📎 This article includes 1 downloadable practice file ↓
In this article
Most poor answers come from prompts that leave the model guessing. A simple structure removes the guesswork.
The four parts
| Part | Say | Example |
|---|---|---|
| Role | Who it should act as | “You are an experienced accounts receivable executive.” |
| Task | Exactly what to produce | “Write a polite payment reminder email.” |
| Context | Facts it needs | “Customer: Gupta Agencies. Invoice INV-2026-0145, ₹1,28,500, due 15 Sep, 20 days overdue. Good customer for 5 years.” |
| Format | Shape of the answer | “Under 120 words, subject line + 3 short paragraphs, firm but friendly.” |
Before and after
Before: “Write an email about late payment.” → generic, often too harsh or too long.
After: the four parts above → ready to send with minor edits.
Add constraints
- “Don’t mention legal action.”
- “Use Indian English and ₹.”
- “If information is missing, ask me instead of guessing.”
💡 The last constraint, ask instead of guessing, is one of the most useful lines you can add to any prompt. It turns silent assumptions into questions.
Practice
Rewrite the three weak prompts in the worksheet using the template.
📎 Practice files for this article
- 📄Lesson 2 prompt worksheetCopy-paste prompts and exercises to try in ChatGPT, Copilot, Gemini or Claude.⬇ TXT · 302 B
Free to use for learning. Files with macros (.bas) are plain text — import them with Alt+F11 → File → Import File, and always test on a copy.
✨ Ask AI about this article
Stuck on a step? Ask a question and the AI answers using this article.
Free · AI can be wrong