RISE vs Chain-of-Thought: Which Prompting Framework Should You Use?

When I teach prompt engineering workshops, one question comes up constantly: "Should I use RISE or Chain-of-Thought?" The honest answer is that they solve different problems—and understanding when to use each can dramatically improve your AI outputs.

The Core Difference

RISE (Role, Instructions, Steps, Expectations) is a structural framework. It's about giving the AI a clear identity and comprehensive context upfront.

Chain-of-Thought (CoT) is a reasoning framework. It's about getting the AI to show its work and think step-by-step through a problem.

Think of it this way: RISE tells the AI who to be and what to do. Chain-of-Thought tells the AI how to think.

RISE Framework Breakdown

RISE structures your prompt into four components:

| Component | Purpose | Example | |-----------|---------|---------| | Role | Establish expertise | "You are a senior financial analyst" | | Instructions | Define the task | "Analyze this quarterly report" | | Steps | Outline the process | "First examine revenue, then costs, then margins" | | Expectations | Set output requirements | "Provide a 500-word summary with bullet points" |

When RISE Works Best

• Content creation tasks (writing, editing, summarizing)
• Tasks requiring specific expertise or perspective
• Outputs that need consistent formatting
• Situations where you want comprehensive, structured responses

RISE Example

Role: You are an experienced B2B copywriter who has written for SaaS companies like Slack, Notion, and Asana.

Instructions: Write a landing page headline and subheadline for a project management tool aimed at remote teams.

Steps:
1. Consider the main pain points of remote team collaboration
2. Focus on the emotional benefit, not features
3. Keep the headline under 10 words
4. Make the subheadline expand on the promise

Expectations: Provide 3 options, each with a headline and subheadline. Explain your reasoning for each.

Chain-of-Thought Breakdown

Chain-of-Thought prompting asks the AI to reason through problems explicitly before arriving at an answer. The key phrase is usually "Let's think step by step" or "Work through this systematically."

When Chain-of-Thought Works Best

• Math and logic problems
• Multi-step reasoning tasks
• Situations where accuracy matters more than speed
• Complex analysis requiring multiple considerations
• Debugging or troubleshooting

Chain-of-Thought Example

A store sells apples for $2 each and oranges for $3 each. Maria bought some apples and oranges for exactly $20. She bought more apples than oranges. How many of each did she buy?

Let's work through this step by step.

The AI will then systematically:

1. Set up the equation (2a + 3o = 20)
2. Find valid combinations
3. Apply the constraint (a > o)
4. Arrive at the answer

Head-to-Head Comparison

| Factor | RISE | Chain-of-Thought | |--------|------|------------------| | Best for | Creative/structured tasks | Analytical/reasoning tasks | | Output style | Comprehensive and formatted | Explanatory and logical | | Speed | Faster (direct output) | Slower (shows reasoning) | | Accuracy | Good for content | Excellent for logic | | Learning | Teaches AI "what" | Teaches AI "how" |

Can You Combine Them?

Absolutely—and this is where things get interesting. For complex tasks, you can use RISE to set up the context and Chain-of-Thought to guide the reasoning.

Combined Example

Role: You are a senior data scientist with expertise in customer churn prediction.

Instructions: Analyze why our churn rate increased from 5% to 8% last quarter.

Steps: Work through this systematically:
1. First, identify potential factors that typically drive churn
2. Then, consider which factors are most likely given a B2B SaaS context
3. Next, prioritize these by likelihood and impact
4. Finally, suggest what data we should examine to confirm each hypothesis

Expectations: Present your reasoning clearly, showing how you arrived at each conclusion. Provide a prioritized list of 5 potential causes with your confidence level for each.

The Decision Framework

Here's how I decide which to use:

Choose RISE when:

• You need creative or written content
• Format and structure matter
• You want the AI to adopt a specific perspective
• The task is relatively straightforward

Choose Chain-of-Thought when:

• The problem requires multi-step reasoning
• Accuracy is critical
• You need to verify the AI's logic
• The task involves math, code, or complex analysis

Combine them when:

• The task is complex AND requires a specific output format
• You need both expert perspective AND transparent reasoning
• You're doing analysis that will be presented to others

Common Mistakes

The most common question I get is: "I used Chain-of-Thought but the output was still wrong." Usually, this happens because:

1. The problem wasn't actually a reasoning problem (should have used RISE)
2. The initial information was ambiguous
3. The task needed domain expertise that CoT alone couldn't provide

Similarly, RISE can fail when the task actually requires step-by-step logic that benefits from being made explicit.

My Recommendation

Start with this simple test: Is your task primarily about what to produce or how to think?

• What to produce → RISE
• How to think → Chain-of-Thought
• Both → Combine them

Once you've used both frameworks for a few weeks, you'll develop an intuition for which fits each situation. The frameworks aren't competing—they're complementary tools for different types of problems.

---

Want to apply these frameworks instantly? Try PromptWizz to automatically optimize your prompts using the best framework for your task.