Chapter 6 / Project Activity Planning: Acknowledgments

292 CHAPTER 6 / PROJECT ACTIVITY PLANNING
problems before they become big ones. Become Gershanov, K. M. (1995). Emergency preparedness in five
sensitive to indicators of impending project fail- easy steps. Occupational Health and Safety, 64 (3), 51–53.
ure. Pay special attention to untracked indicators, Kruse, C. (1993, June). Disaster plan stands test of hur-
because these are the most likely to cause trouble. ricane. Personnel Journal, 36–43.
Develop antennae and know when the project is
Kurstedt, H. A., Jr., Mallak, L. A., and Pacifici, L. C.
going wrong.
(1992). Expand quality management into the customer’s
• Choose a project manager indigenous to the coun- environment to establish effective measures and standards.
try where the project is being conducted. An indig- Proceedings of the 1st International Symposium on Produc-
enous project manager will be sensitive to the social tivity and Quality Improvement, February 1992, 478–485.
and political aspects of the project and its peripheral
Lagadec, P. (1993). Preventing chaos in a crisis. London:
issues and will catch more problems while they’re
McGraw-Hill.
small or otherwise undetectable to the outsider.
Leonard, J. B. (1995). Assessing risk systematically: Illi-
• Be mindful of the social and political consequences of
nois Power’s risk assessment system. Risk Management,
crises or events. Critics, or stakeholders, bear signifi-
42:1, p. 12.
cant influence on project success regardless of what
the indicators of cost, schedule, and quality show. Lloyd, J. (1996, July 28). U.S. official says athletes were
Learn how to satisfy stakeholders (Mallak, Patzak & safe—and feel safe. USA Today, p. 3C.
Kurstedt,1991). Identify one spokesperson as a liai- Mallak, L. A., Patzak, G. R., and Kurstedt, Jr.,
son with the public and prepare a procedure for quick H. A. (1991). Satisfying stakeholders for successful proj-
dissemination of information to all affected parties. ect management. Computers and Industrial Engineering,
21, 429–433.
• Adopt a systems view and separate the crisis from
the origin of the crisis. Consider the basic perfor- O’Connor, P. J. (1996, May 23). Security practice for con-
mance principles and problem analysis techniques vention called a success. Chicago Sun-Times, p. 23.
popularized in total quality management programs. Sawle, W. S. (1991). Concerns of project managers: Crisis
Look forward and backward to access the potential project management. PM Network, 5(1), 25–29.
overall effects of the crisis.
Slack, C. (1996, July 15). Bertha gives VDOT center real-
These tools, recommendations, and strategies should life situation to test computer system. Richmond Times Dis-
help project managers to manage their crises better and patch, p. D–13.
perhaps to avoid some crises altogether. Making time and Walker, J. A., and Middleman, L. I. (1988). Tabletop
resources available to those in charge of crisis planning is exercise programs complement any emergency manage-
essential; otherwise these critical tasks will be subordinated ment system. Proceedings of the ANS Topical Meeting on
to the day-to-day activities, a vicious circle that can increase Emergency Response—Planning, Technologies, and Imple-
the likelihood for a larger crisis going undetected until it’s mentation. Charleston, SC.
too late. The regular and proper use and testing of risk anal-
yses, contingency plans, logic charts, and tabletop exercises Questions
should surface the information, discussion of decisions
and actions, and mitigation techniques that may reduce the 1. Planning for inevitable crises seems to be quite logical,
occurrence and impact of crises in projects. yet is rarely done in projects. Why?
2. Would some of these tools have been of value to Iceland
Acknowledgments in the Project Management in Practice example?
The preparation of this paper was partially funded by 3. Scenario analysis—the brainstorming of possible crises
U.S. Department of Energy (DOE) Grant No. DE-FG05- and anticipation of their outcomes—seems like another
86DP70033. useful tool here. How does this approach compare to the
tools described?
References 4. Which of the four tools would have the most value?
Which would be easiest to implement?
Bradford, M. (1996, February 2). Firms may be caught 5. In their recommendations to project managers regarding
in a five-ring circus: With games on, planning will pay off. implementing these tools, which recommendations are
Business Insurance, p. 3. most important?
Ceniceros, R. (1995, October 23). Nestle resorts to crisis.
Business Insurance, p. 83.
81721_Ch06.indd 292 9/16/08 1:09:47 AM

C H A P T E R
7
Budgeting and
Cost Estimation
In Chapter 6, we reviewed the planning process and gave some guidelines for designing the
project plan. We now begin our discussion of PMBOK knowledge area 4: Project Cost Man-
PMBOK Guide
agement. We treat the subject here in terms of planning (or budgeting) for the costs of project
resources but we will reconsider the issue in Chapter 9 when we discuss the allocation of
resources to project tasks.
First priority is, of course, obtaining resources with which to do the work. Senior man-
agement approval of the project budget does exactly that. A budget is a plan for allocating
resources. Thus, the act of budgeting is the allocation of scarce resources to the various endeav-
ors of an organization. The outcomes of the allocation process often do not satisfy managers of
the organization who must live and work under budget constraints. It is, however, precisely the
pattern of constraints in a budget that embodies organizational policy. The degree to which the
different activities of an organization are fully supported by an allocation of resources is one
measure of the importance placed on the outcome of the activity. Most of the senior managers
we know try hard to be evenhanded in the budgetary process, funding each planned activity at
the “right” level—neither overfunding, which produces waste and encourages slack manage-
ment, nor underfunding, which inhibits accomplishment and frustrates the committed. (This is
not to suggest that subordinate managers necessarily agree with our assessment.)
The budget is not simply one facet of a plan, nor is it merely an expression of organizational
policy; it is also a control mechanism. The budget serves as a standard for comparison, a baseline
from which to measure the difference between the actual and planned uses of resources. As the
manager directs the deployment of resources to accomplish some desired objective, resource
usage should be monitored carefully. This allows deviations from planned usage to be checked
against the progress of the project, and exception reports can be generated if resource expendi-
tures are not consistent with accomplishments. Indeed, the pattern of deviations (variances) can
be examined to see if it is possible, or reasonable, to forecast significant departures from budget.
With sufficient warning, it is sometimes possible to implement corrective actions. In any event,
such forecasting helps to decrease the number of undesirable surprises for senior management.
Budgets play an important role in the entire process of management. It is clear that budget-
ing procedures must associate resource use with the achievement of organizational goals or the
planning/control process becomes useless. If budgets are not tied to achievement, management
may ignore situations where funds are being spent far in advance of accomplishment but are
293
81721_Ch07.indd 293 9/16/08 1:35:10 AM

294 CHAPTER 7 / BUDGETING AND COST ESTIMATION
within budget when viewed by time period. Similarly, management may misinterpret the true
state of affairs when the budget is overspent for a given time period but outlays are appropri-
ate for the level of task completion. Data must be collected and reported in a timely manner, or
the value of the budget in identifying and reporting current problems or anticipating upcoming
problems will be lost. The reporting process must be carefully designed and controlled. It is of
no value if the data are sent to the wrong person or the reports take an inordinately long time to
be processed through the system. For example, one manager of a now defunct, large, computer
company complained that, based on third-quarter reports, he was instructed to act so as to alter
the fourth-quarter results. However, he did not receive the instructions until the first quarter of the
following year.
In Chapter 6, we described a planning process that integrated the planning done at dif-
ferent levels of the project. At the top level is the overall project plan, which is then divided
and divided again and, perhaps, still again into a “nest” of plans. Project plans were shown to
be the verbal equivalents of the WBS. If we cost the WBS, step by step, we develop a project
budget. If we cost project plans, we achieve exactly the same end. Viewed in this way, the
budget is simply the project plan in another form.
Let us now consider some of the various budgeting methods used in organizations. These
are described in general first, then with respect to projects. We also address some problems
of cost estimation, with attention to the details and pitfalls. We consider some of the special
demands and concerns with budgeting for projects. Throughout the chapter attention is paid
to dealing with budgetary risk, although the methods of handling risk will be covered in
greater detail in Chapter 8. Finally, we present a method for reducing the risk in making
estimations, and improving one’s skills at budget estimation, or estimation and forecasting of
any kind. Printouts of project budgets from PM software packages will be shown in Chapter
10 where we cover project management information systems.
7.1 ESTIMATING PROJECT BUDGETS

In order to develop a budget, we must forecast what resources the project will require, the
required quantity of each, when they will be needed, and how much they will cost, including
the effects of potential price inflation. Uncertainty is involved in any forecast, though some
forecasts have less uncertainty than others. An experienced cost estimator can forecast the
number of bricks that will be used to construct a brick wall of known dimensions within 1 to
2 percent. The errors, however, are apt to be much larger for an estimate of the number of pro-
grammer hours or lines of code that will be required to produce a specific piece of software.
Project Management in Practice

Pathfinder Mission to Mars—on a Shoestring
In 1976, NASA’s two Viking Mars-lander missions took mission. This amazing cost reduction was achieved
six years and $3 billion (in 1992 dollars) to develop. through a variety of means but the most important was
Twenty-one years later, on July 4, 1997, Mars Pathfinder perhaps the philosophical one that this was a design-to-
and Sojourner Rover landed on Mars once again, but at cost project rather than a design-to-performance project.
a development cost of only $175 million, representing Given this philosophy, the scope of the mission was inten-
a whopping 94 percent cost reduction over the earlier tionally limited and “scope-creep” was never an issue:
81721_Ch07.indd 294 9/16/08 1:35:10 AM

7.1 ESTIMATING PROJECT BUDGETS 295
The Pathfinder Rover explores Martian terrain.
• to achieve a successful landing • mission designers/builders transitioned into the

testers/operators to save documentation, time,
• return of engineering telemetry
labor cost, and chance of error
• acquisition and transmission of a single, partial
• existing NASA mission infrastructure was used
panoramic image
rather than designing new systems
• successful rover deployment and 7 sol (Martian
• instituting time-phased “what if” and lien lists
day) operation on the surface
for real or potential current and anticipated items
• completion of a 30 sol lander mission meeting of cost growth during the project
all engineering, science, and technology objec-
tives • choosing to use a “single-string” but higher risk
design and offsetting the risk by using more
• one successful alpha proton X-ray spectrom- reliable parts
eter measurement of a Martian rock and soil
sample. • 70 percent of major procurements contracts
were fixed-price rather than cost-plus
The means of limiting the cost of the mission were • creative procurement, such as existing equipment
multiple and creative: spares, and accounting, such as lower burden
• development was cost-capped, with no opportu- rate personnel
nity for more funds On July 5, the Mars Sojourner Rover rolled down its
• identifying a set of “de-scope” options which deployment ramp and the resulting pictures made the
could be implemented in case the cost grew headlines on newspapers around the world. The mis-
beyond the fixed budget sion continued for almost three months and returned
• mission, flight, and ground systems designs 2.6 gigabites of scientific and engineering data, 16,000
were driven by existing hardware and system lander camera images, 550 rover camera images, 8.5
capability million environmental measurements, and the results
• a project cash reserve of 27 percent of the total of 16 chemical rock/soil experiments and 10 technol-
budget was held back and carefully planned for ogy rover experiments.
time-phased release throughout the duration of Source: C. Sholes and N. Chalfin, “Mars Pathfinder Mission,” PM
the project Network, January 1999, pp. 30–35.
81721_Ch07.indd 295 9/16/08 1:35:10 AM

While the field of software science makes such estimates possible, the level of uncertainty is
considerably higher and the typical error size is much larger.
In many fields, cost-estimating methods are well codified. For example, in fields such as
construction, costs can often by estimated by scaling the various cost elements appropriately.
For example, building one mile of a four-lane road can be estimated from the individual cost ele-
ments of previously constructed two-lane roads—e.g., the asphalt cost may be double while the
cost of the road’s shoulders may be the same. Similarly, parametric estimating relies on well-
known statistical correlations between various factors such as the total cost of a house relative to
the square feet of living area. The databases of purchasing departments include multitudes of in-
formation devoted to the techniques of estimating the quantities of materials and labor required
to accomplish specific jobs. Also on the Internet are links detailing what materials, services,
and machines are available, and from whom. Every business has its own rules of thumb for cost
estimating. These usually distill the collective experience gained by many estimators over many
years. An experienced producer of books, for example, can leaf through a manuscript and, after
asking a few questions about the number and type of illustrations and the quality of paper to be
used, can make a fairly accurate estimate of what it will cost to produce a book.
We will have more to say about gathering budget data shortly. Before doing so, however,
it is helpful to understand that developing project budgets is much more difficult than develop-
ing budgets for more permanent organizational activities. The influence of history is strong in
the budget of an ongoing activity. Many entries are simply “last year’s figure plus X percent,”
where X is any number the budgeter feels “can be lived with” and is probably acceptable to the
person or group who approves the budgets. While the project budgeter cannot always depend
on tradition as a basis for estimating the current project budget, it is not uncommon for the bud-
geter to have budgets and audit reports for similar past projects to serve as guides. Although we
maintain that all projects are unique, many are not very different from their predecessors and
can serve as reasonable guides when forecasting current project budgets. Tradition also aids
the estimation process in another way. In the special case of R & D projects, it has been found
(Dean et al., 1969) that project budgets are stable over time when measured as a percent of the
total allocation to R & D from the parent firm, though within the project the budget may be
reallocated among activities. There is no reason to believe that the situation is different for other
kinds of projects, and we have some evidence that shows stability similar to R & D projects.
This notion has been formalized in the practice of “life cycle costing.” The life cycles
of past projects are studied as models for the way costs accrue over the life cycles of similar
projects. Given information about costs during the early life of a project, the model can be
used to forecast the total cost over the project’s life cycle.*
A more interesting estimation technique that also depends on actual costs early in the
life of a project is based on earned value analysis (Zwikael et al., 2000). (For a description of
earned value analysis, see Chapter 10.) Early actual costs on a project are compared to their
estimates, and the remaining costs are adjusted by assuming a constant actual-to-estimate cost
ratio. The assumption of a constant ratio gives the lowest average estimation error (11 percent)
of the five different predictors tested.
For multiyear projects, another problem is raised. The plans and schedules for such projects
are set at the beginning of project life, but over the years, the forecast resource usage may be altered
by the availability of alternate or new materials, machinery, or personnel—available at different
costs than were estimated, giving rise to both the risk of inflation and technological risk. The longer
the project life, the less the PM can trust that traditional methods and costs will be relevant.
*We do not demonstrate it here, but Crystal Ball® can fit distributions to historical data. This is done by selecting
the Fit button in CB’s Distribution Gallery window. Then specify the location of the data. CB considers a wide variety of
probability distributions and offers the user optional goodness-of-fit tests—see the Crystal Ball® User Manual.
81721_Ch07.indd 296 9/16/08 1:35:11 AM

