Exhibit 8.2 A "Two-Stomach" Information Technology Monitoring System (Based on Neste, 1988).

Search Technology implemented an ongoing monitoring system to keep abreast of changing information
technologies in early 1989 (Neste, 1988). Essential criteria for this system included:
Minimal effort required for dispersed participants to contribute information to the system
Convenient access to information for users with quite different needs
Low cost
Ongoing system that facilitates generation of up-to-date reports as needed

One technical staff member takes responsibility to maintain the system and evaluate its performance
(observe usage and suggest improvements). One clerical employee helps input information by copying
documents, filing, and performing other related tasks. Six or so technical staff members each take lead
responsibility for several technologies. They process the information as needed and generate suitable
syntheses and forecasts. Another 20 or so technical people contribute to the system as they find materials
pertaining to the targeted 20 (changing over time) technologies. Users include project managers designing a
particular technological system; management personnel proposing to provide monitoring information per se
to clients or drawing on the system for general background information; and technical personnel writing
articles or reports who want information on one or more of these technologies.

Initiation required that management approve the system, that the target technoIogies be indentified, that the
six staff members agree to take responsibility for particular technologies, and that all staff be alerted to the
new system. Staff pre pared pilot forecasts on certain key technologies and generated initial technological
maps to help identify what other technologies should be monitored. Management and project users were
interviewed to determine their needs. Staff were briefed on how to contribute information at a meeting and
through a memo, and a system manual was prepared. A special bulletin board alerted staff to key items.

The physical setup was based on a dual filing systemthe two stomachs. One folder for each technology
contained all raw (undigested) materials, including copied articles, book references (indicating where the
book could be found), clippings, and other such material. A second folder for each technology contained
monitoring reports prepared by Search Technology staff on an as-needed basis (every six months or so) plus
important supporting sources (well digested). When a report was prepared, that input folder was emptied,
and only vital materials were maintained in the report folder. System materials were kept in a designated set
of easily accessible filing cabinets.

Sometimes monitoring serves more than just forecasting purposes; it may also help in choosing a future
course. As mentioned, the Chinese monitor technology developments in other advanced nations. The state
of Japanese or Western technology provides an excellent menu from which to choose their own
development programs. Obviously, monitoring existing technologies greatly reduces the uncertainty, but it
must also consider technology transfer issues, including cultural and political factors.

Invention and innovation are concepts that denote the issues attendant to different stages of development
(see Table 8.1). However, these concepts also represent a choice of monitoring emphasis. For certain uses,
interest may be invention-orientedthat is. focused on how to reduce technological uncertainty (for in-
stance, it may be necessary to determine technological capabilities by a given date). For innovation-oriented
uses, the key is really economicsWill a technology succeed in some market? Invention focuses on critical
technical "milestones," their prerequisites, and the likelihood of success by a given date. Innovation, on the

Forecasting and Management of Technology 119

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
other hand, focuses on whether a technology will meet needs of some users at an attractive price and/or in a
way that is better than competing technologies. It is often easier to predict the emergence of technical capa-
bilities (particularly when improvements come as a stream of incremental gains rather than discrete
breakthroughs) than to predict the date of the actual innovation (that is, success in some market). These two
types of technological monitoring emphasize different information sources (e.g.. engineering versus market
knowledge).

The monitoring of an established technology can draw upon historical pattern information concerning
generally agreed-upon parameters. Martino (1987) offers a process to monitor established technologies that
uses precursors (see Exhibit 8.3). In contrast, the monitoring of an emerging technology must draw on more
diffuse information (also likely to imply more diffuse sources) with a less well-defined target. The emerging
technology is likely to be more sensitive to contextual influences, and the time frame for it is likely to be
longer, with attendant uncertainty in any forecasts.

Motorola devised a systematic monitoring/forecasting approach that used a technology roadmap process to
sharply differentiate between emerging and established technologies (see Chapter 3). A very different
approach is to track emerging technologies in general. For example, the National Institute for Standards and
Technology (NIST. formerly the National Bureau of Standards) identified 12 emerging technologies for the
1990s (Technology Administration, 1990):

