Signing in

You will be sent to MillerKnoll sign-in.

AI

Prompting

Most people treat AI like a search engine. They type a few words and hope. That is not prompting. That is guessing out loud.

Lesson 1

Ask for what you actually want.

A prompt is an instruction, not a question. The more specific you are about who the output is for, what it should do, what it should avoid, and what format it should take, the more useful the response. Vague brief, useless output. Specific brief, useful output. That relationship is reliable.

Florence Knoll sketched exactly what she wanted on the dinner table before the caterers arrived. Not because she was difficult. Because a sketch is a brief that works.

Prompt as instruction versus vague question

Vague ask"Help with email?"Guessing out loudInstructionAudience · format · factsConstraints · avoid
A prompt is an instruction: audience, constraints, facts, and what to avoid, not a few words hoping for luck.

Core principles

  1. Start with audience and purpose. Who reads this, and what should they do after?
  2. Add constraints. Length, format, tone, what to avoid.
  3. Give it facts. Paste in the actual notes, the actual data, the actual context. Do not make AI guess.
  4. Name what you do not want. "Avoid corporate speak" is as useful as "be clear."

Go deeper: Prompt library & tactics →

Check yourself

What is the key distinction this lesson draws between an effective prompt and a weak one?

Do this in Copilot

Run the before-and-after exercise with your real work context before you mark this lesson complete.

Paste this into Copilot Chat and work through it before moving on.

The prompting prompt

I want to get better at prompting. Before we start, ask me three questions: what kind of work I do most often, what I'm usually trying to produce with AI, and where my outputs most often fall short. Then, based on my answers, give me a before-and-after example of a weak prompt versus a strong prompt for my specific situation. Walk me through what changed and why.
Open Copilot →
  • Before-and-after comparison

Did you run this in Copilot? Mark complete when you have tried it.

Next lesson: Context →