Tradition has still another impact on project budgeting. Every organization has its
idiosyncrasies. One firm charges the project’s R & D budget with the cost of training sales
representatives on the technical aspects of a new product. Another adopts special property
accounting practices for contracts with the government. Unless the PM understands the organi-
zational accounting system, there is no way to exercise budgetary control over the project. The
methods for project budgeting described below are intended to avoid these problems as much
as possible, but complete avoidance is out of the question. The PM simply must be familiar
with the organization’s accounting system!
One aspect of cost estimation and budgeting that is not often discussed has to do with
the actual use of resources as opposed to the accounting department’s assumptions about
how and when the resources will be used. For instance, presume that you have estimated that
$5,000 of a given resource will be used in accomplishing a task that is estimated to require
five weeks. The actual use of the resource may be none in the first week, $3,000 worth in
the second week, none in the third week, $1,500 in the fourth week, and the remaining $500
in the last week. Unless this pattern of expenditure is detailed in the plan, the accounting
department, which takes a linear view of the world, will spread the expenditure equally
over the five-week period. This may not affect the project’s budget, but it most certainly
affects the project’s cash flow. The PM must be aware of both the resource requirements
and the specific time pattern of resource usage. This subject will be mentioned again in
Chapter 9.
Another aspect of preparing budgets is especially important for project budgeting. Every
expenditure (or receipt) must be identified with a specific project task (and with its associated
milestone, as we will see in the next chapter). Referring back to Figure 6-6, we see that each
element in the WBS has a unique account number to which charges are accrued as work is
done. These identifiers are needed for the PM to exercise budgetary control.
With these things in mind, the issue of how to gather input data for the budget becomes a
matter of some concern. There are two fundamentally different strategies for data gathering,
top-down and bottom-up.
Top-Down Budgeting
This strategy is based on collecting the judgments and experiences of top and middle managers,
and available past data concerning similar activities. These managers estimate overall project
cost as well as the costs of the major subprojects that comprise it. These cost estimates are
then given to lower-level managers, who are expected to continue the breakdown into budget
estimates for the specific tasks and work packages that comprise the subprojects. This process
continues to the lowest level.
The process parallels the hierarchical planning process described in the last chapter. The
budget, like the project, is broken down into successively finer detail, starting from the top, or
most aggregated level following the WBS. It is presumed that lower-level managers will argue
for more funds if the budget allocation they have been granted is, in their judgment, insuffi-
cient for the tasks assigned. This presumption is, however, often incorrect. Instead of reasoned
debate, argument sometimes ensues, or simply sullen silence. When senior managers insist
on maintaining their budgetary positions—based on “considerable past experience”—junior
managers feel forced to accept what they perceive to be insufficient allocations to achieve the
objectives to which they must commit.
Discussions between the authors and a large number of managers support the contention
that lower-level managers often treat the entire budgeting process as if it were a zero-sum
game, a game in which any individual’s gain is another individual’s loss. Competition among
junior managers is often quite intense.
81721_Ch07.indd 297 9/16/08 1:35:11 AM

The advantage of this top-down process is that aggregate budgets can often be developed
quite accurately, though a few individual elements may be significantly in error. Not only are
budget categories stable as a percent of the total budget, the statistical distribution of each
category (e.g., 5 percent for R & D) is also stable, making for high predictability (Dean et al.,
1969). Another advantage of the top-down process is that small yet costly tasks need not be
individually identified, nor need it be feared that some small but important aspect has been
overlooked. The experience and judgment of the executive is presumed automatically to factor
all such elements into the overall estimate. Questions put to subordinates, however, indicate
that senior management has a strong bias toward underestimating costs.
Bottom-Up Budgeting
In this method, elemental tasks, their schedules, and their individual budgets are constructed,
again following the WBS. The people doing the work are consulted regarding times and
budgets for the tasks to ensure the best level of accuracy. Initially, estimates are made in terms
of resources, such as labor hours and materials. These are later converted to dollar equivalents.
Standard analytic tools such as learning curve analysis (discussed in the next section) and work
sampling are employed where appropriate to improve the estimates. Differences of opinion are
resolved by the usual discussions between senior and junior managers. If necessary, the project
manager and the functional manager(s) may enter the discussion in order to ensure the accu-
racy of the estimates. The resulting task budgets are aggregated to give the total direct costs of
the project. The PM adds such indirect costs as general and administrative (G&A), possibly a
project reserve for contingencies, and then a profit figure to arrive at the final project budget.
Bottom-up budgets should be, and usually are, more accurate in the detailed tasks, but it is
critical that all elements be included. It is far more difficult to develop a complete list of tasks
when constructing that list from the bottom up than from the top down. Just as the top-down
method may lead to budgetary game playing, the bottom-up process has its unique managerial
budget games. For example, individuals overstate their resource needs because they suspect
that higher management will probably cut all budgets. Their suspicion is, of course, quite
justified, as Gagnon (1982, 1987) and others have shown. Managers who are particularly
persuasive sometimes win, but those who are consistently honest and have high credibility
win more often.
The advantages of the bottom-up process are those generally associated with participative
management. Individuals closer to the work are apt to have a more accurate idea of resource
requirements than their superiors or others not personally involved. In addition, the direct
involvement of low-level managers in budget preparation increases the likelihood that they
will accept the result with a minimum of grumbling. Involvement also is a good managerial
training technique, giving junior managers valuable experience in budget preparation as well
as the knowledge of the operations required to generate a budget.
While top-down budgeting is common, true bottom-up budgets are rare. Senior manag-
ers see the bottom-up process as risky. They tend not to be particularly trusting of ambitious
subordinates who may overstate resource requirements in an attempt to ensure success and
build empires. Besides, as senior managers note with some justification, the budget is the most
important tool for control of the organization. They are understandably reluctant to hand over
that control to subordinates whose experience and motives are questionable. This attitude is
carried to an extreme in one large corporation that conducts several dozen projects simultane-
ously, each of which may last five to eight years and cost millions of dollars. Project managers
do not participate in the budgeting process in this company, nor did they, until recently, have
access to project budgets during their tenure as PMs. Reconciling top-down with bottom-up
81721_Ch07.indd 298 9/16/08 1:35:11 AM

budgets is obviously an area where the earlier principles of negotiation and conflict manage-
ment, as described in Chapter 4, would be useful.
Work Element Costing

The actual process of building a project budget—either top-down or bottom-up or, as we will
suggest, a combination of both—tends to be a straightforward but tedious process. Each work
element in the action plan or WBS is evaluated for its resource requirements, and the cost of
each resource is estimated.
Suppose a work element is estimated to require 25 hours of labor by a technician. The
specific technician assigned to this job is paid $17.50/hr. Overhead charges to the project are
84 percent of direct labor charges. The appropriate cost appears to be
25 hr $17.50 1.84 $805.00
but the accuracy of this calculation depends on the precise assumptions behind the 25-hr
estimate. Industrial engineers have noted that during a normal eight hour day, no one actually
works for all eight hours. Even on an assembly line, workers need breaks called “personal
time.” This covers such activities as visiting the water cooler, the toilet, having a cigarette,
blowing one’s nose, and all the other time consuming activities engaged in by normal people
in a normal workplace. A typical allowance for personal time is 12 percent of total work time.
If personal time was not included in the 25-hr estimate made above, then the cost calculation
becomes
1.12 25 hr $17.50 1.84 $901.60*
The uncertainty in labor cost estimating lies in the estimate of hours to be expended. Not inclu-
ding personal time ensures an underestimate.
Direct costs for resources and machinery are charged directly to the project, and are not
usually subject to overhead charges. If a specific machine is needed by the project and is the
property of a functional department, the project may “pay” for it by transferring funds from
the project budget to the functional department’s budget. The charge for such machines will be
an operating cost ($/hr or $/operating cycle), plus a depreciation charge based on either time
or number of operating cycles. Use of general office equipment, e.g., copy machines, drafting
equipment, and coffeemakers, is often included in the general overhead charge.
In addition to these charges, there is also the General and Administrative (G&A) charge.
This is composed of the cost of senior management, the various staff functions, and any other
expenses not included in overhead. G&A charges are a fixed percent of either the direct costs
or the total of all direct and indirect costs.
Thus, a fully costed work element would include direct costs (labor, resources, and spe-
cial machinery) plus overhead and G&A charges. We advise the PM to prepare two budgets,
one with overheads and G&A charges, and one without. The full cost budget is used by the
accounting group to estimate the profit earned by the project. The budget that contains only
direct costs gives the PM the information required to manage the project without being con-
founded with costs over which the PM has no control. Let us now consider a combination of
top-down and bottom-up budgeting.
*In a weak matrix project, the Technical Assistance Group representing the technician would submit a lump-sum
charge to the project, calculated in much the same way. The charge would, of course, include the costs noted in the rest of
this section.
81721_Ch07.indd 299 9/16/08 1:35:12 AM


Managing Costs at Massachusetts’ Neighborhood Health Plan
Between 1994 and 1996, Medicaid reduced its rate of analysis. From these they determined that to con-
reimbursement by 20 percent while the State of Mas- trol costs effectively, NHP’s contracting philosophy
sachusetts imposed higher eligibility requirements would have to change from the current 95 percent of
for health subscribers, thereby significantly reduc- all line items per episode to a fixed cost per episode
ing Neighborhood Health Plan’s (NHP) revenues and or per day per type of stay. The team then constructed
threatening its viability. In the past, NHP had con- a spreadsheet that allowed cost comparisons to be
trolled costs by controlling hospital bed utilization and made across hospitals which allowed management to
increasing preventive medicine. However, no matter bargain for lower rates or, if hospitals were inflexible,
how low hospital utilization is, if hospital contract rates suggest to health centers what alternative hospitals
are expensive the cost to NHP will be high. Thus, in to refer patients to. This and later developments by
November 1995, NHP chartered a project team to help the team significantly enhanced management’s ability
it manage costs through better selection and manage- to contain their costs while guaranteeing that quality
ment of hospital contracts. More specifically, the team’s care would be available when needed. It also allowed
charter was to develop a method to examine hospital management to examine and respond to contracts and
contracts to assure that proposed rates were financially proposed contract changes in a timely and informed
viable to NHP but high-quality care would be available manner.
when needed. Source: J. H. Hertenstein and K. A. Vallancourt, “Contract
The team first selected the top 10 to 20 hospi- Management Cost Management,” PM Network, July 1997,
tals based on total annual payments from NHP for pp. 31–34.
An Iterative Budgeting Process—Negotiation-in-Action

In Chapter 6, we recommended an iterative planning process with subordinates* developing ac-
tion plans for the tasks for which they were responsible. Superiors review these plans, perhaps
suggesting amendments. (See also the latter part of Section 6.3.) The strength of this planning
technique is that primary responsibility for the design of a task is delegated to the individual
accountable for its completion, and thus it utilizes participative management (or “employee
involvement”). If done correctly, estimated resource usage and schedules are a normal part of
the planning process at all planning levels. Therefore, the superior constructing an action plan
at the highest level would estimate resource requirements and durations for each of the steps in
the action plan. Let us refer to the superior’s resource requirements for a particular task as R.
Similarly, the subordinate responsible for that task estimates the resource requirements as r.
In a perfect world, R would equal r. We do not, however, live in a perfect world. As a
matter of fact, the probable relationship between the original estimates made at the differ-
ent levels is R << r. This is true for several reasons, three of which are practically universal.
First, the farther one moves up the organizational chart away from immediate responsibility
for doing the work, the easier, faster, and cheaper the job looks to the superior than to the one
*We use the terms “superior” and “subordinate” here for the sole purpose of identifying individuals working on
different relative levels of a project’s action plan. We recognize that in a matrix organization it is not uncommon for PMs
(“superiors”) to delegate work to individuals (“subordinates”) who do not report to the PM and who may even be senior to
the PM on the parent firm’s organizational chart.
81721_Ch07.indd 300 9/16/08 1:35:12 AM

who has to do it. This is because the superior either does not know the details of the task, or
has conveniently forgotten the details, as well as how long the job takes and how many prob-
lems can arise. Second, wishful thinking leads the superior to underestimate cost (and time),
because the superior has a stake in representing the project to senior management as a profit-
able venture. Third, the subordinate is led to build-in some level of protection against failure
by adding an allowance for “Murphy’s Law” onto a budget that already may have a healthy
contingency allowance.
Assuming that the superior and subordinate are reasonably honest with one another (any
other assumption leads to a failure in win-win negotiations), the two parties meet and review
the subordinate’s action plan. Usually, the initial step toward reducing the difference in cost
estimates is made by the superior who is “educated” by the subordinate in the realities of
the job. The result is that the superior’s estimate rises. The next step is typically made by the
subordinate. Encouraged by the boss’s positive response to reason, the subordinate surrenders
some of the protection provided for by the budgetary “slop,” and the subordinate’s estimate
falls. The subordinate’s cost estimate is still greater than the superior’s, but the difference is
considerably decreased.
The pair now turn their attention to the technology of the task at hand. They carefully
inspect the subordinate’s work plan, trying to find a more efficient way to accomplish the
desired end; that is, they practice total quality management (TQM) and/or value engineering.
It may be that a major change can be made that allows a lower resource commitment than
either originally imagined. It may be that little or no further improvement is possible. Let us
assume that moderate improvement is made, but that the subordinate’s estimate is still some-
what greater than the superior’s, although both have been altered by the negotiations thus far.
What should the superior do, accept the subordinate’s estimate or insist that the subordinate
make do with the superior’s estimate? In order to answer this question, we must digress and
reconsider the concept of the project life cycle. In Chapter 1, we presented the usual view of
the project life cycle in Figure 1-3, shown here as Figure 7-1 for convenience. This view of the
life cycle shows decreasing returns to inputs as the project nears completion. Figure 1-5 is also
shown here as Figure 7-2 for convenience. In this case, the project shows increasing returns to
inputs as the project nears completion. In order to decide whether to adopt the subordinate’s
100
Slow finish
% Project completion
Quick momentum
Slow start
0
Figure 7-1 The project life cycle. (Figure 1-3
Time reproduced.)
81721_Ch07.indd 301 9/16/08 1:35:12 AM

100
% Project completion
Figure 7-2 Another possible project life

0
Time cycle. (Figure 1-5 reproduced.)
resource estimate or the superior’s, we need to know which picture of the life cycle represents
the task under consideration. Note that we are treating the subordinate’s action plan as if it
were a project, which is perfectly all right because it has the characteristics of a project that
were described in Chapter 1. Also note that we do not need to know the shape of the life cycle
with any precision, merely if its last stage is concave or convex to the horizontal axis.
Remember that the superior’s and subordinate’s resource estimates are not very far apart
as a result of the negotiations preceding this decision. If the latter part of the life cycle curve
is concave (as in Figure 7-1), showing diminishing marginal returns, we opt for the superior’s
estimate because of the small impact on completion that results from withholding a small
amount of resources. The superior might say to the subordinate, “Jeremy, what can you get
me for $R? We will have to live with that.” If, on the other hand, the life cycle curve is con-
vex, showing increasing marginal returns as in Figure 7-2, the subordinate’s estimate should
be chosen because of the potentially drastic effect a resource shortage would have on project
completion. In this event, the superior might say, “OK, Brandon, we have got to be sure of
this job. We’ll go with your numbers.” If the disagreement had concerned schedule (duration)
instead of resources, the negotiation process and underlying logic would be unaltered.
This is a time-consuming process. At the same time the PM is negotiating with the sev-
eral subordinates responsible for the pieces of the PM’s action plan, each of the subordinates
is negotiating with their subordinates, and so on. This multilevel process is messy and not
particularly efficient, but it allows a free-flow of ideas up and down the system at all levels.
This iterative process tends to reduce the uncertainty in budget estimations. The debate over
processes and their associated costs means that the uncertainty in budget estimates is very
likely to be reduced.
It is worth emphasizing that ethics is just as important in negotiations within an organization
as in negotiations between an organization and an outside party. In this case, the superior and
subordinate have the responsibility to be honest with each other. For one thing, they must
continue to work together in the future under the conditions of mutual trust. Second, it is ethi-
cally necessary to be honest in such negotiations.
Comments on the Budget Request Process

The budget process often begins with an invitation from top management for each division
to submit a budget request for the coming year. Division heads pass the invitation along to
departments, sections, and subsections, each of which presumably collects requests from
below, aggregates them, and passes the results back up the organizational ladder.
This sounds like bottom-up budgeting, but there is an important difference between this
procedure and a true bottom-up system. Along with the formal invitation for submission of a
budget request, in the iterative system another message is passed down—a much less formal
message that carries the following kinds of information: the percent by which the wage bill of
81721_Ch07.indd 302 9/16/08 1:35:12 AM