Emerging Materials
1. Advanced materials
2. Superconductors

Exhibit 8.3A Precursors of Technological Change

Martino (1987) illustrates this approach by compiling entries from the Engineering Index (and its on-line
Compendex data base) for selected technologies from 1970 to 1987. The resulting qualitative histories can
be used to anticipate coming changes. Martino recommends watching for
Incomplete inventions that need other elements before they can be deployed economically
Development of performance-improving supporting technologies needed by a basic technology
Development of cost-reducing supporting technologies needed by a basic technology
Development of complementary technologies that mesh with the basic technology to make it
useful (such as sensors for automated controllers)
Use of a technological advance in a prestige application before transition to general use (such as
transition of technology from aviation to automotive usesearly uses in racing cars)
An incentive for use. such as reduced cost or elimination of externalities (for instance, fuel
economy or reduction of automotive emissions)
Different origins for leading indicators (for instance, motor car manufacturer advances are nearer-
term indicators of coming automotive change than are advances coming from parts suppliers)

Martino provides the accompanying chronology of plastic auto body shells to illustrate one such precursor
trail. (His report also provides another trail of improvements in steel auto body shells to "counterattack" the
threat posed by plastics.)
Sometimes, a leading indicator relationship can be found. Martino illustrates this relationship using the
development of new aluminum alloys, followed by their first application in an aircraft, as an example. In
that case, a mathematical relationship (by using regression, for example) may be devised to anticipate when
to expect the application of new alloys (average lag is 4.2 years). Martino suggests treating such leading
indicator relationships by devising probability distributions. In the alloy-aircraft case, he uses the method of

Forecasting and Management of Technology 120

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
maximum entropy (Tribus, 1969) with only the mean specified (to yield a geometric distribution) or with
the standard deviation of the lag specified (to yield a Gaussian distribution truncated on the lower end-lag of
zero). In this case, the Gaussian distribution is most informative, predicting virtually no chance of
application of a new alloy in less than three years and almost certain availability within six years. Such
information would be quite helpful to an aircraft manufacturer planning materials for a new plane.

Exhibit 8.3B Events in the History of Plastic Auto Body Shells

1978
Ford built demonstration Graphite Fiber Reinforced Plastic car; test results reported in 1983.
ICI America introduces resins and urethane thickeners for producing of structural grade sheet molding
compound (SMC). Actual field use still needed to demonstrate usefulness for auto applications.
1980
Ford Motor Co. reports design, development, manufacture of lightweight, one-piece SMC hood for
Econoline Van.
Laminates of steel sheet with plastic core have same formability as steel, lower cost, lighter weight.
1981
About 100 kg polymers now used in medium-sized car.
Reports of several experiments with plastic for body panels, interior components.
1982
FIAT builds demonstration car with steel skeleton, plastic panels for body shell. Skeleton has only 800
welds as compared with 3.000 typical for all-steel car. More resistant to corrosion, better formability for low
drag shape, saved 209c weight over all-steel car.
1983
1984 model Fiero bolt-on body allows .005 tolerances in finish and assembly of bodies; precision is equal
to or better than steel.
1984
Production of Pontiac Fiero. steel "space frame" with all-plastic body skin. Flexible plastic "Bexloy" used in
rear. Proves plastic can match surface finish of steel. Special painting arrangements; separate lines for space
frame, for plastic panels, for flexible front and rear fascias. Painting panels before mounting reduces paint
repair work.
Plastics use in autos. triples that of a decade earlier, now 10% of car total weight.
1985
Finite-element methods used to design steel chassis which carries all torsional loads, no stress on plastic
body.
Successful in-plant coloring of ABS body plastic, reducing inventory costs by eliminating need to store
colored plastic.
1986
2.000-ton plastic molding machine produced for auto plants.

Ford Sierra plastic grille panel is body-color painted and in same plane as painted metal panels,
demonstrating that plastic can be used in conjunction with steel.
Source: Martino. 1987. Reprinted by permission of Elsevier Science Publishing Co.. Inc.

Forecasting and Management of Technology 121

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

Emerging Electronics and Information Systems

3. Advanced semiconductor devices
4. Digital imaging technology
5. High-density data storage
6. High-performance computing
7. Optoelectronics

Emerging Manufacturing Systems

8. Artificial intelligence
9. Flexible computer-integrated manufacturing
10. Sensor technology

Emerging Life-Science Applications

11. Biotechnology
12. Medical devices and diagnostics

In contextual monitoring, macro and micro distinctions are also made. At the macro level, various entities
(typically governmental bodies) compile indicators ideally, time series dataon various socioeconomic
factors of general interest (such as educational achievement test scores, unemployment rates). Private-sector
firms may also compile such data, monitor literature, and/or prevail upon experts to identify contextual
trends of note. For instance, the Roper Organization produces "The Public Pulse" a report on American
preferences. One issue (Roper's. 1987) offered 31 major trends shaping the future of American business.
Two examples from that report are:

The Culture of Convenience the fast-rising number of two-income households lies behind the
takeoff of the "convenience industry." projected to continue. . . . this also explains the tremendous
potential of prepared take-out foods and of appliances like microwave ovens. . . .

