Intro:
What you put into an LLM has a huge effect on what you get out. Small changes in wording can lead to very different answers. Two prompts can be nearly identical except for a couple of words, and the model may respond in completely different ways.
That is why it helps to slow down and provide as much helpful context as you can. When you clearly explain what you want, why you want it, and who it is for, you greatly increase the chances of getting an accurate and useful response.
Why Prompting Matters
AI responds based on the information you provide. If your prompt is missing details, the model will fill in the gaps with assumptions, and those assumptions may not match what you meant. When your prompt is clear, the model can focus on the right goal and deliver a more useful result.
Good prompting can help you:
Get more accurate and useful answers
Save time by reducing follow-up questions
Produce outputs in the style, tone, and/or format you need
Prompt Formula
A simple formula can make prompting easier and more consistent:
“Act as a [persona]. Your task is to [action]. Include [format/details]. The target audience is [audience]. Output should be [output type]. Context: [background and helpful details].”
Here is what each variable means and how to fill it out:
Persona: Choose the role you want the AI to take, such as teacher, editor, product manager, executive, or data analyst
Action: State exactly what you want the AI to do, such as summarize, brainstorm, rewrite, compare, research, or generate a plan
Format or details: Describe how the answer should be structured, any key details you want the response to touch on, and any constraints
Audience: State who the output is for, such as a customer, a beginner, a technical team, or senior leadership
Output type: Specify the type of output you want, such as a report, a table, JSON data, an email draft, or a checklist
Context: Paste or upload any background that matters, such as goals, source material, definitions, examples, or things to avoid
Here’s an example:
“Act as an expert market research analyst. Your task is to research the live shopping market and explain who is winning and what consumer products perform best.
Include a short market overview, a table of top companies with why they are winning, a table of best performing product categories, and three to five recommendations on how to win as a product in this space
The target audience is senior leadership.
Output should be a short report with clear headings, two tables, and sources at the bottom.
Context: We are a clothing start-up and need a practical view of today’s live shopping winners and product trends to guide strategy.”
Extra Tips:
My favorite sentence to add to a prompt is: “Use simple language, don’t change the meaning, and do not use any dashes.”
Always Iterate. Feel free to tweak the prompt and run it again if you don’t get what you need the first time
Make sure to provide as much context as possible by pasting in text or attaching documents
If you are struggling to get the right output, show a short sample of what good output looks like.
If you found this helpful, consider sending it to a colleague or friend in your network!
Have a question or want to connect? Email me at treypezzetti@gmail.com