It's easy to
make good decisions when there are no bad options, a pundit once observed.
But he might have added that it helps if choices are clear, consequences
are minor, and the decision rests with a single decision maker.
Fortunately, for the many day-to-day decisions that meet those criteria, the decision
process is simple: Make an intuitive choice and proceed with little or
no regret.
But what of high-stakes decisions that must be made in the face of competing
priorities, multiple decision makers, and limited information? Common
sense tells us that a high-quality decision making process will give
us the best shot at a good outcome.
The "Trust
Your Gut" Myth
Given their significant consequences, strategic decisions are generally
made with some care and deliberation. But after gathering all of the
available information and completing an often lengthy decision process
that may involve costly studies and multiple meetings, many decision
makers nevertheless base their final choices on something less than
scientific: gut feeling or their initial impressions.
In fact, according to researchers Dan Lovallo and Daniel Kahneman,
executives have a tendency to make decisions based on "delusional
optimism" rather than on a rational weighing of gains, losses
and probabilities. (See "Delusions of Success: How Optimism Undermines
Executives' Decisions," Harvard Business Review, July 2003.)
Other researchers confirm that instead of trying to come to terms
with complexity, decision makers have a marked tendency to push it
aside or ignore it. According to a recent study, 45% of corporate
executives now rely more on instinct than on facts and figures in
running their businesses. (See "Don't Trust Your Gut," Harvard
Business Review, May 2003). Yet researchers agree that the more complex
and difficult a decision, the less likely it is that intuition - when
detached from rigorous analysis - will yield positive results.
Adopting a Transformational
Strategy
Convinced that making good decisions is critical to their success
- particularly in an unforgiving economy - decision makers across
a wide variety of fields and organizations are changing their ways.
With the help of new software and technology, they are adopting a
more consistent, structured and rigorous decision-making strategy:
decision analysis.
Introduced approximately 35 years ago, decision analysis has a proven
ability to help users identify alternatives and options that maximize
value. Corporate decision makers now use the process for all kinds
of decisions involving resource allocation - from building a new plant,
to acquiring a company, to evaluating new market opportunities.
Oil companies such as Conoco and Chevron use decision analysis to
determine leasing strategies and make drilling decisions. Manufacturers
use the methodology to help determine how much inventory to produce.
Medical and pharmaceutical researchers use decision analysis to evaluate
alternative drugs or treatments, or to help set new pharmaceutical
prices. And lawyers use decision analysis to counsel clients on their
litigation risks and help them negotiate optimal settlements.
The common thread? High-impact decisions involving significant investment,
high complexity and elements of uncertainty.
How it Works
Decision analysis makes use of software tools to help users build
models that represent specific decision problems, and then relies
on statistical analysis to either determine the best course of action,
or to discover what information is required to make a good decision.
One of the method's virtues is that it allows decision makers to
quantify the uncertainties involved in a decision by expressing them
in terms of probabilities. This distinction helps ensure the inclusion
of some factors that decision makers might otherwise be tempted to
dismiss.
Modeling options include influence diagrams View Figure 1 >>> which
show all of the relationships relevant to making a decision, and decision
trees, larger branching structures whose paths represent all of the
reasonable alternatives and their outcomes View Figure 2 >>>. These models
provide complementary approaches to looking at a particular decision
problem.
Decision makers find that mapping alternatives on a decision tree
model helps visualize all of the factors affecting a decision and
each of the possible outcomes. Although it might theoretically be
possible to test each and every alternative scenario in the real world,
few would have the time or resources. By enabling virtual testing,
modeling allows decision makers to consider all their possibilities
- and provides the timely information needed to take action.
Time for Timely
Decision Making
Some medical researchers are now using decision modeling to test
different interventions and their effects, studying which treatments
will save more lives or prove to be most cost-effective. Victor Grann,
Clinical Professor of Medicine and Public Health at Columbia University's
School of Public Health, uses decision tree analysis to compare the
benefits of preventive treatment alternatives available to individuals
who test positive for cancer gene mutations. Obtaining that kind of
information from clinical trials, although important, will take many
years, he said. But with the help of decision analysis, Grann is providing
patients and their doctors with the information they need to make
informed decisions now.