Permanent Damage to Nuclear Energy Industrygrowing environmental concerns. combined with

the Chernobyl disaster, have retarded further developments in nuclear energy for the foreseeable
future. . . . Public worries and opposition to new nuclear plants are high and rising. . . .The nuclear
power industry probably cannot recover in the short term. . ..

It appears that the first sample projection is holding true in the 1990s, but the second one could be
overturned by other environmental concerns (such as acid rain and the greenhouse effect).
Micro-level contextual monitoring focuses on those elements of the environment perceived to impinge more
directly on an organization's interests (or a technology of interest). For instance, computer firms monitor
announcements, patents, and rumors concerning their competitors' activities.

Contextual monitoring can address a vast array of possible changes. At times it may be useful to check
coverage of all domains likely to influence the organization as regards its monitoring objectives. The
following list will generally provide a comprehensive catalogue of 10 areas for investigation:
Technological (scientific)the 12 NIST technologies to watch
HealthAIDS rates, life span, nutritional levels
Institutional (legal, organizational)governmental regulation and agency power, expansion of
multinational corporations, industrial concentration in a sector, public participation

Forecasting and Management of Technology 122

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
Social (behavioral)advent of the Information Age, changing educational levels, urbanization,
birth rates, expanding elderly population, minority interests
Politicalgovernmental stability, national and state party power, organizational politics
Economicrecession versus prosperity, union strength, job displacements, service sector growth
Cultural (values)changing work ethic, mass media, cultural homogenization, rising aspirations,
consumerism, women's movement, acceptance of fractured families
International (development)world economy, changing patterns of competition, financial
instabilities, OPEC
Ecologicalgreenhouse effect, aquifer pollution, resource shortages, nuclear waste disposal
Security changing patterns of hostility, evolving alliances, military funding. SDI

8.3 MONITORING STEPS

Given the diversity of possible forms of monitoring, any suggested set of steps must be carefully adapted to
specific needs. However, the following steps are generally useful:
1. Determine the monitoring objectives and focus
2. Describe the technology and map the pertinent context
3. Adapt an appropriate monitoring strategy
4. Interpret and communicate results

8.3.1 Monitoring Objectives and Focus

Sections 8.1 and 8.2 have discussed a range of monitoring information types, possible uses, and prospective
users. The first task of the would-be monitor is to determine which elements of these ranges fit the situation.
Specific objectives should be spelled out (monitoring has a tendency to generate overwhelming amounts of
information unless objectives are kept strictly in view), and a focus must be established. To undertake
technological or contextual monitoring, you need to resolve the choices posed in Table 8.2.

Objectives and focus should be discussed by those engaged in the monitoring effort and with potential users.
Agreement should be reached on issues such as scope, temporal extent, personnel and financial resources,
timetables, whether the effort will be ongoing, and interest in trends or breakthroughs.

8.3.2 Technology Description and Contextual Mapping

As discussed in Chapter 4, you must describe a technology before you forecast it. Monitoring is, however,
likely to generate an iterative situation. It begins with a presumptive definition/description of the
technology. Monitoring enriches understanding of the technology, possibly leading to its redefinition in
terms of level (for instance, addressing microprocessors instead of microcomputers), the pertinent
technological system, and the critical parameters or milestones (critical qualitative or quantitative
improvements) to monitor.
Mapping, whether graphic or text, can help identify relationships among technologies (see Figure 8.1 for
selected information technologies perceived pertinent to computer systems). Likewise, it can help identify
vital contextual socioeconomic influences (see the discussion of technology delivery systems in Section
2.3). Without such mapping, critical signals may be missed because monitoring may be wrong or too
narrow.

Technologies and contextual influences should also be linked to organizational interests. Brown (1980)
suggests composing matrices that indicate the likelihood of significant change on one axis and its likely

Forecasting and Management of Technology 123

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
impact on the company on the other (see Figure 8.2). Such a matrix allows quick identification of which
changes would constitute milestones for the organization.