Source: DILBERT: © Scott Adams/Dist. by United Feature Syndicate, Inc.
the organization may be increased or must be decreased, organizational policy on adding to

or cutting the work force, the general attitude toward capital expenditures, knowledge about
which projects and activities are considered to be high priority and which are not, and a
number of other matters that, in effect, prescribe a set of limits on lower-level managers.
As the budget requests are passed back up the organization, they are carefully inspected for
conformity to guidelines. If they do not conform, they are “adjusted,” often with little or no
consultation with the originating units. Senior management tends to adopt an autocratic stance
on budget making for many reasons, but two are very common: the need to feel in control of
the budget, and the feeling that a tight budget will somehow motivate subordinates to perform
more efficiently. We know of no particular evidence to support such views, but they are quite
common. Moreover, they lead to budgetary game playing and increase the uncertainty sur-
rounding the budgetary process.
The less autocratic the organization (and the less pressured it is by current financial
exigencies), the greater the probability that this process will allow dialogue and some com-
promise between managerial levels. Even the most participative firms, however, will not long
tolerate lower-level managers who are not sensitive to messages relating to budget limitations.
It makes little difference whether budget policy is passed down the system by means of for-
mal, written policy statements or as a haphazard set of oral comments informally transmitted
by some senior managers and practically neglected by others; the PM’s budget request is
expected to conform to policy. Ignorance of the policy is no excuse. Repeated failure to con-
form will be rewarded with a ticket to “corporate Siberia.” It is the budget originator’s respon-
sibility to find out about budget policy. Again we see the importance of political sensitivity.
The PM’s channels of communication must be sensitive enough to receive policy signals even
in the event that a noncommunicative superior blocks those signals.
Cost Category Budgeting vs. Program/Activity Budgeting

Another facet of budgeting has to do with the degree to which a budget is category-oriented
or program/activity-oriented, a distinction we have mentioned before. The traditional orga-
nizational budget is category-oriented often based upon historical data accumulated through
a traditional, category-based, cost accounting system (Coburn, 1997; Vandament et al.,
1993). Individual expenses are classified and assigned to basic budget lines such as phone,
materials, personnel-clerical, utilities, direct labor, etc. or to production centers or processes
(Brimson, 1993). These expense lines are gathered into more inclusive categories, and are
reported by organizational unit—for example, by section, department, and division. In other
words, the budget can be overlaid on the organizational chart. Table 7-1 shows one page of
a typical, category-oriented monthly budget report for a real estate project.
81721_Ch07.indd 303 9/16/08 1:35:13 AM

Table 7-1. Typical Monthly Category Budget for a Real Estate Project (page 1 of 6)
Current
Actual Budget Variance Pct.
Corporate—Income Statement
Revenue
8430 Management fees
8491 Prtnsp reimb—property mgmt 7,410.00 6,222.00 1,188.00 119.0
8492 Prtnsp reimb—owner acquisition .00 3,750.00 3,750.00 – .0
8493 Prtnsp reimb—rehab .00 .00 .00 .0
8494 Other income .00 .00 .00 .0
8495 Reimbursements—others .00 .00 .00 .0
Total revenue 7,410.00 9,972.00 2,562.00– 74.3
Operating expenses
Payroll & P/R benefits
8511 Salaries 29,425.75 34,583.00 5,157.25 85.0
8512 Payroll taxes 1,789.88 3,458.00 1,668.12 51.7
8513 Group ins & med reimb 1,407.45 1,040.00 387.45 – 135.3
8515 Workmen’s compensation 43.04 43.00 .04 – 100.0
8516 Staff apartments .00 .00 .00 .0
8517 Bonus .00 .00 .00 .0
Total payroll & P/R benefits 32,668.12 39,124.00 6,457.88 83.5
Travel & entertainment expenses

8512 Travel 456.65 300.00 156.65 – 152.2
8522 Promotion, entertainment & gift 69.52 500.00 430.48 13.9
8523 Auto 1,295.90 1,729.00 433.10 75.0
Total travel & entertainment exp 1,822.07 2,529.00 706.93 72.1
Professional fees
8531 Legal fees 419.00 50.00 369.00 – 838.0
8532 Accounting fees 289.00 .00 289.00 – .0
8534 Temporary help 234.58 200.00 34.58 – 117.2
With the advent of project organization, it became necessary to organize the budget
in ways that conformed more closely to the actual pattern of fiscal responsibility. Under
traditional budgeting methods, the budget for a project could be split up among many different
organizational units, which diffused control so widely that it was frequently nonexistent. It
was often almost impossible to determine the actual size of major expenditure activities in a
project’s budget. In light of this problem, ways were sought to alter the budgeting process so
that budgets could be associated directly with the projects that used them. This need gave rise
to program budgeting. Table 7-2 shows a program-oriented project budget divided by task/
activity and expected time of expenditure. In an interesting paper, Brimson (1993) critiques
both systems separately, and then combines them.
If a program consists of a set of separate projects, the use of program budgeting for each
project allows those project budgets to be aggregated for the program as a whole by time
periods. Moreover, the program can also have it’s own monthly category budget (as shown in
Table 7-1, with the categories down the left side), but this will require dividing up the revenues
81721_Ch07.indd 304 9/16/08 1:35:13 AM

7.2 IMPROVING THE PROCESS OF COST ESTIMATION 305
Table 7-2. Project Budget by Task and Month

Monthly Budget (£)
Task Estimate 1 2 3 4 5 6 7 8
A 7000 5600 1400
B 9000 3857 5143
C 10000 3750 5000 1250
D 6000 3600 2400
E 12000 4800 4800 2400
F 3000 3000
G 9000 2571 5143 1286
H 5000 3750 1250
I 8000 2667 5333
J 6000 6000
75000 5600 12607 15114 14193 9836 6317 5333 6000
Source: Harrison, 1983.
(if any) and expenses for each of the projects’ activities (tasks) into the appropriate categories.
As well, each organizational unit can also present its projects’ revenues and expenses by add-
ing additional columns to Table 7-1, with one column for “regular operations” and the other
columns for each project (or program). Again, however, this may take extra effort to break out
the tasks by organizational unit as well as activity if the tasks involve more than one organi-
zational unit.
The estimation of capital costs raises special problems. Accounting systems in differ-
ent industries handle capital costs differently. Further, estimation requires highly specialized
knowledge because the prices of some durable goods, e.g., machine tools, rise and fall in
response to much different forces than affect the prices of other equipment, e.g., computer
systems or aircraft. In an interesting two-part article, Sigurdsen (1996a, 1996b) notes that
capital costs are variant with quantity of output and compares two methods of making capital
cost estimates.
7.2 IMPROVING THE PROCESS OF COST ESTIMATION
The cooperation of several people is required to prepare cost estimates for a project. If the
firm is in a business that routinely requires bids to be submitted to its customers, it will have
“professional” (experienced) cost estimators on its staff. The major responsibility of the pro-
fessional estimators is to reduce the level of uncertainty in cost estimations so that the firm’s
bids can be made in the light of expert information about its potential costs. In these cases, it
is the job of the PM to generate a description of the work to be done on the project in sufficient
detail that the estimator can know what cost data must be collected. Frequently, the project
will be too complex for the PM to generate such a description without considerable help from
experts in the functional areas.
Even with the finest of experts working to estimate resource usage, the one thing that is
certain is that things will not go precisely as planned. There are two fundamentally different
ways to manage the risks associated with the chance events that occur on every project. The
simpler and far more common way is to make an allowance for contingencies—usually 5 or
10 percent of the estimated cost. Just why these numbers are chosen in preference to 678 or
914 for instance, we do not know. We strongly prefer another method in which the forecaster
81721_Ch07.indd 305 9/16/08 1:35:14 AM


Completing the Limerick Nuclear Facility Under Budget
When the Limerick nuclear power generating facility to work with the manual workers. In this fashion, the
in Pennsylvania began commercial operation, it set second shift productivity was equal to, if not higher
a construction record for nuclear facilities. In an era than, the first shift’s.
when it is common to hear of nuclear plants that mas- Other decisions and actions further helped either
sively overrun their budgets and completion sched- the cost or the schedule. For example, it was decided
ules, Limerick was completed eight months ahead that overtime would not be worked since a second
of its 49-month schedule and came in $400 million shift was being used. And as a condition of the proj-
under its $3.2 billion budget. Limerick has truly set a ect approval, a project labor agreement with the local
standard for the industry. unions (rather than the national) had to be developed
It was no accident that Limerick was completed that would eliminate strikes, lockouts, and delays and
ahead of schedule and under budget. When construc- provide for peaceful resolution of disputes. Also, an
tion started in February 1986, a project goal was incentive fee contract with the building contractor was
to complete the project eight months ahead of the signed whereby the contractor would share equally in
planned completion, which would help keep the costs cost/schedule overruns or underruns, with limits set.
under the budget limit as well. To achieve this early With such attention to the goal of an early and under-
target, a series of innovative approaches were taken. budget completion, the team, numbering almost 3000
Two of the major ones were to accelerate ramp-up workers at the start, worked diligently and with high
staffing and to use an extensive, fully-supported sec- morale, beating its 4-year deadline by almost a year!
ond shift. The momentum of the speedy start-up set
the fast pace for the remainder of the project. The sec-
ond shift earned a very favorable premium, as well as Source: T. P. Gotzis, “Limerick Generating Station No. 2,” PM
having a full complement of managers and engineers Network, January 1991.
selects “most likely, optimistic, and pessimistic” estimates. We adopted this method in Chap-
ter 2 when we applied simulation to the discounted cash flow problem in the PsychoCeramic
Sciences case. The method is described in detail in Chapter 8 when we cover the issue of
estimating the duration of elements in the action plan. The method described in Chapter 8 is
applicable, unchanged, to the estimation of resource requirements and costs for the determina-
tion of project budgets.
Turning now to the problem of estimating direct costs,* project managers often find it
helpful to collect direct cost estimates on a form that not only lists the estimated level of
resource needs, but also indicates when each resource will be needed, and notes if it is avail-
able (or will be available at the appropriate time). Figure 7-3 shows such a form. It also has
a column for identifying the person to contact in order to get specific resources. This table
can be used for collating the resource requirements for each task element in a project, or for
aggregating the information from a series of tasks onto a single form.
*Our emphasis on estimating direct costs and on focusing on resources that are “direct costed” in the action plan is
based on our belief that the PM should be concerned with only those items over which he or she has some control—which
certainly excludes overheads. The PM, however, may wish to add some nonchargeable items to the resource column of the
action plan simply to “reserve” that item for use at a specific time.
81721_Ch07.indd 306 9/16/08 1:35:14 AM

Note that Figure 7-3 contains no information on overhead costs. The matter of what
overhead costs are to be added and in what amounts is unique to the firm, beyond the PM’s
control, and generally a source of annoyance and frustration to one and all. The allocation of
overhead is arbitrary by its nature, and when the addition of overhead cost causes an otherwise
attractive project to fail to meet the organization’s economic objectives, the project’s supporters
are apt to complain bitterly about the “unfairness” of overhead cost allocation.
At times, firms support projects that show a significant incremental profit over direct
costs but are not profitable when fully costed. Such decisions can be justified for a number of
reasons, such as:
• To develop knowledge of a technology
• To get the organization’s “foot in the door”
Project Name
Date
Task Number
RESOURCES NEEDED
Person to How Many/ Check ( ) If

Resources Contact Much Needed When Needed Available
People:
Managers,
Supervisors
Professional &
Technical
Nontechnical
Money
Materials:
Facilities
Equipment
Tools
Power
Space
Special Services:
Research & Test
Typing/clerical
Reproduction
Others
Figure 7-3 Form for gathering data on project resource needs.
81721_Ch07.indd 307 9/16/08 1:35:15 AM

• To obtain the parts or service portion of the work

• To be in a good position for a follow-on contract
• To improve a competitive position
• To broaden a product line or a line of business
All of these are adequate reasons to fund projects that, in the short term, may lose money
but provide the organization with the impetus for future growth and profitability. It is up to
senior management to decide if such reasons are worth it.
Learning Curves*
If the project being costed is one of many similar projects, the estimation of each cost ele-
ment is fairly routine. If the project involves work in which the firm has little experience, cost
estimation is more difficult, particularly for direct labor costs. For example, consider a project
that requires 25 units of a complex electronic device to be assembled. The firm is experienced
in building electronic equipment but has never before made this specific device, which differs
significantly from the items routinely assembled.
Experience might indicate that if the firm were to build many such devices, it would use
about 70 hours of direct labor per unit. If labor is paid a wage of $12 per hour, and if benefits
equal 28 percent of the wage rate, the estimated labor cost for the 25 units is
(1.28)($12/hr)(25 units)(70 hr/unit) $26,880
In fact, this would be an underestimate of the actual labor cost because more time per unit output
is used early in the production process. Studies have shown that human performance usually
improves when a task is repeated. In general, performance improves by a fixed percent each
time production doubles. More specifically, each time the output doubles, the worker hours per
unit decrease to a fixed percentage of their previous value. That percentage is called the learning
rate. If an individual requires 10 minutes to accomplish a certain task the first time it is attempted
and only 8 minutes the second time, that person is said to have an 80 percent learning rate. If
output is doubled again from two to four, we would expect the fourth item to be produced in
8(0.8) 6.4 min
Similarly, the eighth unit of output should require
6.4(0.8) 5.12 min
and so on. The time required to produce a unit of output follows a well-known formula:
Tn T1nr
where
Tn the time required for the nth unit of output,
T1 the time required for the initial unit of output,
n the number of units to be produced, and
r log decimal learning rate/log 2.
The total time required for all units of a production run of size N is
N
total time T1 a nr
n 1
* Occasionally, particular sections will be shaded, meaning that they can be skipped without loss of continuity.
81721_Ch07.indd 308 9/16/08 1:35:15 AM

Tables are widely available with both unit and total values for the learning curves, and
have been calculated for many different improvement ratios (learning rates—e.g., see Shafer
et al., 1998).
In the example of the electronic device just given, assume that after producing the twenti-
eth unit, there is no significant further improvement (i.e., assembly time has reached a steady
state at 70 hours). Further assume that previous study established that the usual learning rate
for assemblers in this plant is about 85 percent. We can estimate the time required for the first
unit by letting Tn 70 hours by the unit n 20. Then
r log 0.85/log 2
0.1626/0.693
0.235
and
70 5 T1 1 20 2 r
T1 5 141.3 hr
Now we know the time for the initial unit. Using a table that shows the total time multi-
plier (see Shafer et al., 1998, pp. 343–346, for example), we can find the appropriate total time
multiplier for this example—the multiplier for 20 units given a learning rate of 85 percent.
With this multiplier, 12.40, we can calculate the total time required to build all 20 units. It is
1 12.40 2 1 141.3 hr 2 1752.12 hr
The last five units are produced in the steady-state time of 70 hours each. Thus the total
assembly time is
1752.12 5 1 70 hr 2 2102.12 hr
We can now refigure the direct labor cost.
2102.12($12)(1.28) $32,288.56
Our first estimate, which ignored learning effects, understated the cost by
$32,288.56 $26,880 $5,408.56
or about 17 percent. Figure 7-4 illustrates this source of the error.
In recent years, learning curves have received increasing interest from project manag-
ers, particularly in the construction industry. Methods have been developed for the approxi-
mation of composite learning curves for entire projects (Amor et al., 1998), and for the
approximation of total cost from the unit learning curve (Camm et al., 1987). Badiru (1995)
has included learning curve effects in his critical resource diagramming, which is discussed
in Chapter 8.
Remember that we are attempting to reduce the risk inherent in estimating costs. There-
fore, for any task where labor is a significant cost factor and the production run is reasonably
short, the PM should take the learning curve into account when estimating costs.
The implications of this conclusion should not be overlooked. We do not often think of
projects as “production,” but they are. While the construction, electronics, and aircraft assem-
bly industries have used learning curves for many years, other industrial areas have been
slow to follow. For example, research (Gagnon et al., 1987) has shown that the learning curve
effect is important to decisions about the role of engineering consultants on computer-assisted
design (CAD) projects. The same is assuredly true for the design of advertising campaigns
or charity drives. The failure to consider performance improvement is a significant cause of
errors in project cost estimation.
81721_Ch07.indd 309 9/16/08 1:35:15 AM

