Prompt engineering simply means writing clear instructions for an AI assistant. You do not need to code. A few habits make a large difference to the quality of the answers you get.
What is a prompt?
A prompt is the instruction and information you give an AI tool. The more clearly it describes the task, the audience and the result you want, the more useful the answer tends to be.
What makes a good prompt?
- Context: who you are and what the situation is. "I manage a small accounting practice in Pretoria…"
- Goal: what you need. "…and need a one-page checklist for new clients."
- Format: how you want it. "Use a numbered list with no more than ten points."
- Examples: show a sample of the style you want, if you have one.
- Constraints: what to avoid. "Plain English, no jargon."
How should I check AI output?
AI tools can produce confident but wrong answers. Check facts, figures, names and references before you use them, especially for anything legal, financial, medical or client-facing. Treat the output as a first draft written by a fast but unreliable assistant.
What should I never put in a prompt?
Do not paste personal information about clients or staff, passwords, or confidential business information into public AI tools unless your organisation has approved the tool for that use. South Africa's Protection of Personal Information Act (POPIA) applies to how you handle personal information.
How can my team learn this properly?
Hands-on practice with real work tasks is the fastest way to build the skill. You can find trainers for Prompt Engineering and Generative AI Fundamentals, or AI trainers for teams.
This guide gives general information, not professional advice for your specific situation. Spotted something out of date? Let us know.