8.3.3 Monitoring Strategy

A number of strategic monitoring issues have been mentioned (such as making monitoring choices, meeting
different objectives, making do with different resource levels, and performing under accelerated or extended
time schedules). In this section, monitoring approaches are differentiated by how familiar you are with the
subject. (This discussion assumes technological monitoring, but the implications generally fit contextual
monitoring, too.) As a guideline, it is useful to contrast three levels of familiarity:
Level 1 Cold
Level 2 Warm
Level 3 Hot

Figure 8.1 Map of technologies relating to computer systems. (Source: Prepared by S. Cunningham.
1990.)

Forecasting and Management of Technology 124

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

Figure 8.2 Impact and probability of occurrence matrix. (Source: Brown. 1980. p. 31.)

At Level 1, you start "cold"that is. you are unfamiliar with the subject. At this level, the immediate
questions include:
What is the technology? How is it defined and described? What is the state-of-the-art?
How do other technologies relate to it? What other contextual factors effect it?
Who are the key players (individuals, organizations, suppliers, regulators, users)?
What are plausible future development pathways?
To accomplish Level 1 monitoring:
Use a "shotgun" approach to gather informationthat is, grab anything that is convenient and may
be pertinent
Emphasize recent literature and review state-of-the-art articles or books
Locate one or two accessible professionals with sufficient expertise in the technology to point out
information sources and to help ensure that the monitoring does not drift wide of the mark

Level 2 suggests that you have either completed a Level 1 introductory effort or that you have some
familiarity with the subject. Objectives become more focused; pertinent questions include:
What are the driving forces propelling this technology?
Can important interdependences with other technologies or with socioeco-nomic factors be mapped?
What are the key uncertainties along the development path?
To accomplish Level 2. the sources of information shift:
Literature searches become more focused: on-line searches are likely to be more fruitful: and
historical searches now may make sense as a way to identify leading indicators of progress and
significant influences.
Locating other forecasts for the focal technology can help answer the pivotal questions and help
ground forecasts.
Networking can be used to identify experts with various perspectives on the technology.

Forecasting and Management of Technology 125

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
It now makes sense to begin to synthesize the information obtained by formulating an image of what is
happening to the technology.

At Level 3, you are very familiar with the subject. Level 3 objectives are even more focused; pertinent
questions include:
Can you specify the key factors to be watched?
What is the most likely development pattern for the immediate future? For the longer term?
Can you offer specific projections?
What recommendations can you make to help your organization manage development?
Actions undertaken to accomplish this level of monitoring include:
Extending the information search to be as comprehensive as feasible with regard to the key factors
Developing a conceptual model of what drives and what impedes development of this technology
Seeking direct confirmation of this model and a review of your projections from experts
Generating a credible forecast by integrating the monitoring results with other forecasting techniques
Possibly establishing a structure for an ongoing monitoring system

These three levels suggest different requirements. Subject expertise is not essential at Level 1. but it is
necessary at Level 3. Comprehensive information access (such as access to on-line systems) is not critical at
Level 1, but it is at Level 3 to ensure thoroughness. Commitments of time and effort also increase at the
higher levels. It is sensible to commit only to a Level 1 effort in investigating a technology to ascertain its
relevance. The results of such an introductory examination may be sufficient to conclude that the subject is
not critical to the organization. On the other hand, if the subject does seem sufficiently important, you may
need to revamp the monitoring program to involve experts (from inside or outside the organization), boost
resource commitments, and set forth a full-blown monitoring system.

8.3.4 Interpretation and Communication

Data do not speak for themselves: therefore, it is very important that key points be flagged in such a way as
to get management's attention. Possible mechanisms to foster communication with potential users include
maps, milestones, and multiple channels.

Graphical presentation of technology maps (imagine Figure 8.1 with all of the linkages shown) quickly
conveys complex relationships, especially to visually oriented individuals (engineers are great candidates).
Technology delivery system sketches fall in this same category, particularly if a few key contextual
influences can be highlighted. Any model of what drives the development of a given technology should be
interpreted to point out the few, critical factors that should be watched.

Monitoring may generate extensive historical profiles of events and/or trends. These should be offered with
further interpretation. The notion of milestones for which to watch can be quite effective. Table 8.3
illustrates innovation milestones in the history of cement manufacturing in the United States. Milestones
may reflect events or trends advancing dramatically so as to improve performance significantly.