150
Additional time used for learning

125
Time as charged without
allowance for learning
100
Time (hours)
75
(70)
50
25
0
0 4 8 12 16 20 24 28 32 36
Figure 7-4 Effects of ignoring learning curve. Number of units
A Special Case of Learning—Technological Shock

If the parent organization is not experienced in the type of project being considered for selec-
tion, performance measures such as time to installation, time to achieve 80 percent efficiency,
cost to install, and the like are quite uncertain and often will be seriously underestimated. It is
interesting to observe that an almost certain, immediate result of installing a new, cost-saving
technology is that costs rise. Sometimes we blame the cost increases on resistance to change,
but a more sensible explanation is that when we alter a system, we disturb it and it reacts in
unpredictable ways. A steelmaker recalling the installation of the then new technology for
manufacturing tinplate by electrolysis remarked, “We discovered and installed the world’s
first electrolytic method for making scrap. It took a year before we had that line running the
way it was designed.” Of course, if the organization is experienced, underestimation is not
likely to be a serious problem. The Reliance Electric Company undertook several “18-month”
plant construction projects that they predicted, accurately, would require 36 months to build
from decision to the point when the plant was capable of operating at or above three-fourths
capacity. (Note the potential for ethical problems here.) To the extent possible, past knowledge
of system actions and reactions should be built into estimates of future project performance.
Other Factors
Depending on the reference, anywhere from about three-fifths to five-sixths of projects fail
to meet their time, cost, and/or specification objectives (see, for example, Frame, 1998).
The record of Information Technology (IT) projects is particularly poor, according to arti-
cle after article in the journals of the Project Management Institute. Possibly the problem is
that Dilbert’s pointy-haired boss sets arbitrary and impossible goals. Possibly scope-creep
impacts all projects [though cost overruns are not necessarily associated with changing scope
(Christensen et al., 1998)]. Possibly PMs use wildly optimistic estimates in order to influence
81721_Ch07.indd 310 9/16/08 1:35:16 AM

the project selection process. Or maybe they are simply unaware of good cost (or time) esti-
mating practices. For example, there are at least 45 estimating models available for IT projects,
but few IT project managers use any of them (Lawrence, 1994; Martin, 1994). Some IT work-
ers flatly refuse to estimate time and cost for IT projects on the grounds that there is too much
uncertainty—and, we suspect, in an attempt to avoid responsibility. Possibly all of these
things, and even others, act together to produce this unenviable track record.
While the number of things that can increase risk by producing errors in cost estimates
is almost without limit, some problems occur with particularly high frequency. Changes in
resource prices are one of these. The most commonly used solution to this problem is to in-
crease all cost estimates by some fixed percentage. A more useful approach is to identify each
input that accounts for a significant portion of project cost and estimate the direction and rate
of price change for each.
The determination of which inputs account for a “significant” portion of project cost is
not difficult, although it may be somewhat arbitrary. Suppose, for example, that our initial,
rough cost estimate (with no provision for future price changes) for a project with an objec-
tive of setting up a small storefront accounting office is $1 million and is to be spent over a
three-year period in approximately equal amounts per year. If we think personnel costs will
comprise about 60 percent of that total, also spread equally over time, the wage/salary bill will
be about $600,000. Split into three equal amounts, we have expenditures of $200,000 per year.
If we estimate that wage/salary rates will increase by 6 percent per year, our expenses for the
second year rise to $212,000 (an increase of $12,000), and to $224,720 in the third year (a
further increase of $12,720). Failure to account for wage/salary inflation would result in an
underestimate of project cost of $24,720. This is an error of slightly more than 4 percent of the
estimated personnel cost and almost 2.5 percent of the total project budget.
Further improvements can be made by taking into account the fact that the prices of
different inputs often change at very different rates and sometimes in different directions.
A quick examination of the Bureau of Labor Statistics (BLS) wage and price indices, which
cover a very large number of specific commodities and wage rates, will reveal that even in
periods of stable prices, the prices of some things rise while others fall and still others do not
change appreciably. Thus, the PM may wish to use different inflators/deflators for each of sev-
eral different classes of labor or types of commodities. While most PMs are concerned only
with price increases, any industry submitting competitive bids on projects must remember that
failure to be aware of falling prices will lead to cost overestimation and uncompetitive bids.
The proper level of breakdown in estimating the impact of price changes simply depends
on the organization’s willingness to tolerate error. Assume that management is willing to ac-
cept a 5 percent difference between actual and estimated cost for each major cost category.
In the example above, expected increases in wage/salary costs will use more than four-fifths
of that allowance. That leaves less than one-fifth (about $5,300) of the allowable error, and
the need to add one part-time clerk to the project for a single year would more than use the
remaining allowance.
Other elements that need to be factored into the estimated project cost include an allow-
ance for waste and spoilage. No sane builder would order “just enough” lumber to build a
house. Also, personnel costs can be significantly increased by the loss and subsequent re-
placement of project professionals. Not only must new people go through a learning period—
which, as we have seen, will have a negative effect on production—but professional starting
salaries often rise faster than the general rate of annual salary increases. Thus, it may well cost
more to replace a person who leaves the project with a newcomer who has approximately the
same level of experience.
We have already mentioned the inclination PMs and project sponsors have toward under-
stating the costs of a project in order to make it appear more profitable to senior managers, as
81721_Ch07.indd 311 9/16/08 1:35:16 AM

well as the proclivity of lower-level project workers to overestimate costs in order to protect
themselves. If the project is in its initial planning stage as a response to a Request for Proposal
(RFP) from an outside organization, over- and underestimation of cost can have a serious im-
pact on the probability of winning the contract—or on the level of profit, if a win occurs.
Serious ethical problems may arise during the process of estimating costs and submis-
sion of bids in response to an RFP. If the job is to be paid on a cost-plus basis, or even if it is
a fixed-fee project, with fee increases allowed for special circumstances, some bidders may
“low ball” a contract (submit underestimated costs). By doing this, they hope to win the bid,
counting on the opportunity to increase costs or to plead special circumstances once the job is
underway. At times, clients have been known to give favored bidders a “last look” at suppos-
edly sealed bids so that the favored bidder can submit a winning bid, often with an unwritten
agreement to allow some cost escalation at a later date. There is considerable opportunity for
unethical behavior during cost estimation and bidding. Further, estimation and bidding prac-
tices vary widely from industry to industry.
Finally, there is plain bad luck. Delays occur for reasons that cannot be predicted. Ma-
chinery with the reliability of a railroad spike suddenly breaks down. That which has never
failed fails. Every project needs an “allowance for contingencies.”
Some writers and instructors differentiate four bases for estimating costs: experience,
quantitative (statistical) methods, constraints, and worksheets. They discuss the advantages
and disadvantages of each and then, typically, decide that one or another gives the best re-
sults. We feel strongly that all four are useful and that no single approach to cost estimation
should be accepted as the best or rejected out of hand. The best estimators seem to employ
an eclectic approach that uses, as one said, “anything that works.” The wise PM takes into
account as many known influences on the project budget as can be predicted. What cannot
be predicted must then, by experience, simply be “allowed for.” There are two other factors,
particularly common to projects involving intangible outputs such as software programming,
that need to be mentioned relating to cost-estimation and the schedule. These two factors
have been identified in a classic and highly readable work—The Mythical Man-Month—by
Brooks (1975).
First, most projects involve a tangible medium that tends not to be under our control—the
wood splits, the paint smears—and thus we blame implementation problems of our “good”
ideas on these physical elements. So, when we are working with a purely intellectual medium
that has no physical elements, such as computer code, we are highly optimistic and foolishly
assume that all will go well. However, when any project consisting of a series of components
can only be successful if all of the components are successful, and each component has a
small probability of failing, the chances of the overall project being successful may be very
poor. Consider, for example, a software program consisting of 1000 lines of code, each of
which is 0.999 reliable. The chance of the program itself working is only about 36 percent!
Brooks’ experience has led him to the following rule of thumb for software projects. As a
fraction of the total time of the project, planning consumes about 13 , coding consumes 16 , com-
ponent test consumes 14 , and system test consumes 14 . Thus, if a project estimate is made based
on the expected coding time (the main element for which we can derive an estimate), this in
reality will usually represent only about 17 percent of the entire project time rather than the
80 to 90 percent commonly assumed.
The second factor is what Brooks calls “the mythical man-month” and relates to our
tendency to assume that workers and time are interchangeable. Thus, when a schedule slips,
the traditional response is to add labor, which is like trying to douse a fire with gasoline. Our
assumption that workers and time are interchangeable is correct only when a task can be
partitioned such that there is no communication needed between the workers. Most projects,
however, especially computer programming, are not set up that way and the more workers that
81721_Ch07.indd 312 9/16/08 1:35:16 AM

are added require even more workers to train, as well as lines of communication to coordinate
their efforts. Thus, three workers require three times as much pairwise intercommunication as
two, and four require six times as much, etc. This result is captured in Brooks’ law: Adding
manpower to a late software project makes it later.

The Emanon Aircraft Corporation
Emanon Aircraft is a major manufacturer of aircraft They determined that the manufacturing process
parts, specializing in landing gear parts and assem- was reasonably efficient and not significantly differ-
blies. They are located in a highly industrialized mid- ent from Emanon’s competitors. Second, they found
western state. The local area suffers from somewhat that all competitors were using approximately the
higher than average unemployment, partly because same level of mark-up when determining their cost-
Emanon has experienced a downturn in business. In plus price. When examining the cost structure, how-
the past three years, they have lost out on a number of ever, they noted that in the past three years, the firm
landing gear contracts, being underbid by competitors consistently ran negative cost variances in material
from other areas of the country. Senior management accounts. That is, the amount of material actually
studied the problem, but has come to no conclusion used in the construction of landing gears was approx-
about what can be done. They have hired a consulting imately 10 percent less than the plan indicated. The
team from a nearby university to study the situation team was unsure of this finding because there were
and make a recommendation. only a few winning contracts for landing gears dur-
Business in the aircraft industry is not significantly ing the past three years.
different than in many other industries specializing in An investigation was conducted on the estima-
the building of complex machines. Aircraft builders tion and purchase of materials for this department.
are primarily assembly operations. They build planes It exposed the following facts. Three-and-one-half
from subassemblies and parts manufactured by them- years ago, Emanon was late making a delivery of
selves or by subcontractors who, in turn, specialize in landing gear parts. The firm paid a large penalty and
specific subassemblies; for example, landing gear, av- was threatened with loss of further business with the
ionics, passenger seats, heating and air conditioning, prime contractor. The late delivery resulted when
etc. When an order is received to build some number Emanon ordered an insufficient quantity of a special
of a given type of plane, the builder (prime contrac- steel alloy used in landing gear struts, and was un-
tor) requests bids for the proper number of a certain able to purchase any on the open market. The steel
part or subassembly from appropriate subcontractors. company required a manufacturing lead time of more
All relevant specifications for the part or subassem- than 90 days, so Emanon’s delivery was late.
bly are included in the RFP. The subcontractors who As a result, the purchasing official who had re-
wish to participate in the project submit proposals sponsibility for this contract was demoted. The new
that include a complete description of the proposed purchasing official handled the problem in a straight-
subassembly together with price information, deliv- forward fashion by inflating the material estimates
ery dates, and any other pertinent conditions of sale. by approximately 10 percent. The cost of material is
The university consulting team studied three as- about half of the total cost of landing gear produc-
pects of Emanon’s landing gear operation: the manu- tion, which resulted in bids that were approximately
facturing process, the cost structure, and the bidding 5 percent above the competition..
behavior and profit structure on landing gear bids. Source: S. J. Mantel, Jr. Consulting project.
81721_Ch07.indd 313 9/16/08 1:35:17 AM

On Making Better Estimates*

Let us begin with the assumption that budget estimation errors are not the result of deliberate
dishonesty, but derive from honest errors on the part of the PM, project cost estimators, or
anyone else involved. We have already noted that there are many reasons why “honest” errors
occur—that projects look easier, faster, and cheaper to senior managers than to those who
must do the work that nonexpert estimators tend to overlook details required to do a complete
job—that everyone seems to assume that Murphy’s law has been repealed for their personal
project. But these things should be learned by experience. Why do experienced managers err
on their budget and schedule estimates?
Ambrose Bierce, in The Devil’s Dictionary, defined “experience” as “The wisdom that
enables us to recognize as an undesirable old acquaintance the folly that we have already em-
braced.” It is axiomatic that we should learn through experience. It is a truism that we do not.
Nowhere is this more evident than in project management, and yet it is not difficult to improve
one’s estimation/forecasting skills.
Recall that there are two generic types of estimation error. First, there is random error in
which overestimates and underestimates are equally likely. These are errors that tend to cancel
each other. If we sum these errors over many estimates, their sum will approach zero. Second,
there is bias, which is systematic error. For biased estimates, the chance of over- and under-
estimates are not equally likely, and their sum, either positive or negative will increase as the
number of estimates increases. A measure for bias, the “tracking signal” (TS), is given in Row 4
of Figure 7-5. Using the ubiquitous Excel® we can construct a spreadsheet that captures the
essence of a person’s performance as an estimator. Two simple statistical measures are used:
the mean absolute deviation (MAD), and the tracking signal (TS). The printout† of such a
spreadsheet is shown in Figure 7-5. Appendix A of this book (on the book’s website) includes
information on probability, statistics, and forecasting.
Figure 7-5 assumes that for each period (Column A) someone has made an estimate of
a variable (Column B), and that the actual value of that variable is, sooner or later, known
(Column C). (It should be noted that Column A need not be time periods. This column simply
counts the number of estimates made and links estimates with their respective actuals.) Col-
umn D calculates the difference between the actual value, A(t), and the estimate or forecast
for that period, F(t). Column E contains the absolute value of that difference. We can now
calculate a statistic known as the mean absolute deviation (MAD).
As the information in Row 3 of the spreadsheet shows,
MAD a 1 0 A 1 t 2 F 1 t 2 0 2 /n
where n is the number of differences. The MAD is therefore the arithmetic average of the
absolute values of the differences—the mean absolute deviation.
Students of statistics may note that the MAD has certain logical similarities to the
standard deviation. Assuming that the forecast errors are normally distributed, the MAD is
approximately 80 percent of a standard deviation (see Chase et al., 1989). Thus, if the MAD
is a sizable fraction of the variable being estimated, the average error is large and the forecast
or estimate is not very accurate, but this does not mean that it is biased.
Now, consider Column D. The sum of the entries in this column for any number of peri-
ods is the sum of the forecast errors, often referred to as the “running sum of the forecast
errors” (RSFE). As we noted, if the estimator’s errors are truly random, and unbiased, their
sum should approach zero. In Figure 7-5, RSFE 133, so the forecast is quite biased.
* Occasionally, particular sections will be shaded, meaning that they can be skipped without loss of continuity.
†Any of the common spreadsheet programs can easily handle all of the calculations shown in this chapter and will
accept formulas and calculations from any of the others.
81721_Ch07.indd 314 9/16/08 1:35:17 AM

