What it does: Writes the prompt for Figma Make to build a 4 screen booking flow you can click through and show people before anyone codes it.
Use it when: You want to test how customers will book with you before paying for a website change.
You get: The words for each screen, one Figma Make prompt ready to paste, and 5 questions to ask the people who test it.
You are my product designer, a senior designer with 10 years of making booking screens that customers finish on a phone. I will build a clickable test version in Figma Make. Goal: write the words for a 4 screen booking flow and ONE prompt for Figma Make to build it, so I have a prompt ready to paste and a flow I can click through with real customers before anyone codes it. Context: I run [what you do and for who]. My customers book on their phone, often in a hurry. They want to book in a few taps and know what happens next, because a long or unclear form makes them leave. First, check that you can read the details I gave you about my business and what people book. Tell me how many services and booking details you found. If any are missing, use the defaults below and say so, or ask me to paste them instead. If I already have a Figma file or Figma Make project for this, check that you can open it and tell me how many screens it has. If you can't open it, stop and follow the Figma steps at simpleai.community/connect/, or I'll paste the screen names for you. Ask me these, then run. Skip any I already answered. 1. What do people book, how long, and the price if you show it? (Default: one service, [check: service], no price shown.) 2. What do you need from them to book? (Default: name, phone and a time.) 3. What happens after they book, and your colors? (Default: [check: what happens next] on Screen 4, and plain black and white.) Do this. The customer is the hero, my business stays in the background: 1. Write the words for 4 screens: - Screen 1: what they get, in under 10 words, and one button "Book a time". - Screen 2: pick a service and a time. - Screen 3: the fewest details I need. - Screen 4: done, and only what I said happens after they book. 2. Check each screen from the customer's side. Ask for nothing I didn't list, and ask for each detail only once. If a choice on Screen 2 already tells you something (like size), don't ask it again. 3. Write ONE prompt for Figma Make that holds those words and says: phone size first, big buttons, one color for buttons, the back button works on each screen, no log in. 4. Give me 5 questions to ask 3 real customers while they click it, like "What would you tap first?" Give me: the words for each screen as a short list, the Figma Make prompt in one block under 250 words, and the 5 questions as a numbered list. Example of Screen 1, match this: "A clean home by Friday. [Book a time]" Before you show me, check: - Every word on the 4 screens is in the Figma Make prompt, and every [check] note is carried into it. - No detail is asked twice, and Screen 4 says only what I told you happens next. Rules: - Drafts only. Never publish, share or invite anyone to the test version, and ask me before you change anything in Figma. - Show times as [time] unless I gave my open times. - If I already have a Figma Make project for this, ask me if the prompt should change that one or start a new one, and write it that way. - Use only what is in my data and what I tell you. Never make up a name, number, date, price, link or promise. - If something you need is missing, write [check: what is missing] in that spot and keep going. Last, save these steps as a skill called 'Booking Flow Test' so you do it this way each time I test a new booking flow. To run it again, just say: "Run my Booking Flow Test skill."
You need
Figma![]()
Any AITested on sample data
Built on Building a StoryBrand's rule that the customer is the hero.
Copilot version
There is no separate Copilot version of this one yet. Paste the prompt above into Microsoft 365 Copilot Chat.
See a real example
Tested on sample data Oct 29 2026, after the rewrite: a Claude agent ran it on made-up sample input (tool results simulated). A separate Claude agent judged the run and the card.
Watch: no video for this one yet.
AuthorCorey Leon