Most people give AI vague requests and accept whatever emerges, but the gap between 'write something professional' and 'write exactly what I need to say' is enormous—and bridging it is a learnable skill. The difference between a prompt that produces boilerplate and one that produces something actually useful to you is usually just a few more sentences of specificity about context, tone, and what success looks like.
Think of a prompt like ordering coffee at a café. If you say "coffee," you might get burnt diner coffee. If you say "medium oat milk latte, 150 degrees," you get exactly what you want. Same with AI—the more specific your question, the better your answer.
A prompt is simply the instruction you give an AI tool. Most people ask vague questions and wonder why answers are unhelpful. The secret isn't that AI is limited—it's that you need to be precise about what you want.
A good workplace prompt includes context, a specific task, and an output format. Instead of "Analyze this email," try: "I received this email from my manager. Identify any passive-aggressive language, contradictions with previous statements, and underlying accusations. Format as a bulleted list with direct quotes."
Notice the difference? The second prompt tells AI what to look for, how to structure it, and what matters to you.
Weak: "Help me with this HR situation."
Strong: "My manager told me in a meeting that I 'didn't contribute,' but my email shows I sent three proposals. Identify the contradiction and suggest how to address it professionally."
Weak: "Look at these messages."
Strong: "Here are five messages from my team lead over two months. Do you see a pattern of exclusion, dismissal, or shifting goalposts? List specific examples of each if present."
Always tell AI how you want the answer. Want a timeline? Say "Format as a chronological timeline." Want evidence for HR? Say "Structure with direct quotes and dates for documentation purposes." Want it simple? Say "Use plain language, no jargon."
Try this: Take a workplace issue you're facing. Write two prompts—a weak one and a strong one using the template above. Submit both to Claude or ChatGPT and compare. You'll immediately see how specificity changes output quality.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.