Decision-Making Tools for Better Outcomes

The ability to make a decision with a high probability of a successful outcome is one of the most coveted skills in the business world. Choosing between divergent paths can often spell the difference between organizational growth and stagnation.

Innovation and success can come about as the result of thoughtful and strategic decisions, but some entrepreneurs and decision-makers don’t always have the necessary tools at hand. Companies and strategic planners can learn a model for applying specific tools to varying types of decisions with the help of a Harvard Business Review paper entitled, Deciding How to Decide: A Tool Kit for Executives Making High-Risk Strategic Bets, by Hugh Courtney, Dan Lovallo, and Carmina Clarke.

In their paper, the authors assert that most executives over-rely on basic tools like cash flow analysis when it comes to the decision-making process because they don’t know how to match the tools to the situation. These basic decision tools can be unproductive in highly complex and uncertain contexts.

Decisions today tend to be made in environments with incomplete information. Although tools such as case-based decision analysis and qualitative scenario analysis are helpful in these situations, the decision-maker is often unclear about which tool or tools should be used. The authors describe a model for matching the decision to the decision-making tool. They utilize three factors:

  1. How well are the variables that determine success understood?
  2. How predictable are the possible outcomes?
  3. How centralized is relevant information?

Building a Decision Profile

Decision-makers will need to ask themselves if they know what it takes to succeed; that is, what success factors and economic conditions will lead to an outcome that can be labeled as successful. A strong causal model enables the decision-maker to be able to answer certain “if-then” scenarios with a high degree of certainty.

Once the first two factors are understood, it is possible to identify which decision-support tool will be most suited to the circumstances. Depending on the situation, causal model, and ability to predict an outcome, some of the decision support tools might include:

  • Conventional capital-budgeting tools such as discounted cash flow, expected rate of return and net present value
  • Quantitative multiple scenario tools such as Monte Carlo methods, decision analysis and real options
  • Qualitative scenario analysis using representative scenarios and likely consequences
  • Case-based decision analysis using equivalent past experiences and examples

The following decision tools should be used for each of the situation profiles:

If you understand the causal model and it is possible to predict outcomes or a range of outcomes with relative certainty then use conventional tools such as:

  • Discounted cash flow
  • Expected rate of return

If you understand the causal model, as well as probabilities for possible outcomes then use:

  • Quantitative multiple scenario tools

If you understand the causal model, but cannot predict outcomes, then use:

  • Qualitative scenario analysis combined with case-based decision analysis

If you don’t understand the causal model, but can predict outcomes, then use:

  • Case-based decision analysis

If you don’t understand the causal model and also cannot predict outcomes, then use:

  • Case-based decision analysis

If analogies are to be used as part of a case-based analysis to inform decisions, business leaders must be sure to do so in a rigorous manner. They cannot focus only on those cases which support the decision they would like to make, but must include a broader range of analogous cases.

How Centralized is Relevant Information?

Information aggregation tools function independently of the first two factors. If information is decentralized, the decision-makers must tap the knowledge of experts and aggregate their knowledge using tools such as the Delphi approach to help predict possible outcomes. Other possible approaches to gathering decentralized information include tapping prediction markets, using incentivized estimates, and accessing similarity-based forecasting. Each tool has its advantages and limitations depending on the amount of known information and amount of confidential information that the company might need to disseminate.

For more specific scenarios and a glossary of decision support tools, view the complete report.

facebooktwittergoogle_plusredditpinterestlinkedinmailfacebooktwittergoogle_plusredditpinterestlinkedinmailby feather