Multiple communication channels improve the chances of successfully transmitting information. Both
parallel channels and serial repetition (same or similar messages repeated at different times) may make
sense. For example, the monitoring system discussed in Exhibit 8.2 has multiple channels. Some channels
work well for some types of information, but not for others. Paper folders fulfill the configuring objectives
quite well, but they do not accomplish the alerting or forecasting functions. Wall maps of relationships (see

Forecasting and Management of Technology 126

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

Figure 8.1) seem excellent for configuring and helpful for alerting. A bulletin board provides a nice alerting
mechanism, but little more. Reports appear best for presenting forecasts, but they are too infrequent to alert
organizational staff in a timely manner. Finally, meetings can help round out rough spots by providing rich,
two-way communication opportunities, but they are costly in personnel time.

In addition to the inherent strengths and weaknesses of the communication channels, consider the individual
quirks of the providers and users. Some people favor graphical presentations; some want it written; others
thrive on verbal exchange. When feasible, it may be best to use multimedia schemes to reach as many as
possible.

8.4 IMPLEMENTATION

Implementation involves three issues: Who should perform the monitoring? How can they put together a
monitoring program? How can they integrate monitoring results with other forecasting techniques?

8.4.1 Who Should Conduct Monitoring?

Informal monitoring can be done by anyone who alertly keeps track of information obtained from routine
reading of journals, casual browsing, and other such activities. Some sources provide a sort of ready-made
monitoring. For instance. IEEE Spectrum publishes a "newslog" each month of items pertaining to
electronics developments. Predicasts Forecasts compiles forecasts from a wide range of government reports
and open literature: these are compiled quarterly (and maintained on-line as well) using Standard Industrial
Classification (SIC) codes. An example issue contains some 95 items pertaining to industrial robots,
reporting projections for robot sales in units or dollars, related employment, prices, or numbers in use
(specifying years that the projections address and providing source information on the projections). Any
person with a professional level of expertise related to the topic should be able to select pertinent
information from these sources.

Forecasting and Management of Technology 127

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.

Formal monitoring varies so greatly in scale that formulating rules would be foolhardy. Instead, consider
the range of possible performers. Outside information services may be able to provide the information
desired. Options include targeted newsletters, clipping services, trade associations, or contracted monitoring
studies done by research firms. Possible disadvantages include less knowledge of the organization's needs,
weaker communication bonds to organizational management, and less control over the accessibility of
information to others. Advantages include possible strong subject expertise and established access channels.

Within the organization, a single staff person could be assigned to monitor a subject; alternately, that person
might be supported by some outside information services. Although the single-person approach is efficient,
it may be less effective than a more broadly based approach.

A study by Allen (1977) on where and how scientists and engineers obtain information has pointed to the
importance of "gatekeepers." These professionals (perhaps, but not necessarily, managers) serve as critical
bridges to the outside world. Most of the flow of ideas in technical development takes place within project
groups. According to Allen, relatively few individuals in R&D read much technical literature or. especially
in the case of engineers, actively participate in technical meetings. The gatekeeper funnels in a wide sweep
of information for use by the project team in their relatively narrow application focus. Gatekeepers tend to
emerge, rather than to be created by their managers? They should be well networked within their
organizational unit (whether this is a single project team or a large company) and be avid information
hounds. Gatekeepers make ideal monitors.

The next logical step is to assign an ad hoc team, from some central staff unit, to conduct monitoring.
Alternatively, a team might be set up on a permanent basis to run a monitoring system (see Exhibit 8.2), or

Forecasting and Management of Technology 128

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
monitoring could involve more widely dispersed personnel (see Exhibit 8.4). The dispersed effort can
directly involve those with subject expertise and decision-making responsibilities, but it requires incentives
to participate and checkpoints to ensure performance and communicate results. Coates et al. (1986)
elaborate on possible arrangements.

Exhibit 8.4 Dispersed Monitoring in a Food Processing Company

Technology forecasting is a new venture for this company. In the past, innovation has not been a high
priority for this marketing-oriented firm. Now. however, management is making new technology a major
corporate objective to save on costs through process improvement and to expand markets through new
products and packaging.

