Your choice also saves this guide’s current worked-example checkmarks. Only this browser; no account or sync. View saved guides
Saving needs JavaScript. Reading and downloading this guide does not save it.
Make a start
One step at a time.
3 steps
Start with the synthetic worked example or other non-sensitive inputs. Run one prompt at a time in your chosen AI tool. Replace any bracketed placeholders with the requested material, check each reply, and supply the source or earlier outputs that step needs. Paste them again if the tool lacks that context.
Start with the synthetic example, then adapt each prompt to your task.
01
Open a chat and use the complete sample
Open https://chatgpt.com. If you have access, choose New chat. Copy all the instructions below (the prompt), including the SAMPLE INPUT, into the message box and send it. You do not need to upload a file. You should see six ordered questions that draw on each panelist’s supplied experience without inventing outcomes. The wording can differ from the worked example, but the facts and constraints must agree. If you cannot access this service, use another text assistant available to you; account requirements and limits may differ.
Your starting prompt
Write a six-question panel sequence with named recipients and a one-sentence purpose for each question. Ask for examples and limitations rather than assuming success. Avoid assigning experience that a panelist does not have. Use only the supplied facts; mark missing information rather than inventing it. Return written text only. Do not send, publish or change anything in another service.
SAMPLE INPUT
Made-up example for practice. All people, organizations, records and product behavior below are fictional.
Fictional panel: Making shared studios easier to use. Panelist A, Sol: introduced a tool sign-out sheet in a ceramics studio. Panelist B, Ren: redesigned shelf labels in a print studio. Panelist C, Dev: coordinates volunteer opening and closing duties at a shared workshop. Only those experiences are approved. Goal: help attendees think about one small improvement to a shared workspace. Need six ordered moderator questions: opener for all, one specific question for each panelist, one comparison for all, one closing question for all. No claim that any change improved results; no private member details.
02
Compare, correct and save
Compare the answer with the complete sample input and the checks below. There are six questions: all, Sol, Ren, Dev, all, all. If a fact is missing, changed or invented, paste the correction prompt below into the same chat. Read the revised result again; if it still fails, correct the text yourself before using it. Copy the reviewed result into a new note or text document. Keep real publication, decisions and sending for a separate human review.
Your starting prompt
Check your previous answer against the SAMPLE INPUT in my first message. Test each point below. For a failure, show the problem and then return a complete corrected answer; for a missing source fact, keep it explicitly unknown. Do not invent a successful check.
- There are six questions: all, Sol, Ren, Dev, all, all.
- Each targeted question matches the person’s supplied experience.
- No question assumes success, quantified impact or private member information.
03
Optional: use your own facts
Open a new chat. Copy the prompt below and replace [PASTE YOUR INPUT HERE] with your own non-sensitive source text, facts and limits for this task. Include any desired headings, count or length. Send the whole message. Expect the same kind of written result, using your facts. Check every source fact and limit yourself; the example’s names and totals do not apply. If a detail differs, show the exact mismatch and ask for a complete correction. Save the checked draft before any separate decision to use or send it.
Your starting prompt
Write the requested number of panel questions from the supplied roles, actions and topic below. Name recipients and give each question a short purpose. Ask for examples and limitations without assuming success, difficulties or experience not in the source.
Use only my input. Ask for information needed to finish rather than guessing. Return written text only; do not send, publish, book, submit or change another service.
MY INPUT
[PASTE YOUR INPUT HERE]
Optional: improve the result
Prepare non-leading follow-ups
Have a useful prompt ready when an answer stays abstract. This extension uses the worked example. Return to that example chat, or paste its input and your checked example answer into a new chat first. Do not apply its sample values unchanged to your own input.
Advanced prompt
Add one optional follow-up to questions 2, 3 and 4 that asks for a specific example or unresolved limitation. Do not invent what any panelist will say. Keep the same SAMPLE INPUT from my first message unless I explicitly replace it.
Show, then try
See the shape of the result.
This input is synthetic. The output is a written illustration of the intended result, not a verified AI-tool run.
01 / The starting point
Synthetic input
Made-up example for practice. All people, organizations, records and product behavior below are fictional.
Fictional panel: Making shared studios easier to use. Panelist A, Sol: introduced a tool sign-out sheet in a ceramics studio. Panelist B, Ren: redesigned shelf labels in a print studio. Panelist C, Dev: coordinates volunteer opening and closing duties at a shared workshop. Only those experiences are approved. Goal: help attendees think about one small improvement to a shared workspace. Need six ordered moderator questions: opener for all, one specific question for each panelist, one comparison for all, one closing question for all. No claim that any change improved results; no private member details.
Copies text only. For image or file tasks, use the source files named in the steps too.
Worked example written for this guide, not output recorded from a live AI session.
1. All: What does your work in this part of the studio involve, and what needs or difficulties, if any, have you noticed?
Purpose: Establish each person’s role and any reported need without assuming there is a problem.
2. Sol: How did you decide what the tool sign-out sheet needed to record, and what did you leave out?
Purpose: Explore the design choices behind one concrete change.
3. Ren: Can you describe one shelf label you changed and what evidence, if any, you have about whether people understood it?
Purpose: Seek evidence of understanding without assuming improvement.
4. Dev: What information does a volunteer need to complete opening or closing duties, and where do uncertainties arise?
Purpose: Surface coordination needs in routine work.
5. All: Which part of another panelist’s approach might transfer to your space, and what would need to change?
Purpose: Compare approaches while respecting context.
6. All: What small first step could an attendee try, and what should they observe before deciding it helped?
Purpose: Close with a practical experiment and an honest check.
These checks use the worked example and its particular facts. With your own material, check your actual names, numbers, sources and rules; do not reuse the sample’s values as the answer.
Unsaved checks stay in this tab and reset when you reload. If you choose Save guide, its current example checkmarks are kept in this browser when storage is available.
!
A few things to watch for.
“How much did this improve things?” assumes an outcome that may not exist.
Questions should open a conversation, not prescribe the answers panelists must give.