Prompts For Everyone logo
PFE
Back to Glossary
Technique

Chain-of-Thought Prompting

Asking the AI to reason through a problem step by step before giving an answer.

Share

Definition

Chain-of-thought (CoT) prompting is a technique that instructs the AI to break down complex problems into sequential reasoning steps before arriving at a final answer. Instead of jumping straight to a conclusion, the AI "thinks out loud" — explaining its reasoning at each step. This dramatically improves accuracy on tasks that require logic, math, multi-step analysis, or decision-making.

Research from Google showed that simply adding "Let's think step by step" to a prompt can improve accuracy on math problems by over 40%. Chain-of-thought works because it forces the AI to process information sequentially rather than trying to pattern-match to an answer directly.

Examples

1

Adding "Think through this step by step before giving your final answer" to a complex analysis prompt

2

"First, identify the key variables. Then, analyze how they interact. Finally, recommend a course of action with reasoning for each recommendation." — structured CoT prompt

Related Terms

Prompt EngineeringFew-Shot PromptingAI Agent

Frequently Asked Questions

When should I use chain-of-thought prompting?
Use it for complex tasks: math problems, logical reasoning, multi-step analysis, decision-making, and debugging. For simple tasks like "write a tweet," it's usually unnecessary.
Does chain-of-thought work with all AI models?
It works best with larger, more capable models like GPT-4, Claude, and Gemini Pro. Smaller models may struggle to maintain coherent reasoning chains.

Build prompts using this concept

Explore our prompt library and put chain-of-thought prompting into practice with ready-to-use templates.

Build prompts using this concept