A B C D E F G
1 This is a template for improving one’s estimating skills
2
3 MAD = SUM (|A(t) − F (t)|)/n The Average Absolute Error
4 Tracking Signal = SUM(A(t) − F(t))/MAD A Measure of Bias
5 Tracking
6 Period Estimate Actual A(t) – F(t) |A(t) – F(t)| MAD Signal
7 ====== ======== ======== ======== ======== ======== ========
8
9 1 155 163 8 8
10 2 242 240 −2 2 5.00 1.20
11 3 46 67 21 21 10.33 2.61
12 4 69 78 9 9 10.00 3.60
13 5 75 71 −4 4 8.80 3.64
14 6 344 423 79 79 20.50 5.41
15 7 56 49 −7 7 18.57 5.60
16 8 128 157 29 29 19.88 6.69
17
18 Key Formulae Total = 133 159
19 Cell D9 = (C9–B9) copy to D10:D16
20 Cell E9 = ABS(C9–B9) copy to E10:E16
21 Cell F10 = (Sum($E$9:E10))/A10 copy to F11:F16
22 Cell G10 = (Sum($D$9:D10))/F10 copy to G11:G16
Figure 7-5 Excel® template for cost estimation.
The tracking signal is a special measure of the estimator’s bias. It is easily found:
TS RSFE/MAD
Note that it calculates the number of MADs in the RSFE (see column G in Figure 7-5, and recall
the similarity between MAD and standard deviation). If the RSFE is small, approaching zero,
the TS will also approach zero. As the RSFE grows, the TS will grow, indicating bias. Division
of the RSFE by the MAD creates a sort of “index number,” the TS, a ratio that is independent of
the size of the variables being considered. We cannot say just how much bias is acceptable in an
estimator/forecaster. We feel that a TS 3 is too high unless the estimator is a rank beginner. Cer-
tainly, an experienced estimator should have a much lower TS. (It should be obvious that the TS
may be either negative or positive. Our comment actually refers to the absolute value of the TS.)
Perhaps more important than worrying about an acceptable limit on the size of the track-
ing signal is the practice of tracking it and analyzing why the estimator’s bias, if there is one,
exists. If the estimator can identify the source of the bias, then the forecasting process should
be corrected, with a concomitant improvement in the forecasts. If not, but the bias seems to
continue, the estimator can simply correct the forecasts by the amount of the known bias. It
has often been said that manufacturing managers do this when automatically cutting the Sales
Department’s upcoming quarterly sales forecasts by 20 percent.
81721_Ch07.indd 315 9/16/08 1:35:17 AM

Similarly, the estimator should consider how to reduce the MAD, the random error. Such
analysis is the embodiment of “learning by experience.” The Excel® template makes the analy-
sis simple to conduct, and should result in decreasing the size of both the MAD and the TS.
Some estimators would like to speed up the process of improving their estimation skills by
grouping forecasts of different resources to generate more data points when calculating their
MADs and TSs. Use of the tracking signal requires that the input data, estimates (forecasts)
and actuals, be collected and processed separately for each variable being estimated. Cost
estimates and actuals for different resources, for instance, would be used to find the MAD and
TS for each individual resource. The reason for this inconvenience is that resources come in
different units and the traditional caution about adding apples and oranges applies. (Even if all
resources are measured in dollars, we still have scale problems when we mix resource costs of
very different sizes.) Fortunately, there is a way around the problem.
Instead of defining the estimation error as the difference between actual and forecast, we
can define it as the ratio of actual to forecast. Therefore, the new error for the first forecast
(Period 1) in Figure 7-5 is not eight units, but rather is
A 1 t 2 /F 1 t 2 163/155 1.052
A B C D E F G
1 This is a template for improving one’s estimating skills
2
3 MAR = SUM {|(A(t) / F (t)) − 1|} /n
4 Tracking Signal = SUM{(A(t) / F (t)) − 1}/MAR
5 Tracking
6 Period Estimate Actual (A(t)/F(t)) − 1 |(A(t)/F(t)) − l| MAR Signal
7 ===== ======= ======= ======== ========== ===== ======
8
9 1 155 163 0.052 0.052
10 2 242 240 − 0.008 0.008 0.030 1.448
11 3 46 67 0.457 0.457 0.172 2.904
12 4 69 78 0.130 0.130 0.162 3.898
13 5 75 71 − 0.053 0.053 0.140 4.120
14 6 344 423 0.230 0.230 0.155 5.205
15 7 56 49 − 0.125 0.125 0.151 4.523
16 8 128 157 0.227 0.227 0.160 5.670
17
18 Key Formulae Total = 0.908 1.281
19 Cell D9 = (C9–B9)− 1 copy to D10:D16
20 Cell E9 = ABS((C9–B9)− 1) copy to E10:E16
21 Cell F10 = (Sum($E$9:E10))/A10 copy to F11:F16
22 Cell G10 = (Sum($D$9:D10))/F10 copy to G11:G16
Figure 7-6 Estimation template using ratios.
81721_Ch07.indd 316 9/16/08 1:35:18 AM

SUMMARY 317
or a 5.2 percent error. In order to produce measures similar in nature and concept to the MAD
and TS, we will subtract 1 from the ratio. Thus, when the actual is greater than the forecast,
the measure (i.e., the error ratio minus 1) will be positive, and if the actual is less than the
forecast, the measure will be negative. Figure 7-6 shows the calculations of {A(t)/F(t) – 1} for
the data used in Figure 7-5. Column E shows the absolute value of column D, and column F
lists the MAR (mean absolute ratio). The tracking signal is calculated as usual by dividing the
“running sum of the forecast ratios” (RSFR) by the MAR,
TS 5 RSFR/MAR
Notice that this calculation does not suffer from unit or scale effects because the ratio of ac-
tual to forecast is a dimensionless number and we are finding the percent error rather than the
“real” error.
One caution remains. While this technique will allow one to aggregate dissimilar data
and, thereby, measure the degree of random error and bias faster than when using differences,
care must be exercised to aggregate only data for which there is good reason to believe that the
amount of bias and uncertainty is roughly the same for all resource estimates.
At the beginning of this discussion, we made the assumption that estimation errors were
“honest.” That assumption is not necessary. If a manager suspects that costs are purposely
being under- or overestimated, it is usually not difficult to collect appropriate data and calcu-
late the tracking signal for an individual estimator—or even for an entire project team. If it is
known that such information is being collected, one likely result is that the most purposeful
bias will be sharply reduced.
It may occur to the reader that we have not applied simulation analysis, similar to that illus-
trated in Chapter 2, to deal with budgetary uncertainty. It should be obvious that such an appli-
cation is appropriate. For example, in addition to optimistic, pessimistic, and most likely cost
estimates, a number of other variables are uncertain—the inflation rate for various cost elements,
the learning rate, the amount of resource usage for individual tasks, and the timing of expendi-
tures, to mention a few. As we have noted above, we are saving this application for a detailed
description for the scheduling problem in Chapter 8. Its application to budgeting is quite similar.
A final note: In the next chapter, we discuss a method for managing schedule and budget
risk. We also describe some software used for risk management. Most of the useful methods
require that the forecaster specify the mode and range of the variable being forecast. These
models tacitly assume that the data supplied represent unbiased estimates. As we noted previ-
ously, if the input estimates are biased, the output will incorporate that error and the danger is
not “Garbage in, garbage out” but “Garbage in, gospel out.”
SUMMARY
This chapter initiated the subject of project implementa- Specific points made in the chapter were these:
tion by focusing on the project budget, which authorizes
the project manager to obtain the resources needed to • The intent of a budget is to communicate organiza-
tional policy concerning the organization’s goals and
begin work. Different methods of budgeting were de-
priorities.
scribed along with their impacts on project management.
Then, a number of issues concerning cost estimation were • There are a number of common budgeting methods:
discussed, particularly the effect of learning on the cost of top-down, bottom-up, the program budget.
repetitive tasks and how to use the concept of the learn- • A form identifying the level of resource need, when
ing curve. Finally, methods for improving cost estimation it will be needed, who the contact is, and its avail-
skills were described. ability is especially helpful in estimating costs.
81721_Ch07.indd 317 9/16/08 1:35:18 AM

• It is common for organizations to fund projects whose • Other major factors, in addition to learning, that
returns cover direct but not full costs in order to should be considered when making project cost
achieve long-run strategic goals of the organization. estimates are inflation, differential changes in the
• If projects include repetitive tasks with significant cost factors, waste and spoilage, personnel re-
human input, the learning phenomenon should placement costs, and contingencies for unexpected
be taken into consideration when preparing cost difficulties.
estimates. In the next chapter, we address the subject of task sched-
• The learning curve is based on the observation that uling, a topic of major importance in project management.
the amount of time required to produce one unit de- More research and investigation have probably been con-
creases a constant percentage every time the cumu- ducted on the subject of scheduling than any other element
lative output doubles. of project management.
• A method for determining whether or not cost esti-
mates are biased is described. The method can be used
to improve any estimation/forecasting process.
GLOSSARY
Bottom-up Budgeting A budgeting method that be- are charged as a lump sum or as a fixed percent of some
gins with those who will be doing the tasks estimating the direct cost such as labor.
resources needed. The advantage is more accurate esti- Overhead Cost Costs incurred by the firm, but not
mates. associated with any specific product or class of prod-
Cost Categories ucts; e.g., cost of building and ground maintenance,
Direct (or Variable) Cost These costs vary with utilities, cost of plant security, cost of health insurance
output; e.g., labor costs, material costs, and some- and pension plans. Typically charged as a fixed percent
times the cost of capital equipment such as machin- of some direct cost such as labor.
ery that performs a specific function on each unit of Learning Rate The percentage of the previous worker
output. hours per unit required for doubling the output.
General & Administrative Cost (G&A) The cost Program Budgeting Budgeting by project task/activity
of administration; e.g., Accounting, Human Resources, and then aggregating income and expenditures by project
and Legal not charged as an Indirect Cost and not in- or program, often in addition to aggregation by organiza-
cluded in Overhead Cost. Sometimes the G&A is not tional unit or category.
reported as a separate item but is included in overhead Top-Down Budgeting A budgeting method that begins
cost. G&A is usually charged as a fixed percent of a di- with top managers’ estimates of the resources needed for a
rect cost such as labor. project. Its primary advantage is that the aggregate budget
Indirect (or Fixed) Cost These costs are associ- is typically quite accurate because no element has been left
ated with output, but do not vary with each unit of out- out. Individual elements, however, may be quite inaccurate.
put; e.g., the cost of capital equipment not charged per Variances The pattern of deviations in costs and usage
piece of output, advertising, distribution, or sales. Costs used for exception reporting to management.
QUESTIONS
Material Review Questions

1. What are the advantages of top-down budgeting? Of 5. What is a variance?
bottom-up budgeting? What is the most important task 6. Describe the learning curve phenomenon.
for top management to do in bottom-up budgeting?
7. How might you determine if cost estimates are biased?
2. In preparing a budget, what indirect costs should be
8. What is “program budgeting”?
considered?
9. What is the difference between activity- and category-
3. Describe the purpose and use of a tracking signal.
oriented budgets?
4. Describe the top-down budgeting process.
81721_Ch07.indd 318 9/16/08 1:35:18 AM

PROBLEMS 319
Class Discussion Questions

10. Discuss ways in which to keep budget planning from 16. How is the budget planning process like a game?
becoming a game. 17. Would any of the conflict resolution methods described
11. List some of the pitfalls in cost estimating. What steps in the previous chapter be useful in the budget plan-
can a manager take to correct cost overruns? ning process? Which?
12. Why do consulting firms frequently subsidize some 18. How does the fact that capital costs vary with different
projects? Is this ethical? factors complicate the budgeting process?
13. What steps can be taken to make controlling costs 19. Why is learning curve analysis important to project
easier? Can these steps also be used to control other management?
project parameters, such as performance? 20. Why is it “ethically necessary to be honest” in negotia-
14. Which budgeting method is likely to be used with tions between a superior and subordinate?
which type of organizational structure? 21. The chapter describes the problems of budgeting for
15. What are some potential problems with the top-down S-shaped and exponential life-cycle projects. What
and bottom-up budgeting processes? What are some might be the budgeting characteristics of a project with
ways of dealing with these potential problems? a straight line life cycle?
Questions for Project Management in Practice
Pathfinder Mission to Mars—On a Shoestring Managing Costs at Massachusetts’ Neighborhood Health Plan
22. How did a change in philosophy make such a drastic 28. Wouldn’t higher eligibility requirements for subscrib-
difference in project cost? ers cut NHP’s health care costs? Why did this exacer-
23. Why was the mission scope so limited? Why even spend bate NHP’s situation?
the money to go to Mars with such limited objectives? 29. Explain the trade-off between hospital utilization and
24. Describe their “de-scope,” “lien list,” and “cash re- contract rates.
serve” approaches. 30. How did changing from a line item pay plan to an epi-
25. Recent design-to-cost interplanetary projects have also sode plan allow comparisons and save costs?
had some spectacular failures. Is this the natural result
of this new philosophy?
Completing the Limerick Nuclear Facility Under Budget
26. What “trick” did the construction firm use to come in
ahead of schedule and under budget?
27. What extra expenses did the contractor incur in order
to finish ahead of schedule? How can one tell whether
they are spending too much to finish early, thereby sav-
ing not only time but also overhead costs?
PROBLEMS
1. Using the cost estimation template and Actuals in Fig- rate of return of 0.2. The project will cost $75,000 but
ure 7-5, compare the model in the figure with the fol- will result in cash inflows of $20,000, $25,000, $30,000,
lowing estimates derived from a multiplicative model. and $50,000 in each of the next four years.
Base your comparison on the mean bias, the MAD, and 4. In Problem 3, assume that the inflows are uncertain but
the tracking signal. Comment. normally distributed with standard deviations of $1000,
Period: 1 2 3 4 5 6 7 8 $1500, $2000, and $3500, respectively. Find the mean
Estimated: 179 217 91 51 76 438 64 170 forecast NPV using Crystal Ball®. What is the prob-
2. Repeat Problem 1 but using the template in Figure 7-6 ability the actual NPV will be positive?
based on the MAR instead. 5. A production lot of 25 units required 103.6 hours of
3. Conduct a discounted cash flow calculation to determine effort. Accounting records show that the first unit took
the NPV of the following project, assuming a required 7 hours. What was the learning rate?
81721_Ch07.indd 319 9/16/08 1:35:19 AM

