What Is Decision Analysis? (With Definition And Examples)
By Indeed Editorial Team
Published 25 October 2022
The Indeed Editorial Team comprises a diverse and talented team of writers, researchers and subject matter experts equipped with Indeed's data and insights to deliver useful tips to help guide your career journey.
Businesses take various strategies to find an optimal solution to a problem. One such strategy is to conduct a decision analysis (DA) that helps them find the best possible outcome from a set of potential outcomes and make an informed decision. If you are a manager or work in a similar role, you can benefit from learning about this technique and how you can apply it in the workplace.
In this article, we discuss what decision analysis is, highlight its importance, share steps on how to make an effective decision, outline the limitations of this technique and provide examples to help you understand the concept better.
What Is Decision Analysis?
Decision analysis is the process of taking a systematic, structured and visual approach to finding the most optimal solution to a business problem. It involves finding all potential options and choosing one that either minimises or maximises the objective function. For example, if you are looking to reduce operational costs in a company and you have eight potential solutions that you can implement, you require choosing one that reduces the cost to its minimum value. Here, the objective function is to minimise operational costs.
Some applications of this technique in the workplace include relocating, restructuring, outsourcing, project feasibility analysis, launching new products and pricing. To make an effective decision, you require:
Information related to the decision
Viable options to a specified problem
Quantitative data to make a data-driven decision
Visual tools to better understand the workflow
How To Make An Effective Decision
Follow these steps to learn how to make an effective decision in the workplace:
1. Identify the problem
Identifying the problem is the first step in making an effective decision. You can do this by collecting the required information and talking to stakeholders. For example, a manager observes that a company's sales are declining over a period. The manager identifies the problem may be because of the quality of customer services, competitor's pricing and lack of innovation in the product.
2. Research the available options
After identifying the problem, start looking at the potential solutions or options to solve the problem. This requires conducting thorough research and collaboration with other experts. This provides data you can use to assess the potential outcomes of an option and to build a decision model.
Continuing the previous example, if the manager wants to focus on creating new products to increase innovation, they can organise brainstorming sessions with the team, conduct customer and competitor research, design an implementation roadmap and analyse the risks involved. After meticulously researching potential solutions, they can list a set of features that they want to implement in the new product.
3. Create a framework
After researching all workable options, it is essential to evaluate their outcomes. You can do this by introducing key performance indicators (KPIs) that measure the progress towards the intended goal. The criteria of measurement may either be qualitative or quantitative. From the previous example, the manager wants to maximise the return and improve sales by launching a new product in the market. After having listed the features to be incorporated into the new product, they design various KPIs to choose from features that are most applicable to meet their goals.
Consider another scenario where the manager wants to focus on building product awareness and capturing new customers instead of improving sales and revenue. This would require them to use a different set of KPIs.
4. Develop a decision model
This step involves integrating the framework with a decision model to visualise the potential solutions and their outcomes. You can use the following models:
Decision trees are complex flowcharts that display multiple pathways for decisions and outcomes. They comprise chance, decision and end nodes, where chance nodes contain the probability of results, the decision nodes are points where the flow branches into several other options and the end nodes show the ultimate outcome of a particular decision path.
When drawing a decision tree, draw a square or rectangle, which is the root node, to represent the initial decision. Next, draw circles to represent the potential outcomes of that decision. These circles are the leaf nodes of a decision tree. The root node remains on the left, whereas the outcomes branch out to the right. Draw triangles to denote the final outcomes of a decision. These triangles can contain values, such as revenue. Connect the root node, the leaf nodes and the end nodes. Add descriptions to these branches, making it easier to follow various pathways.
Influence diagrams are visual representations of a decision problem. You can use this model to visualise the major factors that influence a decision, the impact of one factor on others and how various outcomes relate to one another. Follow these steps to draw an influence diagram:
Start the tree. The first node contains the primary question or factor related to a decision. You can use a rectangle for this.
Add various factors. Add a series of secondary factors, sub-decisions and chances that constitute a decision. You can use various shapes, such as circles for uncertainty and parallelograms for values.
