KoreField
AI Foundation School/Phase 2: Quick Wins & Confidence

Module 3: Prompting That Actually Works

1-2 daysTeach effective prompting

AI Avatar Lesson

Video will be available when Cloudflare Stream is configured

1-2 days
Coming Soon

Why Most Prompts Fail

Most people write prompts the way they would text a friend — vague, incomplete, and missing context. AI models are powerful but literal. They respond to exactly what you give them. A prompt like 'tell me about marketing' will produce a generic essay. A prompt like 'List 5 low-budget digital marketing strategies for a small bakery in Lagos targeting young professionals' will produce something you can actually use.

The Anatomy of a Good Prompt

  • Context: who you are and what situation you are in
  • Task: exactly what you want the AI to do
  • Format: how you want the output structured (list, table, paragraph, code)
  • Constraints: length limits, tone, audience, what to include or avoid
  • Examples: show the AI what good output looks like (few-shot prompting)

Before and After

Bad prompt: 'Write about climate change.' Good prompt: 'Write a 200-word summary of how climate change affects smallholder farmers in East Africa. Use simple language suitable for a community newsletter. Include one actionable recommendation.' The difference is specificity. The good prompt tells the AI exactly what to produce, for whom, and how long it should be.

Prompting is a skill, not a talent. The more you practice writing specific, structured prompts, the better your AI outputs will be. Treat every prompt as a brief for a skilled assistant.

Key Takeaway

Good prompts are specific, structured, and include context, task, format, and constraints. Vague prompts produce vague results. Practice transforms prompting from guesswork into a reliable professional skill.

Review Questions

1. What is the most important difference between a bad prompt and a good prompt?

2. Which prompt would produce the most useful output?