Decision Modeling with Tables and Trees

From Ambiguous Rules to Clear Logic

Business processes are guided by business rules. These are the specific, operational policies or constraints that direct decision-making. For example:

• "Premium customers who spend over $500 get a 15% discount."
• "If an ad campaign's click-through rate drops below 0.1% for three consecutive days, send an alert to the account manager."

When rules are simple, you can write them down as text. But as the number of conditions and outcomes grows, text becomes a tangled mess of "if-then-else" statements, which is hard to understand and easy to get wrong.

To solve this problem, we use two powerful techniques: Decision Tables and Decision Trees.

Decision Tables: Structuring Complex Logic

A decision table is a tabular method for organizing business rules in a way that ensures all possible combinations of conditions are considered. It's a fantastic tool for ensuring completeness and communicating complex logic to both business stakeholders and technical teams.

Anatomy of a Decision Table

A decision table is typically divided into four quadrants:

1. Conditions: The factors or criteria that are evaluated (e.g., "Customer Type," "Purchase Amount").
2. Actions: The possible outcomes or operations that can be taken (e.g., "Apply 10% Discount," "Offer Free Shipping").
3. Condition Entries (Rules): The different combinations of condition values (e.g., Customer Type = 'New', Purchase Amount > $100). Each column represents a unique rule.
4. Action Entries: An indicator (like an 'X') that marks which action to take for a given rule.

Let's look at a concrete example.

Why and How to Use Decision Tables

Decision tables are more than just an organizational tool. They bring significant benefits to a project, which is important to articulate during interviews or stakeholder discussions.

The key steps to creating a decision table are:

1. Identify all business rules related to a specific decision.
2. Deconstruct the rules into a set of unique conditions and actions.
3. Create the table structure with conditions on top and actions on the bottom.
4. Populate the table by creating a column for each unique and relevant combination of conditions (a rule).
5. Mark the corresponding action(s) for each rule.

The following video provides a detailed walkthrough of this process, starting from a set of written business rules.

Decision Trees: Visualizing Decision Paths

A decision tree is another way to model business rules, but it uses a graphical, tree-like structure. It's often preferred when the number of outcomes for each condition is small and when you want to create a model that is very easy for stakeholders to follow visually.

Anatomy of a Decision Tree

• Root Node: The starting point of the tree, representing the first condition to be evaluated.
• Branches: The lines connecting nodes, representing the possible outcomes or values of a condition (e.g., 'Yes'/'No' or 'High'/'Low').
• Internal Nodes: Subsequent conditions that need to be evaluated along a path.
• Leaf Nodes: The endpoints of the tree, representing the final actions or decisions.

Here is a simple example for a hiring decision.

Decision Trees vs. Decision Tables

Decision tables and trees can represent the same information. The choice between them often comes down to the nature of the problem and your audience.

For a more detailed look at creating decision trees and the best practices involved, let's turn to another resource.

Key best practices for decision trees include:

• Start with the most general condition at the root.
• Ensure the branches from any node are mutually exclusive and cover all possibilities.
• Keep labels concise and clear.

                      ┌─> [Pay Weekly]
(Publisher Tier?) ─┬─ (Premium)
                    |
                    │             ┌─> (> $100) ─> [Pay Weekly]
                    └─ (Standard) ─┤ (Account Balance?)
                                  │
                                  └─> (<= $100) ─> [Pay Monthly]

Conclusion

In this lesson, you've learned how to model complex business logic using two essential techniques: decision tables and decision trees. These tools help you move from ambiguous textual rules to precise, complete, and verifiable models.

Key Takeaways:

• Decision Tables excel at organizing complex rules with many conditions. They provide a structured, grid-based format that ensures no scenario is missed.
• Decision Trees are excellent for visualizing decision pathways in a flowchart-like manner, making them very effective for validation with non-technical stakeholders.
• Both techniques are critical for a Business Analyst to create clear requirements, reduce development errors, and ensure the final system behaves as the business expects.

Preview of the Next Lesson:
This lesson marks the end of our module on process and system modeling. You now have a powerful toolkit for analyzing and documenting what a system should do and how it should behave.

In the next module, we will pivot to a critical technical skill for any modern Business Analyst: data interrogation using SQL. Our first lesson will introduce the foundational SELECT statement, teaching you how to write your first queries to retrieve and filter data directly from a database. This is a crucial first step toward performing independent data analysis and providing data-driven insights.