6. If unit 1 requires 200 hours to produce and the labor lines. The most likely estimated cost of the project it-
records for an Air Force contract of 50 units indicates self is $1,000,000, but the most optimistic estimate is
an average labor content of 63.1 hours per unit, what $900,000 while the pessimists predict a project cost
was the learning rate? What total additional number of $1,200,000. The real problem is that even if the
of labor-hours would be required for a follow-on Air project costs are within those limits, if the project it-
Force contract of 50 units? What would be the average self plus its implementation costs exceed $1,425,000,
labor content of this second contract? Of both contracts the project will not meet the firm’s NPV hurdle. There
combined? If labor costs the vendor $10/hour on this are four cost categories involved in adding the pro-
second contract and the price to the Air Force is fixed spective new machine to the production line: (1) engi-
at $550 each, what can you say about the profitability neering labor cost, (2) nonengineering labor cost, (3)
of the first and second contracts, and hence the bidding assorted material cost, and (4) production line down-
process in general? time cost.
7. Your firm designs PowerPoint slides for computer The engineering labor requirement has been
training classes, and you have just received a request estimated to be 600 hours, plus or minus 15 percent
to bid on a contract to produce the slides for an eight- at a cost of $80 per hour. The nonengineering labor
session class. From previous experience, you know that requirement is estimated to be 1500 hours, but could
your firm follows an 85 percent learning rate. For this be as low as 1200 hours or as high as 2200 hours at a
contract it appears the effort will be substantial, run- cost of $35 per hour. Assorted material may run as high
ning 50 hours for the first session. Your firm bills at the as $155,000 or as low as $100,000, but is most likely
rate of $100/hour and the overhead is expected to run to be about $135,000. The best guess of time lost on
a fixed $600 per session. The customer will pay you a the production line is 110 hours, possibly as low as 105
flat fixed rate per session. If your nominal profit margin hours and as high as 120 hours. The line contributes
is 20 percent, what will be the total bid price, the per about $500 per hour to the firm’s profit and overhead.
session price, and at what session will you break even? What is the probability that the new machine project
will meet the firm’s NPV hurdle?
8. A light manufacturing firm has set up a project for
developing a new machine for one of its production
INCIDENTS FOR DISCUSSION
Preferred Sensor Company manufacturing process. Sean is an experienced executive

Sean Cole has been appointed project manager of the Pre- and feels comfortable with his ability to come close to
ferred Sensor Company’s new sensor manufacturing proc- estimating the cost of the project. However, the recent fir-
ess project. Sensors are extremely price-sensitive, and ing of his colleague has made him a bit gun-shy. Only one
Preferred has done a great deal of quantitative work so it stage out of the traditional four-stage sensor manufacturing
can accurately forecast changes in sales volume relative to process is being changed, so he has detailed cost informa-
changes in pricing. tion about a good percentage of the process. Unfortunately,
The company president, “Dude” Sensor, has consider- the tasks involved in the process stage being modified are
able faith in the firm’s sensitivity model and insists that unclear at this point. Sean also believes that the new modi-
all projects that affect the manufacturing cost of sensors fication will cause some minor changes in the other three
be run against the sensitivity model in order to generate stages, but these changes have not been clearly identified.
data to calculate the return on investment. The net result The stage being addressed by the project represents almost
is that project managers, like Sean, are under a great deal 50 percent of the manufacturing cost.
of pressure to submit realistic budgets so go/no-go project Questions: Under these circumstances, would Sean
decisions can be made quickly. Dude has canceled several be wise to pursue a top-down or a bottom-up budgeting
projects that appeared marginal during their feasibility approach? Why? What factors are most relevant here?
stages and recently fired a project manager for overestimat-
General Ship Company
ing project costs on a new model sensor. The project was
killed very early in the design stage and six months later General Ship Company has been building nuclear destroy-
a competitor introduced a similar sensor that proved to be ers for the Navy for the last 20 years. It has recently com-
highly successful. pleted the design of a new class of nuclear destroyer and
Sean’s dilemma is how to go about constructing a will be preparing a detailed budget to be followed during
budget that accurately reflects the cost of the proposed new construction of the first destroyer.
81721_Ch07.indd 320 9/16/08 1:35:19 AM

BIBLIOGRAPHY 321
The total budget for this first destroyer is $90 million. the project and wants to work closely with the project man-
The controller feels the initial project cost estimate prepa- ager to control the costs.
red by the planning department was too low because the
waste and spoilage allowance was underestimated. Thus, Question: How would you monitor the costs of this
she is concerned that there may be a large cost overrun on project?
CONTINUING INTEGRATIVE CLASS PROJECT
Although probably a minor task for this project, create a pro- Reconcile any differences by discussion between the PM,
gram budget using both top-down and bottom-up budgeting. subgroup leaders, and the student(s) responsible for the task.
BIBLIOGRAPHY
Amor, J. P., and C. J. Teplitz. “An Efficient Approxima- Consulting, Ph.D. dissertation. Cincinnati: University of
tion for Project Composite Learning Curves.” Project Man- Cincinnati, 1982.
agement Journal, September 1998. Gagnon, R. J., and S. J. Mantel, Jr. “Strategies and Per-
Badiru, A. B. “Incorporating Learning Curve Effects into formance Improvement for Computer-Assisted Design.”
Critical Resource Diagramming.” Project Management IEEE Transactions on Engineering Management, November
Journal, June 1995. 1987.
Bierce, A. The Devil’s Dictionary. New York: Dell Pub- Harrison, F. L. Advanced Project Management. Hants,
lishing, 1991. England: Gower, 1983.
Brimson, J. A. “Activity Product Cost.” IEEE Engineering Lawrence, A. O. “Using Automated Estimating Tools to
Management Review, Spring 1993. Improve Project Estimating.” PM Network, December 1994.
Brooks, F. P. The Mythical Man-Month. Reading, MA: Martin, J. E. “Selecting an Automated Estimating Tool for
Addison-Wesley, 1975. IS Development Projects.” PM Network, December 1994.
Camm, J. D., J. R. Evans, and N. K. Womer. “The Unit Shafer, S. M., and J. R. Meredith. Operations Manage-
Learning Curve Approximation of Total Cost.” Computers ment: A Process Approach with Spreadsheets. New York:
in Industrial Engineering. Vol. 12, No. 3, 1987. Wiley, 1998.
Chase, R. B., and N. J. Aquilano. Production and Opera- Sigurdsen, A. “Principal Errors in Capital Cost Estimating
tions Management, 5th ed. Homewood, IL: Irwin, 1989. Work, Part 1: Appreciate the Relevance of the Quantity-
Christensen, D. S., and J. A. Gordon, “Does a Rubber Dependent Estimating Norms.” Project Management Jour-
Baseline Guarantee Cost Overruns on Defense Acquisi- nal, September 1996a.
tion Contracts?” Project Management Journal, September Sigurdsen, A. “Principal Errors in Capital Cost Estimat-
1998. ing, Part 2: Appreciate the Relevance of the Objective Cost
Coburn, S. “How Activity Based Costing Was Used in Risk Analysis Method.” Project Management Journal,
Capital Budgeting.” Management Accounting, May 1997. December 1996b.
Dean, B. V., S. J. Mantel, Jr., and L. A. Roepcke. “Re- Vandament, W. F., and D. P. Jones. Financial Manage-
search Project Cost Distributions and Budget Forecasting.” ment: Progress and Challenges. San Francisco: Jossey-
IEEE Transactions on Engineering Management, November Bass, 1993.
1969. Zwikael, O., S. Globerson, and T. Raz. “Evaluation of
Frame, J. D. “Risk Assessment Groups: Key Component Models for Forecasting the Final Cost of a Project.” Project
of Project Offices.” PM Network, March 1998. Management Journal, March 2000.
Gagnon, R. J. An Exploratory Analysis of the Relevant
Cost Structure of Internal and External Engineering
81721_Ch07.indd 321 9/16/08 1:35:19 AM

The following case and the answers to the questions at the end describe the stringent criteria this disguised but well-known
firm uses to select among projects that offer major profit opportunities for the firm. In addition, the firm intentionally ties
the criteria to their strategic goals so that each adopted project moves the organization farther in the competitive direction
they have chosen by adding to their core competencies. The case also illustrates how the firm integrates their marketing,
operations, engineering, and finance functions to forge a competitive advantage for the firm in the marketplace.
C A S E
AUTOMOTIVE BUILDERS, INC.:
THE STANHOPE PROJECT
Jack Meredith
It was a cold, gray October day as Jim Wickes pulled A Major Capital Project
his car into ABI’s corporate offices parking lot in subur-
The farm equipment industry in the 1970s had been
ban Detroit. The leaves, in yellows and browns, swirled
doing well, but there were some disturbing trends. Japa-
around his feet as he walked into the wind toward the
nese manufacturers had entered the industry and were
lobby. “Good morning, Mr. Wickes,” said his secre-
beginning to take a significant share of the domestic
tary as he came into the office. “That proposal on the
market. More significantly, domestic labor costs were
Stanhope project just arrived a minute ago. It’s on your
significantly higher than overseas and resulted in price
desk.” “Good morning, Debbie. Thanks. I’ve been anxi-
disadvantages that couldn’t be ignored any longer. Per-
ous to see it.”
haps most important of all, quality differences between
This was the day Jim had scheduled to review the
American and Japanese farm equipment, including trac-
1986 supplemental capital request and he didn’t want any
tors, were becoming quite noticeable.
interruptions as he scrutinized the details of the flexible
To improve the quality and costs of their incoming
manufacturing project planned for Stanhope, Iowa. The
materials, many of the domestic tractor manufacturers
Stanhope proposal, compiled by Ann Williamson, PM
were beginning to single-source a number of their trac-
and managerial “champion” of this effort, looked like
tor components. This allowed them better control over
just the type of project to fit ABI’s new strategic plan,
both quality and cost, and made it easier to coordinate
but there was a large element of risk in the project. Be-
delivery schedules at the same time.
fore recommending the project to Steve White, executive
In this vein, one of the major tractor engine manufac-
vice president of ABI, Jim wanted to review all the de-
turers, code-named “Big Red” within ABI, let its suppli-
tails one more time.
ers know that it was interested in negotiating a contract
for a possible 100 percent sourcing of 17 versions of
History of ABI
special piston heads destined for a new line of high-
ABI started operations as the Farm Equipment Company efficiency tractor engines expected to replace the current
just after the First World War. Employing new technol- conventional engines in both new and existing tractors.
ogy to produce diesel engine parts for tractors, the firm These were all six-cylinder diesel engines and thus would
flourished with the growth of farming and became a require six pistons each.
multimillion dollar company by 1940. This put ABI in an interesting situation. If they failed
During the Second World War, the firm switched to to bid on this contract, they would be inviting compe-
producing tank and truck parts in volume for the mili- tition into their very successful and profitable diesel
tary. At the war’s end, the firm converted its equipment engine parts business. Thus, to protect their existing suc-
to the production of automotive parts for the expand- cessful business, and to pursue more such business, ABI
ing automobile industry. To reflect this major change in seemed required to bid on this contract. Should ABI be
their product line, the company was renamed Automo- successful in their bid, this would result in 100 percent
tive Builders, Inc. (ABI), though they remained a major sourcing in both the original equipment market (OEM)
supplier to the farm equipment market. as well as the replacement market with its high margins.
81721_Ch07.indd 322 9/16/08 1:35:20 AM

CASE 323
Furthermore, the high investment required to produce 2. Pursue oly new products whose design or production
these special pistons at ABI’s costs would virtually rule process is of a proprietary nature and that exist in
out future competition. areas where our technical abilities enable us to main-
ABI had two plants producing diesel engine compo- tain a long-term position.
nents for other manufacturers and believed they had a 3. Employ, if at all possible, the most advanced tech-
competitive edge in engineering of this type. These plants, nology in new projects that is either within our expe-
however, could not accommodate the volume Big Red ex- rience or requires the next step up in experience.
pected for the new engine. Big Red insisted at their nego-
4. Foster the “project champion” approach to innova-
tiations that a 100 percent supplier be able to meet peak
tion and creativity. The idea is to encourage entrepre-
capacity at their assembly plant for this new line.
neurship by approving projects to which individual
As Jim reviewed the proposal, he decided to refer
managers are committed and that they have adopted
back to the memos that restated their business strategy
as personal “causes” based on their belief that the
and started them thinking about a new Iowa plant lo-
idea, product, or process is in our best interest.
cated in the heart of the farm equipment industry for this
project. In addition, Steve White had asked the follow- 5. Maintain small plants of no more than 480 employ-
ing basic, yet rather difficult questions about the pro- ees. These have been found to be the most efficient,
posal at their last meeting and Jim wanted to be sure he and they enjoy the best labor relations.
had them clearly in mind as he reviewed the files. With these in mind, Jim reopened the proposal and star-
• ABI is already achieving an excellent return on ted reading critical sections.
investment (ROI). Won’t this investment simply
tend to dilute it? Demand Forecasts and Scenarios
• Will the cost in new equipment be returned by an For this proposal, three scenarios were analyzed in terms
equivalent reduction in labor? Where’s the payoff? of future demand and financial impacts. The baseline
• What asset protection is there? This proposal “Scenario I” assumed that the new line would be suc-
requires an investment in new facilities before cessful. “Scenario II” assumed that the Japanese would
knowing whether a long-term contract will be soon follow and compete successfully with Big Red in
procured to reimburse us for our investment. this line. “Scenario III” assumed that the new line was
a failure. The sales volume forecasts under these three
• Does this proposal maximize ROI, sales potential,
scenarios are shown in Table 1.
or total profit?
There was, however, little confidence in any of these
To address these questions adequately, Jim decided to forecasts. In the preceding few years Japan had become
recheck the expected after-tax profits and average rate of a formidable competitor, not only in price but also in
return (based on sales of 70,000 engines per year) when more difficult areas of competition, such as quality and
he reached the financial portion of the proposals. These reliability. Furthermore, the economic situation in 1986
figures should give a clear indication of the “quality” was taking a severe toll on American farmers and eco-
of the investment. There were, however, other aspects nomic forecasts indicated there was no relief in sight.
of capital resource allocation to consider besides the Thus, as stated in the proposal:
financial elements. One of these was the new business
strategy of the firm, as recently articulated by ABI’s The U.S. farm market will be a difficult battle-
executive committee. ground for world farm equipment manufacturers
and any forecast of a particular engine’s potential
in this market must be considered as particularly
The Business Strategy risky. How much risk do we want to accept? Every
A number of elements of ABI’s business strategy were effort should be made to minimize our exposure
directly relevant to this proposal. Jim took out a note on this investment and maximize our flexibility.
pad to jot down each of them and assign them a priority
as follows: Manufacturing Plan
1. Bid only on good margin products that have the po- The proposal stressed two primary aspects of the manu-
tential for maintaining their margins over a long term. facturing process. First, a learning curve was employed
81721_Ch07.indd 323 9/16/08 1:35:20 AM