This company is pursuing a diffused monitoring strategy. The director of en gineering has initiated a
comprehensive search strategy that categorizes possible application areas for new technology. Within each
category, the director has specified technologies to monitor. For instance:

Pumps, agitators, and mixers

Ultrafiltration
Microprocessors
Wastewater treatment
Wire-guided vehicles

One engineer is assigned responsibility for each area. That person is charged with monitoring the
technology, identifying potential innovations for the company to pursue, and developing specific proposals
for action. Such proposals can come forth at any time. (In addition, regular semiannual progress reports are
due.) Superior proposals become development projects with specific objectives and budgets.
The great advantage of this approach is that the engineers responsible for development and implementation
do the forecasting. No separate "sales pitch" is needed to convince engineers to act on the forecast.

8.4.2 How Should Monitoring Be Conducted?

Traditionally, monitoring relied heavily on printed materials. In decades past, technology monitoring relied
on composing journals with entries interpreted with respect to prescribed parameters and concerns (see
Bright. 1972). That appears too restrictive for ongoing systems [hat try to track rapidly advancing families
of technologies.

Advances in electronic information accessibility represented by the some 5,000 on-line databases (as of
1990). the gateway services (such as BRS and DIALOG), and the wide availability of networked personal
computers offer attractive options. Instead of maintaining a monitoring file with journal reports keyed on
certain identified parameters, on-line searches on old or new parameters can be generated. As of early 1990,
however, searches are still not satisfactory for many purposes. Hard copy systems supplemented by on-line
searches to capture additional new items on specific topics appear to be most cost-effective (see Exhibit
8.2). However, as data bases become increasingly "full text" and include more graphic information, elec-
tronically based monitoring should emerge as the method of choice. Watch for intelligent information
retrieval mechanisms that go beyond Boolean and probabilistic searches (using systems such as SIRE, for
instance) to help the user select valuable information (Wiederholt et al.. 1989). Indeed, artificial

Forecasting and Management of Technology 129

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
intelligence-supported electronic monitoring could, one day. help the user pose the monitoring questions
and come back with suitable answers, not just unprocessed sources.

Today it may be worthwhile to develop monitoring files on the computer for ease of cross-referencing
and/or browsing (that is, using hypermedia systems). Input and maintenance of such a system is more time-
consuming than use of hard copy files, but the payoff can be worthwhile gains in effectiveness in some large
monitoring efforts with many people involved and many cross-linked interests (such as those suggested by
Figure 8.1).

8.4.3 Integrating Monitoring Results with Other Forecasting Techniques

Monitoring provides the basis for most technology or socioeconomic forecasting. Sometimes it provides all
that is needed or affordable. Most of the time, however, monitoring provides one component of a forecast. If
monitoring is restricted to mean gathering of information from published sources, then it must usually be
enriched by expert opinion. For instance, one forecast of intelligent information retrieval (Wiederholt et al.,
1989) began with a casual discussion with in-house experts, continued with consolidation of published
literature, and then formally requested review and elaboration by outside experts. The process continued
with revision and further monitoring over several months before the final report. In such a process,
monitoring is tightly integrated with expert opinion methods. Monitoring may initiate forecasting studies
that utilize other forecasting techniques as well. For instance, monitoring of developments in personal
computers has uncovered salient time series data, suggesting trend extrapolation.

Monitoring also provides a vehicle to keep forecasts alive. Rather than allow a study to be published and
become out of date, assign someone to track developments and update the study periodically (this can be
done at relatively low cost). Should such monitoring uncover milestone changes, then reexamine the
forecast. In the previous paragraph, monitoring was cast as the precursor to a forecast; here it is the follow-
on activity.

Because changes in different technologies are interdependent and technological change depends on
contextual social influences, it is attractive to consider forecasting such sociotechnical systems rather than
focusing on a single technology. Again, monitoring constitutes the main activity of such forecasting efforts.
If monitoring turns up a critical change then initiate a more focused study. Such integrated, monitoring-
based forecasting systems serve the organization best.

8.5 RECOMMENDED SOURCES

Here are the general categories of published information sources with examples of each:
Journals: The Futurist. Technological Forecasting and Social Change, Futures, Science
Trade journals: PC Week, Iron Age
Newsmagazines: Time, Business Week
Newspapers: New York Times, Washington Post, The Wall Street Journal (es pecially indexes of
these, on-line or hard copy)
Government activity reports:

Legislative reportsCongressional Research Service studies. Office of Technology Assessment studies.