In
some situations, delaying a decision can be more costly than
embarking on a less than perfect path. For example, while researchers
and policy makers debate the pro's and con's of alternative
screening methods for cervical cancer, 200,000 South African
women die from the disease each year for want of access to cancer
screening. |
| Decision analysis
can provide a basis for expedient action, enabling policy
makers to make the best decision given the circumstances. |
|
Professor Sue Goldie, who has been working to model various screening
strategies for women in third world countries, believes that in cases
like these, decision analysis can provide a basis for expedient action,
enabling policy makers to make the best decision given the circumstances.
(See "Good as Goldie," Harvard Public Health Review,"
Summer 2002.)
Decision makers know that complexity can often lead them to focus
on the wrong problems.
But because decision analysis incorporates a process called sensitivity
analysis, it can help determine which factors are really critical
to a decision's outcome. That knowledge keeps decision makers from
focusing on distinctions that make little difference, helping them
avoid "analysis paralysis" and move on to a final decision.
|
"You only want to spend time discussing something
relevant," said Gary Bush, Managing Director of Decision
Strategies Inc., an international leader in the training and
coaching of decision practices. By incorporating sensitivity
analysis, decision analysis provides actionable parameters that
help clients who are facing litigation reach a timely settlement. |
| When client
Chevron Texaco was negotiating the terms of a joint project
with an oil company partner, Decision Strategies used
sensitivity analysis to help the company identify the
variables that could make a real difference in the economics
of the project … The difference between the project
as originally proposed and the terms that Chevron finally
negotiated amounted to a few hundred million dollars in
value to Chevron. |
|
"If one lawyer says we have a 50% chance of winning, and another
lawyer says there's a 75% chance, with sensitivity analysis, we can
turn around and say, 'If it's less than 80%, we settle,'" he said.
When client Chevron Texaco was negotiating the terms of a joint
project with an oil company partner, Decision Strategies used sensitivity
analysis to help the company identify the variables that could make
a real difference in the economics of the project.That information
was then used to establish Chevron Texaco's negotiating position.
The difference between the project as originally proposed and the
terms that Chevron finally negotiated amounted to a few hundred million
dollars in value to Chevron, Bush said.
Rigorous, Objective
Analysis
With TreeAge Software's TreeAge Pro package, decision makers can
also consider the cost-effectiveness of possible alternatives. "I
don't think there is another tool that has those functions, "
Bush said.
The feature's users include leading pharmaceutical companies. By
establishing the cost effectiveness of new products vis a vis other
therapies, management can decide whether to invest in further development.
The same cost effectiveness tools can also help managed care organizations
or government policy makers (in Europe and Asia) decide if they will
provide reimbursement for particular therapies.
Decision analysis doesn't guarantee a good decision. Like other software-based
tools, it is subject to the perils of "garbage in, garbage out."
But when it is well executed, incorporating probabilities based on
accepted data and expert opinion, decision analysis generates highly
credible results. It is easy to trace the thought process behind an
analysis back through its decision tree. And familiar graphical tools,
including probability wheels and bar charts, help communicate results
in a way decision makers understand.
Thanks to the tool's credibility, decision analysis is beginning
to play an important role in mediation. In a recent class action suit
filed against Microsoft, Microsoft and lead counsel for the plaintiffs
retained Marc Victor - an expert in the use of decision analysis for
litigation risk analysis, and founder and president of Litigation
Risk Analysis, Inc. - to ensure that the mediation was firmly grounded
upon a rigorous and objective analysis of the legal and evidentiary
merits. The settlement agreement obligates Microsoft to provide the
plaintiff class up to $1.1 billion in vouchers.
Effective leaders have always prided themselves on their ability
to shape the future by making good choices. But today's strategic
decisions are just too complex to evaluate intuitively. By adopting
decision analysis, decision makers can implement a rigorous strategy
that will improve decision making for years to come - and lay a solid
foundation for future success.
Elsewhere in This Section
> Markov Models - State Transition Models
> Discrete Event Simulation Models
> Cost-Effectiveness Analysis
> Monte Carlo Simulation
Learn
how TreeAge Pro users are using decision analysis. >