Add connectors. Add connectors to each of the shapes to showcase the flow of influence in the diagram. It is necessary that each box or shape has at least one connection.
5. Find the expected value
The expected value (EV) provides the weighted average of each outcome in the decision-making process. This requires assigning numerical values to various outcomes. After assigning values to the options, multiply them with each outcome's probability. This gives the partial value of each outcome. Add all the partial values to obtain the expected value. After this, compare all the values and choose an outcome with the maximum value. Review the formula:
EV = ∑ (probability of an event x value)
Why Is This Analysis Important?
DA is important for the following reasons:
Helps make complex decisions
Provides clarity in the decision-making process
Helps justify the decision to the management
Enables individuals and team members to visualise the decision-making process
Challenges To Decision Analysis
The following factors are challenges to the decision-making process:
Uncertainty: There are situations where there may not be enough data available to make a decision. This makes it difficult for a decision-maker to evaluate the consequences of potential outcomes.
Bias: Biases and personal views may influence the decision-making process. This can cause systematic errors and result in misjudging threats and risks.
Time constraint: There are situations when decision-makers require making immediate decisions. This does not allow them to collect enough data to perform the analysis.
Rationale: Decision-makers may look for solutions that can benefit them instead of ones that are actually optimal, also known as bounded rationality.
Related: Types Of Graphs And Charts
Examples Of The Decision-Making Process
Here are some examples of a decision-making process using DA:
Here is an example of a retail store that wants to open another store in a different city:
A retail store wants to expand its operations to another city. They perform a thorough analysis of the potential retail location, demographics, local competitors, real estate pricing and incentives. They determine the probability of success and the forecasted revenue of opening a store in one of the two cities.
City A: Launching a store in city A can generate up to ₹30,00,000 with a probability of 70% and may result in a loss of ₹9,00,000 with a probability of 30%.
City B: Launching a store in city B can generate up to ₹40,00,000 with a probability of 60% and may result in a loss of ₹10,00,000 with a probability of 40%.
Expected value of option = (potential revenue x probability of success) + (expected loss x probability of failure)
Expected value of option A = (₹30,00,000 x 0.7) + (-₹9,00,000 x 0.3) = ₹18,30,000
Expected value of option B = (₹40,00,000 x 0.6) + (-₹10,00,000 x 0.4) = ₹20,00,000
Based on the above calculation, the retail store can consider expanding its store in city B.
Here is an example of an IT company that wants to choose a project which can generate more revenue:
An IT company wants to determine the profitability of two projects and choose one that generates more revenue. They determine the probability of success and the forecasted revenue for each of these projects.
Project A: It can generate around ₹20,00,000 with a probability of 70% and ₹10,00,000 with a probability of 30%.
Project B: It can generate approximate ₹15,00,000 with a probability of 50% and ₹10,00,000 with a probability of 50%.
Expected value of project A = (₹20,00,000 x 0.7 + ₹10,00,000 x 0.3) = ₹17,00,000
Expected value of project B = (₹15,00,000 x 0.5 + ₹10,00,000 x 0.5) = ₹12,50,000
Based on the above calculation, the company can generate an additional ₹4,50,000 if they choose project A.
Explore more articles
- 6 Software Tools For A Financial Advisor (With Tips)
- Computer Scientist Skills (With Definition And Examples)
- What Is Account Balance? (Definition, Types And Examples)
- 10 Characteristics Of Big Data And How You Can Use Them
- 10 Digital Signage Software Tools (With Salient Features)
- 5 Useful Tips To Achieve Career Fulfilment (With Definition)
- How To Become An Ecommerce Specialist (With Skills)
- Continuous Employee Growth: Benefits And How To Encourage It
- A Guide On Financial Planning And Analysis (With Functions)
- What Is A BSN Degree? Definition And Career Opportunities
- 8 Types Of Estimation For Projects (With Definition)
- What Is The Bottom Of The Funnel? With Ideas And Tips