Zero-Shot Chain of Thought
Just add "Let's think step by step."
What It Is
Zero-Shot Chain of Thought is the simplest way to improve AI reasoning: add 'Let's think step by step' (or similar phrase) to any prompt. This triggers AI to break down its response into logical steps without you providing examples or detailed instructions.
It combines the simplicity of direct prompting with the improved accuracy of step-by-step reasoning — no extra setup required.
Why It Works
Step-by-step pattern activation: The phrase 'think step by step' appeared in training data alongside methodical problem-solving. It activates patterns where AI shows its work.
Intermediate context: Each step AI generates becomes context for the next prediction, improving accuracy on complex reasoning tasks compared to jumping straight to an answer.
When To Use It
- Problem has multiple steps
- Logical reasoning is required
- Math or calculation is involved
- Decision requires weighing factors
- You want to see AI's thought process
- Simple factual questions
- Creative writing (makes it rigid)
- Already providing step-by-step instructions
- Single-step tasks
- Speed is more important than accuracy
Ready-to-Use Templates
"We have 500 monthly visitors but only 10 conversions. Our target is 50 conversions. Let's think step by step."
"Should I hire two junior developers or one senior developer with my $150K budget? Let's approach this systematically."
"Analyze why our customer churn rate increased from 5% to 12% last quarter. Let's break this down."
"How do I migrate 10,000 customers from our old platform to the new one without service interruption? Let's work through this carefully."
"Calculate our break-even point if fixed costs are $50K/month, variable costs are $15 per unit, and we sell at $45 per unit. Let's think through this logically."
Pro Tips
- Position matters: 'Let's think step by step' works at the beginning or end of your prompt — end placement often works slightly better.
- Don't overuse: Save this for problems that actually need reasoning. Simple tasks don't benefit and it just adds length.
- Check the steps: AI's intermediate steps might reveal faulty logic even if the conclusion sounds good. Read the reasoning, not just the answer.
- Combine with other techniques: 'You are an accountant. Calculate... Let's think step by step.' Role + reasoning = strong combination.
- Upgrade when needed: If zero-shot CoT isn't enough, move to full Chain of Thought with explicit numbered step structure.
Understanding the Fundamentals
These techniques work because of how AI actually processes language. Our Making AI Make Sense series breaks down exactly why — making techniques like this intuitive instead of mysterious. 34 videos, free on YouTube.
Watch on YouTube →Need Help Applying AI to Your Business?
Templates teach the technique. We help you figure out where it actually fits in your work.
Eau Claire AI — Your path to AI, made clear.