Business analysis glossary
Prompt Engineering
What it is
Prompt engineering is the practice of writing clear instructions and context for an AI model to get useful, reliable output. A good prompt states the role, the task, the context, the format wanted, constraints and examples, and it is refined by testing.
Why it matters
The same AI tool gives very different results depending on the prompt. Analysts who prompt well get better first drafts of stories, test cases and summaries. Writing a good prompt is similar to writing a good requirement: specific, unambiguous and testable.
Example
Instead of asking an assistant to write user stories, the analyst writes: you are a business analyst at an insurance company; from the meeting notes below, write user stories in the as a, I want, so that format, with three Gherkin acceptance criteria each, including one error case, and list any open questions separately.