Table 1. Demand Forecasts (000s engines)* of machining other metals, such as aluminum or nodular
Year Baseline I Scenario II Scenario III iron, and would be fitted with variable feed and speed
motors, feed-force monitors, pressure-controlled clamp-
1987 69 69 69 ing of workpieces, and air-leveling pallets. Also, fully
1988 73 72 72 interchangeable chucks, spindles, pallets, tooling, and
1989 90 81 77 risers would be purchased to minimize the spare parts
1990 113 95 68
inventories.
1991 125 87 62
1992 145 74 47
Plant Operation and Organization
*Each engine requires six pistons.
As stated in the proposal, many innovative practices
were to be employed at the new plant:
in calculating production during the 1000-unit ramp-up • Machine operators will be trained to do almost all
implementation period in order to not be overly opti- of their own machine maintenance.
mistic. A learning rate of 80 percent was assumed.
Second, an advanced technology process using a flexi-
• All employees will conduct their own statistical
process control and piston heads will be subject
ble manufacturing system, based largely on turning cen- to 100 percent inspection.
ters, was recommended since it came in at $1 million
less than conventional equipment and met the strategy • There will only be four skill classes in the plant.
guidelines of using sophisticated technology when Every employee in each of those classes will be
appropriate. trained to do any work within that class.
Since ABI had closely monitored Big Red’s progress • There will not be any time clocks in the plant.
in the engine market, the request for bids had been fore-
The organizational structure for the 11 salaried work-
seen. In preparation for this, Jim had authorized a spe-
ers in the new plant is shown in Figure 1, and the
cial manufacturing-process study to determine more
complete labor summary is illustrated in Figure 2,
efficient and effective ways of producing piston heads.
including the shift breakdown. As can be seen, the
The study considered product design, process selection,
plant will be relatively small, with 65 employees in
quality considerations, productivity, and manufacturing
the ratio of 1:5 salaried to hourly. The eight-month
system planning. Three piston manufacturing methods
acquisition of the employees during the ramp-up is
were considered in the study: (1) batch manufacture
illustrated in Figure 3, with full employment occur-
via computer numerically controlled (CNC) equipment;
ring by March 1987.
(2) a flexible manufacturing system (FMS); and (3) a
high-volume, low-unit-cost transfer machine.
Financial Considerations
The resulting recommendation was to install a
carefully designed FMS, if it appeared that additional Financial aspects of new proposals at ABI were consid-
flexibility might be required in the future for other ered from a number of perspectives, in part because of
versions, or even other manufacturers. Though such the interdependent nature of many proposals. The re-
a system would be expensive, the volume of produc- sults of not investing in a proposal are normally com-
tion over the FMS’s longer lifetime would offset that pared with the results of investing and the differences
expense. Four preferred machine builders were con- noted. Variations on the investment assumptions are also
tacted for equipment specifications and bids. It was tested, including errors in the forecast sales volumes,
ABI’s plan to work closely with the selected vendor learning rates, productivities, selling prices, and cancel-
in designing and installing the equipment, thus build- lations of both current and future orders for existing and
ing quality and reliability into both the product and potential business.
the process and learning about the equipment at the For the Stanhope proposal, the site investment re-
same time. quired is $3,012,000. The details of this investment
To add further flexibility for the expensive machi- are shown in Table 2. The total investment required
nery, all design features that would facilitate retool or amounts to $7,108,000 (plus required working capital
changeover to other products were incorporated. For of $1,380,000). The equipment is depreciated over an
example, the machining centers would also be capable eight-year life. ABI, under the revised tax laws, is in the
81721_Ch07.indd 324 9/16/08 1:35:20 AM

CASE 325
Plant
manager
Engineering/ Manufacturing
Personnel
Quality manager
Engineer Shift 2 Clerk

supervisor
Quality Shift 3
Clerk
control supervisor
Clerk
Figure 1 Stanhope organization.
34 percent tax bracket. The price of the piston heads has Salaried Labor Number of Staff
been tentatively set at $25.45 apiece. ABI’s expected Plant manager 1
costs are shown in Table 3. Manufacturing managers (3 shifts) 3
Quality control manager 1
Engineering 2
Some Concerns Personnel manager 1
Jim had spoken with some of his colleagues about the Clerical 23
FMS concept after the preliminary financial results had 11
been tabulated. Their concerns were what now inter-
ested him. For example, he remembered one manager Hourly Labor Days Afternoons Night
asking: “Suppose Big Red’s sales only reach 70 percent Direct 14 14 10
of our projections in the 1989–90 time period, or say, Inspection 1 1 1
perhaps as much as 150 percent; how would this affect Maintenance 2 1 1
the project? Does the FMS still apply or would you Tooling 2 2 1
Rec./shp./mtl. 22 21 21
consider some other form of manufacturing equipment,
Total 21 19 14
possibly conventional or CNC with potential aftermar-
ket application in the former case or a transfer machine Summary
in the latter case?”
Salary 11
Another manager wrote down his thoughts as a memo
Hourly 54
to forward to Jim. He had two major concerns: Total 65
• “Scenario II” analysis assumes the loss of sub-
Figure 2 Stanhope labor summary.
stantial volume to competition. This seems rather
unlikely. Jim wondered what these changes in their assumptions
• After-tax margins seem unreasonably high. Can would do to the ROI of the proposal and its overall
we get such margins on a sole-source contract? profitability.
81721_Ch07.indd 325 9/16/08 1:35:21 AM

80
Total
60
Hourly
40
20
Salaried
0
Aug Sept Oct Nov Dec Jan Feb March
1986 1987 Figure 3 Stanhope labor buildup.
Table 2. Stanhope Site Capital Costs Table 3. Piston Head Cost Summary
Land and Site Preparation Material $8.47
land $246,000 Labor 1.06
access roads/parking lot 124,000 Variable overhead 2.23
landscaping 22,000 Fixed overhead 2.44
Freight 0.31
Building Costs Total Factory Cost 14.51
building (67,000 sq ft) 1,560,000 General & administrative 1.43
air conditioning 226,000 Scrap 0.82
power 205,000 Testing 0.39
employee services 177,000 Total Cost 17.15
legal fees and permits 26,000
Auxiliary Equipment
ABI company sign 25,000
containers, racks, etc. 33,000
flume 148,000
coolant disposal 97,000 do. If the demand didn’t materialize, ABI might be
furnishings 51,000 sorry they had invested in such an expensive piece
forklift trucks 72,000 of equipment as an FMS.
Total 3,012,000
Strategically, it seemed like ABI had to make this
investment to protect its profitable position in the die-
sel engine business; but how far should this argument
be carried? Were they letting their past investments
Conclusion
color their judgment on new ones? He was also con-
Jim had concerns about the project also. He won- cerned about the memo questioning the high profit
dered how realistic the demand forecasts were, given margins. They did seem high in the midst of a slug-
the weak economy and what the Japanese might gish economy.
81721_Ch07.indd 326 9/16/08 1:35:21 AM

DIRECTED READING 327
QUESTIONS
1. How did ABI handle forecast risk? 4. What other factors are relevant to this issue?
2. Were ABI’s Stanhope site costs in Table 2 derived by a 5. How do the changes in assumptions mentioned by the
top-down or bottom-up process? Why? other managers affect the proposal?
3. What are the answers to Steve White’s questions? 6. What position should Jim take? Why?
This article clearly describes the importance and impact of cost-related issues on a project. These issues can
significantly alter the profitability and even success of a project. Costs are discussed from three viewpoints:
that of the project manager, the accountant, and the controller. Not only are the amounts of expenditures and
encumbrances important, but their timing is critical also. Perhaps most important is having a project cost sys-
tem that accurately reports costs and variances in a way that can be useful for managerial decisions.
D I R E C T E D R E A D I N G
THREE PERCEPTIONS OF PROJECT COST*
D. H. Hamburger
Project cost seems to be a relatively simple expression, but are reported. With this knowledge, the project manager
“cost” is more than a four letter word. Different elements of can control more than the project’s cost of goods sold (a
the organization perceive cost differently, as the timing of function often viewed as the project manager’s sole finan-
project cost identification affects their particular organiza- cial responsibility). The project manager can also influence
tional function. The project manager charged with on-time, the timing of cost to improve cash flow and the cost of
on-cost, on-spec execution of a project views the “on cost” financing the work, in addition to affecting revenue and
component of his responsibility as a requirement to stay expense reporting in the P&L statement.
within the allocated budget, while satisfying a given set
of specified conditions (scope of work), within a required Three Perceptions of Cost
time frame (schedule). To most project managers this sim-
ply means a commitment to project funds in accordance To understand the three perceptions of cost—commitments,
with a prescribed plan (time-based budget). Others in the expenses, and cash flow—consider the purchase of a major
organization are less concerned with the commitment of project component. Assume that a $120,000 compressor
funds. The accounting department addresses expense rec- with delivery quoted at six months was purchased. Figure 1
ognition related to a project or an organizational profit and depicts the order execution cycle. At time 0 an order is pla-
loss statement. The accountant’s ultimate goal is reporting ced. Six months later the vendor makes two shipments, a
profitability, while positively influencing the firm’s tax li- large box containing the compressor and a small envelope
ability. The comptroller (finance department) is primarily
Ship compressor
concerned with the organization’s cash flow. It is that per- invoice
son’s responsibility to provide the funds for paying the Place P.O. Pay invoice
bills, and putting the unused or available money to work
for the company. Cash outflow
$120,000
To be an effective project manager, one must under-
Commitment Expense
stand each cost, and also realize that the timing of cost $120,000 $120,000
identification can affect both project and corporate finan-
cial performance. The project manager must be aware of
the different cost perceptions and the manner in which they
*Three Perceptions of Project Cost—Cost Is More Than a Four 0 1 2 3 4 5 6 7 8
Letter Word. Project Management Journal, June 1986. ©1987 by the Time (months)
Project Management Institute. Reprinted by permission. Figure 1 Three perceptions of project cost.
81721_Ch07.indd 327 9/16/08 1:35:21 AM

containing an invoice. The received invoice is processed assigning an appropriate level of expense to arrive at an ac-
immediately, but payment is usually delayed to comply ceptable period gross margin. At the end of each account-
with corporate payment policy (30, 60, 90, or more days ing year and at the end of the project, adjustments are made
may pass before a check is actually mailed to the vendor). to the recorded expenses to account for the differences be-
In this example, payment was made 60 days after receipt of tween actual expenses incurred and the theoretical expenses
the invoice or 8 months after the order for the compressor recorded in the P&L statement. This can be a complex pro-
was given to the vendor. cedure. The misinformed or uninformed project manager
can place the firm in an untenable position by erroneously
Commitments—The Project Manager’s misrepresenting the project’s P&L status; and the rare un-
Concern scrupulous project manager can use an arbitrary assessment
of the project’s percentage of completion to manipulate the
Placement of the purchase order represents a commitment to firm’s P&L statement.
pay the vendor $120,000 following satisfactory delivery of There are several ways by which the project’s percent-
the compressor. As far as the project manager is concerned, age of completion can be assessed to avoid these risks. A
once this commitment is made to the vendor, the available typical method, which removes subjective judgments and
funds in the project budget are reduced by that amount. the potential for manipulation by relying on strict account-
When planning and reporting project costs the project ing procedures, is to be described. In this process a theo-
manager deals with commitments. Unfortunately, many ac- retical period expense is determined, which is divided by
counting systems are not structured to support project cost the total estimated project expense budget to compute the
reporting needs and do not identify commitments. In fact, percentage of total budget expense for the period. This be-
the value of a purchase order may not be recorded until an comes the project’s percentage of completion which is then
invoice is received. This plays havoc with the project man- used to determine the revenue to be “taken down” for the
ager’s fiscal control process, as he cannot get a “handle” on period. In this process, long delivery purchased items are
the exact budget status at a particular time. In the absence not expensed on receipt of an invoice, but have the value of
of a suitable information system, a conscientious project their purchase order prorated over the term of order execu-
manager will maintain personal (manual or computer) tion. Figure 2 shows the $120,000 compressor in the exam-
records to track his project’s commitments. ple being expensed over the six-month delivery period at
the rate of $20,000 per month.
Expenses—The Accountant’s Concern
Preparation of the project’s financial report requires identi- Cash Flow—The Comptroller’s Concern
fication of the project’s revenues (when applicable) and all The comptroller and the finance department are respon-
project expenses. In most conventional accounting systems, sible for managing the organization’s funds, and also as-
expenses for financial reporting purposes are recognized suring the availability of the appropriate amount of cash
upon receipt of an invoice for a purchased item (not when for payment of the project’s bills. Unused funds are put to
the payment is made—a common misconception). Thus, work for the organization in interest-bearing accounts or in
the compressor would be treated as an expense in the sixth other ventures. The finance department’s primary concern
month. is in knowing when funds will be needed for invoice pay-
In a conventional accounting system, revenue is re- ment in order to minimize the time that these funds are not
corded when the project is completed. This can create se- being used productively. Therefore, the comptroller really
rious problems in a long-term project in which expenses
are accrued during each reporting period with no attendant
revenue, and the revenue is reported in the final period with Place P.O. Pay invoice
little or no associated expenses shown. The project runs at Ship compressor
invoice
an apparent loss in each of the early periods and records an
inordinately large profit at the time revenue is ultimately Commitment Cash outflow
$120,000 $120,000
reported—the final reporting period. This can be seriously Expense Expense Expense
$20,000 $20,000 $20,000
misleading in a long-term project which runs over a multi-
year period. Expense Expense Expense
To avoid such confusion, most long-term project P&L $20,000 $20,000 $20,000
statements report revenue and expenses based on a “per-
centage of completion” formulation. The general intent is
to “take down” an equitable percentage of the total project 0 1 2 3 4 5 6 7 8
revenue (approximately equal to the proportion of the Time (months)
project work completed) during each accounting period, Figure 2 Percentage of completion expensing.
81721_Ch07.indd 328 9/16/08 1:35:22 AM

views project cost as a cash outflow. Placement of a pur- the delivery period for major equipment items; subcontract
chase order merely identifies a future cash outflow to the performance period and the schedule of its work; and the
comptroller, requiring no action on his part. Receipt of the effect of the project schedule on when and how labor will
invoice generates a little more interest, as the comptroller be expended in relation to equipment procurement.
now knows that a finite amount of cash will be required for The conscientious project manager must understand
a particular payment at the end of a fixed period. Once a these different perceptions of cost and should be prepared
payment becomes due, the comptroller provides the funds, to plan and report on any and all approaches required by
payment is made, and the actual cash outflow is recorded. management. The project manager should also be aware of
It should be noted that the compressor example is a the manner in which the accounting department collects and
simplistic representation of an actual procurement cycle, reports “costs.” Since the project manager’s primary con-
as vendor progress payments for portions of the work (i.e., cern is in the commitments, he or she should insist on an
engineering, material, and delivery) may be included in the accounting system which is compatible with the project’s
purchase order. In this case, commitment timing will not reporting needs. Why must a project manager resort to a
change, but the timing of the expenses and cash outflow manual control system when the appropriate data can be
will be consistent with the agreed-upon terms of payment. made available through an adjustment in the accounting
The example describes the procurement aspect of project department’s data processing system?
cost, but other project cost types are treated similarly. In
the case of project labor, little time elapses between actual Putting Your Understanding of Cost to Work
work execution (a commitment), the recording of the labor
hours on a time sheet (an expense), and the payment of Most project managers believe that their total contribution
wages (cash outflow). Therefore, the three perceptions of to the firm’s profitability is restricted by the ability to limit
cost are treated as if they each occur simultaneously. Sub- and control project cost, but they can do much more. Once
contracts are treated in a manner similar to equipment pur- the different perceptions of cost have been recognized, the
chases. A commitment is recorded when the subcontract is project manager’s effectiveness is greatly enhanced. The
placed and cash outflow occurs when the monthly invoice manner in which the project manager plans and executes
for the work is paid. Expenses are treated in a slightly dif- the project can improve company profitability through influ-
ferent manner. Instead of prorating the subcontract sum ence on financing expenses, cash flow, and the reporting of
over the performance period, the individual invoices for the revenue and expenses. To be a completely effective project
actual work performed are used to determine the expense manager one must be totally versed in the cost accounting
for the period covered by each invoice. practices which affect the firm’s project cost reporting.
Thus the three different perceptions of cost can result in Examination of the typical project profit & loss state-
three different time-based cost curves for a given project ment (see Table 1) shows how a project sold for profit is
budget. Figure 3 shows a typical relationship between subjected to costs other than the project’s costs (cost of
commitments, expenses, and cash outflow. The commit- goods sold). The project manager also influences other
ment curve leads and the cash outflow curve lags, with the areas of cost as well, addressing all aspects of the P&L to
expense curve falling in the middle. The actual shape and influence project profitability positively.
the degree of lag/lead between the curves are a function Specific areas of cost with examples of what a project
of several factors, including: the project’s labor, material, manager can do to influence cost of goods sold, interest
and subcontract mix; the firm’s invoice payment policy; expense, tax expense, and profit are given next.
800
Table 1. Typical Project Profit & Loss Statement
700
Commitments Revenue (project sell price) ($1,000,000)
600
Cost x $1,000
(less) cost of goods sold

