How to make and use decision trees
The success of a business depends on every team member making the best decisions they can, moment to moment, quarter to quarter, year to year. Sometimes these decisions are big: Where do we want to build our headquarters? And sometimes they’re smaller: Which social media management tool should we purchase? No matter the decision, a decision tree is a simple tool to explore your options and get to the ideal solution.
What’s a decision tree?
In essence, a decision tree is just a spicy flowchart. There are three parts to a decision tree:
- Root node
- Leaf node
The root node is the ultimate decision you’re trying to make. Each leaf node is a refining question. Branches connect everything to show the flow from questions to answers.
For instance, in the decision tree below, you’re trying to decide between building a completely distributed team or maintaining a hybrid group of in-person and remote employees. That’s the big question asked in the root node. In the decision tree model below, you can follow each choice to its probable outcome, ranked in order of high, medium, and low probability.
Advantages of using decision tree analysis
The advantages of decision trees really come down to the benefits of data-driven decision making. Here are some key advantages of decision trees:
- Comprehensive—Decision trees force you to look at all the possible outcomes of a choice. You’re able to better understand the risks and consequences of your decisions.
- Visual—Decision trees don’t rely on formulas. They’re easy-to-understand and the benefit of using models in decision making is that you can easily share them with others for input. This can help you gain buy-in from stakeholders.
- Adaptable—Just about every question can be answered using a decision tree and they can be as simple or as complex as you’d like.
- Reduce bias—Decision trees help cut through emotions, letting you properly weigh the results of one decision against another.
- Simple—Decision trees don’t require gathering much data. If there are gaps in the data, you can identify where you’ll need more information.
- Low-cost—Making a decision tree diagram is free and doesn’t require advanced training. There are plenty of free templates available.
- Save time—Depending on the complexity of the decision, a decision tree may be the fastest way to find a solution.
The benefits of decision trees far outweigh the learning curve, so they’re a useful addition to your decision-making process.
When to use decision trees
Decision trees are best used for more complex decisions. A decision tree analysis can help break through uncertainty and bring clarity. But bear in mind, a decision tree is best suited for clear-cut decisions and not for brainstorming solutions.
So, for instance, an appropriate decision tree question would be: “Should we build a new website or overhaul the existing one?” Then, you can start investigating the consequences of building or overhauling the website. An inappropriate decision tree question would be: “What soup is Jared’s favorite?” That’s not a decision: it’s just a question (and the answer is minestrone).
Common use cases/applications for decision trees
- Product planning—For instance, coming up with an entirely new product and choosing between two ideas or determining which feature to add first.
- General business decisions—Keep the business remote or move headquarters to Prague?
- Software design—Should you use Drupal or Blogspot?
- Loan approval—Knowing when to lend money is a big choice. Mitigate financial risk by using a decision tree when determining if a loan should be approved.
- Personal decisions—Your commute says Prius, but your heart says Mustang. Decision trees can help you make a more rational decision.
Decision tree symbols
Before we dissect a decision tree, it's helpful to know some of the common symbols. Here are some normal decision tree symbols below:
⃞ Root node: This is the decision that must be made.
⬦ Leaf node: This is a decision or a test.
͢ Branch: This connects outcomes to their decisions or tests.
⃝ Connector: This can be used to show yes/no decisions or simple answers.
◀ Endpoint: This is the ultimate outcome
How to make a decision tree
1. Begin with the decision
In the root node box, enter the decision you’d like to make. For instance, if you’re deciding whether to outsource your customer service or make an in-house department, your root node would say: “How should we deploy customer service?”
2. List your options
Next, use connectors to list out your options and connect them to the root node with branches. Here, our connectors say “In-house” or “Outsource.”
3. Put it to the test
Apply the same tests to each option. In this example, using a leaf node connected with branches to the connectors, ask: “Is it cheaper?” Then answer the question. Other questions you might ask include:
- Will we have direct oversight?
- How much training will it require?
- Will it improve customer satisfaction?
4. List out conclusions
At the endpoint of each stream of questions, write the ultimate conclusion. You may also choose to show risk in your conclusions by showing degrees of probability (high, medium, low).
Decision tree tips:
- The bulk of your decision tree will be leaf nodes. Each branch should have a leaf node.
- Don’t be afraid to add branches if you think of more outcomes that should be considered.
- You don’t have to go it alone. This can be a group project and should involve stakeholders.
- Make sure to apply the same test to each outcome. If you ask, for instance, “will it be easy to implement?” of one option, you must ask the same question of the other options.
Tree diagramming challenges
There are some things to be aware of as you begin your decision tree journey. Knowing their limitations before you begin is great.
- Watch out for chain reactions—If you’ve tested a question early on in your tree and later realize you made a mistake, all branches and nodes will be affected. That’s why it’s best to use cloud software to design your tree so that it can be easily rearranged if data changes.
- Remember that decisions are based on expectations—Decision trees don’t plan for all variables (say, for instance, your business having to become fully remote due to a pandemic). While decision trees are great for helping make better decisions, they’re only as good as the information you have to work with.
- Plan for some level of complexity—Even simple decisions can lead to massive decision tree diagrams. These can quickly become unwieldy. Worse, they can become confusing and lead to errors that could lead to disingenuous outcomes.
That said, despite its limitations, a decision tree model may just be the guidance you need to make an important decision. If you benefit from visual representation, are managing a team, have been struggling with indecision, give them a try. You’ll likely feel much more confident knowing you’ve carefully weighed your options and made the best choice with the data you have.
Weigh all the possible options and make your own decision tree today in Lucidspark.Get started
Create your own decision tree when you sign up for Lucidspark today!Sign up now
Lucidspark is a virtual whiteboard that helps you and your team collaborate to bring the best ideas to light. It comes packed with all of the sticky notes, freehand drawing tools, and infinite canvas space you need to capture that next big idea. And it’s built for collaboration. Think of it like a sandbox where your team can bounce ideas around and innovate together in real time.
Brought to you by the makers of Lucidchart, trusted by 25 million users worldwide, including 99% of the Fortune 500.