CIS Index (Congressional Committee hearings and reports), Congressional Record, Congressional Quarterly
Executive reportsweekly compilation of presidential documents, monthly catalog of government reports,
agency indexes (such as NASA, DoE)

Forecasting and Management of Technology 130

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
Indices: Engineering Index (known as Compendex on-line), Science Citation Index (also Social
Science Citation Index), Trade and Industry Index, Computer Index (all these are available on-line).
Library of Congress Catalog of Books by Subject
On-Line Database Gateways: services such as BRS and DIALOG provide access to multiple
electronic data bases; college library collections increasingly are cataloged on-line and may provide
reduced price (or free) access to outside data bases
Information Services: International Data Corporation (Framingham, Massachusetts). Bacon's
Clipping Bureau (Chicago)

EXERCISES
8.1 Exhibit 8.1 chronicles Whirlpool's successful monitoring of advances in chemicals and textiles to
affect home appliances. Consider the following industries:
(A) A major American steel company
(B) A Japanese home entertainment company
(C) A multinational oil company subsidiary in the pesticide business What domains would you monitor
for these industries today and why?
8.2 Monitoring can be configured very differently as a function of its purposes and the situation in which
it is occurring. Imagine yourself formulating a monitoring effort in three of the following situations:
Information technology developmentsfor the government of a small developing country in Latin
America
Information technology developments for a major American bank
New materials developmentsfor a "Big Three" American automaker
Genetic engineering developments for a Congressional Committee
New drug possibilities for a multinational pharmaceutical company
Socioeconomic contextual changes for a multinational pharmaceutical company
(A) Identify the key type of uncertainty for each of these three and your major information acquisition
concerns (refer to Table 8.1).
(B) Following Table 8.2, make each of the pertinent choices and briefly explain your reasoning for these
three situations.
8.3 For one of the situations in Exercise 8.2 (or another of your choosing), lay out the "big picture" aspects
of a monitoring program in the form of a proposal to top management:
(A) Set out the objectives:
Identify the key user(s) and the prime use(s)
Pose the key questions to be answered by the monitoring program
(B) Will this be a single, ad hoc study or a monitoring system? Why?
(C) Describe how the monitoring program will be managed:
Who will do the monitoring?
What incentives will you use?
What resistances do you anticipate?
How will you evaluate performance?
(D) Propose a schedule for the monitoring effort.
(E) Designate the resources required and justify them.
8.4 (Project-suitable) Select a topic of interest (a technology or the context for one company's business
interests) and conduct Level 1 (or cold start) monitoring. Follow Section 8.3.3. using suggested sources of
information to fulfill the four objectives. Your brief report should include a technology description,
identification of related technologies (sketch relationships as done in Figure 8.1). identification of key
players, and setting out possible development pathways. In addition, note what you found to be promising
hard copy sources (such as key serials, books) and on-line data bases.

Forecasting and Management of Technology 131

 Porter, A. Roper, T. Mason, T. Rossini, F. Banks, J. Wiedweholt.
8.5 (Project-suitable) Carry your project forward, conducting a Level 2 (or warm) monitoring program on
the topic developed in Exercise 8.4 or on a topic with which you are already familiar. Choose among the
following activities:
(A) Conduct a focused on-line search to round out what you know about a particular technology. Classify
how pertinent various sources are.
(B) Interview two experts to verify the accuracy of your information. Have 5 to 10 questions ready to help
you find out about unpublished sources, forecasts, other experts, who is promoting the technology, and so
forth. Summarize what you learn from them.
(C) Create a historical chronology of the development of a particular technology or product line. Use this
to suggest driving forces and key uncertainties. Consider a precursor analysis, qualitative or probabilistic
(see Exhibit 8.3). if your data warrant. Alternatively, consider preparing a product roadmap. as discussed in
Chapter 3. If neither approach appears useful in this case, explain why and offer your own analysis of the
chronology.
(D) Map the competing and supporting technologies that affect this technology or product line. Use an
impact matrix (Figure 8.2) to identify those in which developments could constitute milestones for your
technology or product line.
(E) Map the pertinent contextual factors that affect, or are affected by, your technology or product line. Use
an impact matrix to identify potential milestones.
8.6 (Project-suitable) Set up a "two-stomach system" for your technology (see Exhibit 8.2). You will need
manila folders and an accessible storage site. If you have a project team, write your understanding of the
topics to be covered, a deadline for information gathering, and who will be responsible for analyzing each
of the topics. Agree on what those analyses will contain, their format, and their initial due date. If sensible,
arrange to continue monitoring and have later topic updates scheduled.

Forecasting and Management of Technology 132