500 (project costs) ($ 750,000)
400 Gross margin $ 250,000)
Expenses
300 (less) selling, general &
administrative expenses ($ 180,000)
200 Cash outflow Profit before interest and taxes $ 70,000)
100
(less) financial expense ($ 30,000)
0 Profit before taxes $ 40,000)
1 2 3 4 5 6 7 8 9 10 11 12
(less) taxes ($ 20,000)
Months
Net profit $ 20,000
Figure 3 Three perceptions of cost.
81721_Ch07.indd 329 9/16/08 1:35:22 AM

Cost of Goods Sold (Project Cost) should be built into the project schedule. The Mer-
cury project proved to be safe and successful even
• Evaluation of alternate design concepts and the use though John Glenn, perched in the Mercury capsule
of “trade-off” studies during the development phase atop the Atlas rocket prior to America’s first earth or-
of a project can result in a lower project cost, with- biting flight, expressed his now famous concern that
out sacrificing the technical quality of the project’s “all this hardware was built by the low bidder.” The
output. The application of value engineering princi- project manager should ensure that the initial project
ples during the initial design period will also reduce budget is commensurate with the project’s required
cost. A directed and controlled investment in the level of reliability. The project manager should not
evaluation of alternative design concepts can result be put in the position of having to buy project reli-
in significant savings of project cost. ability with unavailable funds.
• Excessive safety factors employed to ensure “on- • Procurement of material and services based on par-
spec” performance should be avoided. Too frequently tially completed drawings and specifications should
the functional members of the project team will be avoided. The time necessary for preparing a
apply large safety factors in their effort to meet or complete documentation package before soliciting
exceed the technical specifications. The project team bids should be considered in the preparation of the
must realize that such excesses increase the project’s project schedule. Should an order be awarded based
cost. The functional staff should be prepared to jus- on incomplete data and the vendor then asked to
tify an incremental investment which was made to alter the original scope of supply, the project will be
gain additional performance insurance. Arbitrary and controlled by the vendor. In executing a “fast track”
excessive conservatism must be avoided. project, the project manager should make certain
• Execution of the project work must be controlled. The that the budget contains an adequate contingency for
functional groups should not be allowed to stretch out the change orders which will follow release of a par-
the project for the sake of improvement, refinement, tially defined work scope.
or the investigation of the most remote potential risk. • Changes should not be incorporated in the project
When a functional task has been completed to the scope without client and/or management approval
project manager’s satisfaction (meeting the task’s and the allocation of the requisite funds. Making
objectives), cut off further spending to prevent accu- changes without approval will erode the existing
mulation of “miscellaneous” charges. budget and reduce project profitability; meeting the
• The project manager is usually responsible for con- project manager’s “on-cost” commitment will be-
trolling the project’s contingency budget. This budget come extremely difficult, if not impossible.
represents money that one expects to expend during • During periods of inflation, the project manager
the term of the project for specific requirements not must effectively deal with the influence of the
identified at the project onset. Therefore, these funds economy on the project budget. This is best accom-
must be carefully monitored to prevent indiscrimi- plished during the planning or estimating stage of
nate spending. A functional group’s need for a por- the work, and entails recognition of planning in an
tion of the contingency budget must be justified and inflationary environment for its effect by estimat-
disbursement of these funds should only be made ing the potential cost of two distinct factors. First, a
after the functional group has exhibited an effort to “price protection” contingency budget is needed to
avoid or limit its use. It is imperative that the con- cover the cost increases that will occur between the
tingency budget be held for its intended purpose. time a vendor provides a firm quotation for a limited
Unexpected problems will ultimately arise, at which period and the actual date the order will be placed.
time the funds will be needed. Use of this budget to (Vendor quotations used to prepare an estimate usu-
finance a scope change is neither advantageous to ally expire long before the material is actually pur-
the project manager nor to management. The con- chased.) Second, components containing certain
tingency budget represents the project manager’s price-volatile materials (e.g., gold, silver, etc.) may
authority in dealing with corrections to the project not be quoted firm, but will be offered by the sup-
work. Management must be made aware of the true plier as “price in effect at time of delivery.” In this
cost of a change so that financing the change will be case an “escalation” contingency budget is needed
based on its true value (cost-benefit relationship). to cover the added expense that will accrue between
• In the procurement of equipment, material, and order placement and material delivery. Once the
subcontract services, the specified requirements project manager has established these inflation
should be identified and the lowest priced, qualified related contingency budgets, the PM’s role becomes
supplier found. Adequate time for price “shopping” one of ensuring controlled use.
81721_Ch07.indd 330 9/16/08 1:35:22 AM

Financial Expense inflow. A reduction in retention should be pursued

as the project nears completion. Allowing a project’s
• The project’s financial cost (interest expense) can be schedule to indiscriminately slip delays project
minimized by the project manager through the tim- acceptance, thereby delaying final payment. Incur-
ing of order placement. Schedule slack time can be ring an additional expense to resolve a questionable
used to defer the placement of a purchase order so problem should be considered whenever the expense
that the material is not available too early and the will result in rapid project acceptance and a favora-
related cash outflow is not premature. There are sev- ble interest expense reduction.
eral risks associated with this concept. Delaying an
order too long could backfire if the desired material • On internally funded projects, where retention,
is unavailable when needed. Allowing a reasonable progress payments, and other client related financial
margin for error in the delivery cycle, saving some of considerations are not a factor, management expects
the available slack time for potential delivery prob- to achieve payback in the shortest reasonable time.
lems, will reduce this risk. Waiting too long to place In this case, project spending is a continuous cash
a purchase order could result in a price increase outflow process which cannot be reversed until the
which can more than offset the interest savings. It is project is completed and its anticipated financial
possible to “lock-up” a vendor’s price without com- benefits begin to accrue from the completed work.
mitting to a required delivery date, but this has its Unnecessary project delays, schedule slippages, and
limitations. If vendor drawings are a project require- long-term investigations extend system startup and
ment, an “engineering only” order can be placed to defer the start of payback. Early completion will re-
be followed by hardware release at the appropriate sult in an early start of the investment payback pro-
time. Deferred procurement which takes advantage cess. Therefore, management’s payback goal should
of available slack time should be considered in the be considered when planning and controlling project
execution of all projects, especially during periods work, and additional expenditures in the execution of
when the cost of money is excessively high. the work should be considered if a shortened sched-
ule will adequately hasten the start of payback.
• Vendors are frequently used to help “finance the
project” by placing purchase orders which contain
Tax Expense and Profit
extended payment terms. Financially astute ven-
dors will build the cost of financing the project into • On occasion, management will demand project
their sell price, but only to the extent of remaining completion by a given date to ensure inclusion of
competitive. A vendor’s pricing structure should be the project’s revenue and profit within a particular
checked to determine if progress payments would accounting period. This demand usually results from
result in a reduced price and a net project benefit. a need to fulfill a prior financial performance fore-
A discount for prompt payment should be taken if cast. Delayed project completion by only a few days
the discount exceeds the interest savings that could could shift the project’s entire revenue and profit
result from deferring payment. from one accounting period to the next. The volatile
• Although frequently beyond the project manager’s nature of this situation, large sums of revenue and
control, properly structured progress payment terms profit shifting from one period to the next, results in
can serve to negate most or all project financial ex- erratic financial performance which negatively re-
penses. The intent is simple. A client’s progress pay- flects on management’s ability to plan and execute
ment terms can be structured to provide scheduled their efforts.
cash inflows which offset the project’s actual cash • To avoid the stigma of erratic financial performance,
outflow. In other words, maintenance of a zero net management has been known to suddenly redirect a
cash position throughout the period of project execu- carefully planned, cost-effective project team effort
tion will minimize the project’s financial expense. to a short-term, usually costly, crash exercise, di-
In fact, a positive net cash position resulting from rected toward a project completion date, artificially
favorable payment terms can actually result in a necessitated by a corporate financial reporting need.
project which creates interest income rather than one Unfortunately, a project schedule driven by influ-
that incurs an interest expense. Invoices to the client ences external to the project’s fundamental objec-
should be processed quickly, to minimize the lost in- tives usually results in additional cost and reduces
terest resulting from a delay in receiving payment. profitability.
• Similarly, the project manager can influence receipt • In this particular case, the solution is simple if a
of withheld funds (retention) and the project’s percentage of completion accounting process can
final payment to improve the project’s rate of cash be applied. Partial revenue and margin take-down
81721_Ch07.indd 331 9/16/08 1:35:23 AM

during each of the project’s accounting periods,

resulting from this procedure (rather than lump sum
• Revenue/profit accrual based on percentage of com-
pletion can create a financial problem if actual
take-down in a single period at the end of the project, expenses greatly exceed the project budget. In this
as occurs using conventional accounting methods) case the project’s percentage of completion will accu-
will mitigate the undesirable wild swings in reported mulate more quickly than justified and the project
revenue and profit. Two specific benefits will result. will approach a theoretical 100 percent completion
First, management’s revenue/profit forecast will be before all work is done. This will “front load” rev-
more accurate and less sensitive to project schedule enue/profit take-down and will ultimately require a
changes. Each project’s contribution to the over- profit reversal at project completion. Some managers
all forecast will be spread over several accounting may find this desirable, since profits are being shifted
periods and any individual performance change will into earlier periods, but most reputable firms do not
cause the shift of a significantly smaller sum from wish to overstate profits in an early period which will
one accounting period to the next. Second, a project have to be reversed at a later time. Therefore, the
nearing completion will have had 90–95 percent project manager should be aware of cost overruns
of its revenue/profit taken down in earlier periods, and, when necessary, reforecast the project’s “cost on
which will lessen or completely eliminate manage- completion” (increasing the projected cost and reduc-
ment pressure to complete the work to satisfy a ing the expected profit) to reduce the level of profit
financial reporting demand. Inordinate, unneces- taken down in the early periods to a realistic level.
sary spending to meet such unnatural demands can
thereby be avoided.
Conclusion
• An Investment Tax Credit,* a net reduction in cor-
porate taxes gained from a capital investment project Cost is not a four letter word to be viewed with disdain by
(a fixed percentage of the project’s installed cost), the project manager. It is a necessary element of the proj-
can be earned when the project actually provides its ect management process which the project manager must
intended benefit to the owner. The project manager comprehend despite the apparent mysteries of the account-
should consider this factor in scheduling the project ing systems employed to report cost. The concept of cost is
work, recognizing that it is not necessary to com- more than the expenses incurred in the execution of the proj-
plete the entire project to obtain this tax benefit as ect work: the manner in which cost is treated by the organi-
early as possible. Failure to substantiate beneficial zation’s functional elements can affect project performance,
use within a tax year can shift this savings into the interest expenses, and profitability. Therefore, the conscien-
next tax year. The project manager should consider tious project manager must develop a complete understand-
this factor in establishing the project’s objectives, ing of project cost and the accounting systems used to record
diligently working toward attainment by scheduling and report costs. The project manager should also recognize
the related tasks to meet the tax deadline. Consid- the effect of the timing of project cost, and the differences
eration should also be given to expenditures (to the between commitments, expenses, and cash flow. The project
extent they do not offset the potential tax savings) to manager should insist on the accounting system modifica-
reach this milestone by the desired date. tions needed to accommodate project cost reporting and con-
trol requirements. Once an appreciation for these concepts
• In managing the corporate P&L statement, the
has been gained, the project manager can apply this knowl-
need to shift revenue, expenses, and profit from
one tax period to the next often exists. By manag- edge towards positively influencing project and organiza-
ing the project schedule (expediting or delaying tional profitability in all areas of cost through control of the
major component procurements or shifting expen- project schedule and the execution of the project’s work.
sive activities), the project manager can support
this requirement. Each individual project affords Questions
a limited benefit, but this can be maximized if the 1. What is the major point of the article?
project manager is given adequate notice regarding
2. How does the accountant view project costs?
the necessary scheduling adjustments.
3. How does the controller view project costs?
4. How does the project manager view project costs?
*The proposed tax law revisions under consideration in Congress 5. What other costs does the project manager need to be
at the time this article was written include a provision which elimi- cognizant of? What actions should the PM take con-
nates the Investment Tax Credit. cerning these other costs?
81721_Ch07.indd 332 9/16/08 1:35:23 AM

Chapter 6 / Project Activity Planning: Acknowledgments

Uploaded by

Copyright:

Available Formats

Chapter 6 / Project Activity Planning: Acknowledgments

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 6 / Project Activity Planning: Acknowledgments

Uploaded by

Copyright:

Available Formats

292 CHAPTER 6 / PROJECT ACTIVITY PLANNING

81721_Ch06.indd 292 9/16/08 1:09:47 AM

81721_Ch07.indd 293 9/16/08 1:35:10 AM

7.1 ESTIMATING PROJECT BUDGETS

Project Management in Practice

81721_Ch07.indd 294 9/16/08 1:35:10 AM

The Pathfinder Rover explores Martian terrain.

• to achieve a successful landing • mission designers/builders transitioned into the

81721_Ch07.indd 295 9/16/08 1:35:10 AM

81721_Ch07.indd 296 9/16/08 1:35:11 AM

81721_Ch07.indd 297 9/16/08 1:35:11 AM

81721_Ch07.indd 298 9/16/08 1:35:11 AM

Work Element Costing

81721_Ch07.indd 299 9/16/08 1:35:12 AM

Project Management in Practice

An Iterative Budgeting Process—Negotiation-in-Action

81721_Ch07.indd 300 9/16/08 1:35:12 AM

81721_Ch07.indd 301 9/16/08 1:35:12 AM

Figure 7-2 Another possible project life

Comments on the Budget Request Process

81721_Ch07.indd 302 9/16/08 1:35:12 AM

Source: DILBERT: © Scott Adams/Dist. by United Feature Syndicate, Inc.

the organization may be increased or must be decreased, organizational policy on adding to

Cost Category Budgeting vs. Program/Activity Budgeting

81721_Ch07.indd 303 9/16/08 1:35:13 AM

Travel & entertainment expenses

81721_Ch07.indd 304 9/16/08 1:35:13 AM

Table 7-2. Project Budget by Task and Month

7.2 IMPROVING THE PROCESS OF COST ESTIMATION

81721_Ch07.indd 305 9/16/08 1:35:14 AM

Project Management in Practice

81721_Ch07.indd 306 9/16/08 1:35:14 AM

Person to How Many/ Check ( ) If

81721_Ch07.indd 307 9/16/08 1:35:15 AM

• To obtain the parts or service portion of the work

81721_Ch07.indd 308 9/16/08 1:35:15 AM

81721_Ch07.indd 309 9/16/08 1:35:15 AM

Additional time used for learning

A Special Case of Learning—Technological Shock

81721_Ch07.indd 310 9/16/08 1:35:16 AM

81721_Ch07.indd 311 9/16/08 1:35:16 AM

81721_Ch07.indd 312 9/16/08 1:35:16 AM

Project Management in Practice

81721_Ch07.indd 313 9/16/08 1:35:17 AM

On Making Better Estimates*

81721_Ch07.indd 314 9/16/08 1:35:17 AM

81721_Ch07.indd 315 9/16/08 1:35:17 AM

81721_Ch07.indd 316 9/16/08 1:35:18 AM

81721_Ch07.indd 317 9/16/08 1:35:18 AM

Material Review Questions

81721_Ch07.indd 318 9/16/08 1:35:18 AM

Class Discussion Questions

81721_Ch07.indd 319 9/16/08 1:35:19 AM

INCIDENTS FOR DISCUSSION

Preferred Sensor Company manufacturing process. Sean is an experienced executive

81721_Ch07.indd 320 9/16/08 1:35:19 AM

CONTINUING INTEGRATIVE CLASS PROJECT

81721_Ch07.indd 321 9/16/08 1:35:19 AM

81721_Ch07.indd 322 9/16/08 1:35:20 AM