C.W.K.
Stream
Lesson 01 of 10 · published

Chain-of-Thought — When It Helps and When It Hurts

~18 min · reasoning, cot

Level 0Apprentice
0 XP0/100 lessons0/14 achievements
0/120 XP to next level120 XP to go0% complete

The most cited prompt trick of the GPT-3 era

Adding "Let's think step by step" to a prompt was the original Chain-of-Thought (CoT) finding: it raised math-word-problem accuracy on small models by double digits. The technique survived into 2026 but its leverage shifted. On a frontier model, CoT helps in a narrower band — and on the wrong tasks, it hurts.

Where CoT still helps

  • Multi-step problems where the steps must align (math, logic, planning, debugging traces).
  • Tasks where the model has been overshooting — guessing without checking.
  • Non-reasoning models that don't think internally.

Where CoT hurts

  • Trivial tasks (classification, formatting). The thinking burns tokens for no quality gain.
  • Reasoning models with native thinking. "Think step by step" double-thinks.
  • Tasks where you need a tight, structured output. The thinking pollutes the response.

The compromise: structured CoT

Ask the model to think inside a tagged region the consumer ignores: <thinking>...</thinking> followed by <answer>...</answer>. You get the reasoning trace for debugging without polluting the output schema.

Code

Structured CoT pattern·markdown
Reason in <thinking> tags. Then return the answer in <answer> tags.

<thinking>
  Step through the problem. Show edge cases.
</thinking>
<answer>
  {"verdict": "...", "reason_codes": [...]}
</answer>

External links

Exercise

Find one prompt where you reflexively added 'think step by step.' Remove it on a reasoning model and see whether output quality changes. Add it back where it earned its place.

Progress

Progress is local-only — sign in to sync across devices.
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.