Better answers with structure and examples

You send: “Write something about our Q3 results.” ChatGPT returns three polished paragraphs, a fake 14% growth line, a soft strategy section, and a closing that could belong to any company on earth. You almost paste it into Slack. Then you notice you never said what Q3 was, who the reader is, or which four bullets are true. The model filled the silence with confident fog.

This is Part 2 of the ChatGPT everyday tutorial. Part 1 ran a Mon-Fri week of write, plan, and learn with one real output per day. Here we stay in Chat and fix the quality of those outputs: a structure stack (role, goal, constraints, example, format), weak vs strong prompts side by side, and a calm iteration loop when the first reply is wrong. If foundations or modes still blur, use Learn ChatGPT from scratch and ChatGPT product map first. Do not open Work or Codex for this skill. Structure is a Chat habit.

You do not need a secret prompt dialect. You need to stop leaving the model to guess the reader, the facts, the length, and what “done” looks like.

What you’ll learn

• A five-part structure stack you can reuse for most everyday Chat jobs
• How weak and strong prompts differ on the same task (with worked pairs)
• How to add one good example (or a counter-example) without writing a novel
• How to iterate when the answer is wrong: diagnose, patch, re-run, not “try harder”
• A short checklist before you hit Send on any important prompt
• Where this leads next: long docs and multi-file work in Part 3

The structure stack

Think of five slots. You do not always fill all five. Empty slots are where the model improvises, and improvisation is where invented numbers and generic tone sneak in.

Role is not cosplay. “Act as a pirate CEO” is entertainment. “Write as a careful analyst who will not invent numbers” is a constraint with a hat. Prefer job and reader over movie characters.

Goal should name a deliverable you could drop into a file. “Help me think about Q3” is a mood. “Produce a 200-word update for the ops channel from these bullets” is a goal.

Constraints are where most quality lives. Length, audience, refuse lists, facts-only rules, reading level, languages, and “mark unknowns as [UNKNOWN]” all belong here. If you remember one slot from this post, make it constraints.

Example can be positive (“match this tone”) or negative (“do not sound like this”). One short sample beats five adjectives like “professional, friendly, crisp.” Format is the shape: table, bullets, email, JSON-like fields, quiz with answers hidden, two columns. Shape saves you ten minutes of cleanup.

Weak prompt vs strong prompt

Same task, two prompts. Watch which slots are empty.

Pair 1: the Q3 update

Weak

Write something about our Q3 results.

Strong

Role: careful ops writer. I will edit and send under my name.
Goal: 200-word update for non-finance managers in Slack.
Constraints:
- Use ONLY these bullets. If a detail is missing, write [UNKNOWN], do not invent.
  • East region dashboard v2 live Mon 7/14
  • West still missing store #118 (data eng, ETA Wed)
  • Freeze new KPI requests until Fri
  • South pilot target next Tue if West feed is clean
- No growth percentages unless I provided them (I did not).
- No “excited,” “thrilled,” or “synergy.”
Format:
1) What changed (short paragraph)
2) Blocker (2 bullets max)
3) Ask (one sentence)
4) Two discussion questions
Example of tone I want: “East is live. West still has a feed gap on store #118. I need a freeze on new KPI asks until Friday so we can fix grain.”

The weak prompt leaves every slot empty. The strong prompt fills role, goal, constraints, example, and format. The model still might drift. You still edit. But you stopped asking for fog.

Pair 2: a meeting agenda

Weak

Make an agenda for our metrics meeting.

Strong

Goal: 30-minute metrics sync agenda for 6 people (ops + one data eng).
Constraints:
- Start at :00, hard stop :30.
- Must include 8 minutes on active store grain decision.
- Must include 5 minutes on store #118 only (not a full data platform review).
- No “round robin updates” longer than 6 minutes total.
- I will paste owner names; do not invent attendees.
Attendees and owns:
- Alex: grain proposal
- Priya: #118 status
- Sam: South pilot readiness
Format:
- Table: time | topic | owner | decision needed? (Y/N)
- End with a 3-bullet parking lot for topics we will not cover today
Refuse: strategy offsites, hiring, tools shopping.

Weak agendas become TED-style topic lists. Strong agendas become something you can run with a clock.

Pair 3: explain a concept

Weak

Explain primary keys.

Strong

Role: patient teammate teaching a new analyst in week one.
Goal: 12th-grade explanation of primary keys for retail ops tables.
Constraints:
- Assume they know what a spreadsheet row is.
- Use a stores table and a daily_sales table as examples.
- Do not start with academic database history.
- Flag what is a team decision vs a hard rule.
Format:
1) One-sentence definition
2) Tiny table example (3 stores)
3) What breaks if the key is wrong
4) 3 quiz questions (answers in a separate ANSWERS section)
Example of voice: short sentences, no “in the realm of data management.”

Same topic. Different chance you can teach it back without reopening Wikipedia.

Fill the slots without writing a novel

People skip structure because they think it takes twenty minutes. It takes ninety seconds once you have a pattern. Use a sticky note or a snippet expander if you repeat jobs.

Ninety-second template

Role: [job + reader]
Goal: [deliverable + length]
Constraints:
- Facts: [bullets or “none yet, ask me questions first”]
- Refuse: [invented numbers / legal advice / hype words / …]
- Other: [deadline voice, reading level, language]
Example: [2 sentences of good tone OR “avoid: …”]
Format: [headings / table / email / quiz]

If you do not have facts yet, say so. A strong early move is: “Ask me up to 5 clarifying questions before drafting. Do not draft until I answer.” That is still structure. Goal becomes “questions,” not “fake draft.”

When to skip a slot

Examples that pull weight

One good example beats a pile of adjectives. Two kinds work well.

Positive micro-example

Match this tone and density:
"East is live. West is blocked on store #118 until data eng finishes the feed fix (ETA Wed). I need leadership to freeze new KPI requests until Friday."

Negative micro-example

Do not sound like this:
"We are thrilled to announce a holistic journey toward cross-functional excellence in our Q3 landscape."
If you catch yourself writing that way, rewrite shorter and more concrete.

You can also give a format example: paste a skeleton with placeholders. “Fill this outline; keep headings; replace BRACKETS only.” That is example + format in one move.

Fill this skeleton. Keep headings. Replace only the bracketed parts. Do not add sections.

## Status
[3 sentences max]

## Blocker
- [owner]: [issue] ([ETA if known])

## Ask
[one sentence]

## Next 7 days
- [ ]
- [ ]

Iterate when the answer is wrong

First replies fail. That is normal. The failure mode is restarting from a brand new vague prompt, or arguing with the model in emotional paragraphs. Use a diagnose, patch, re-run loop.

Step 1: name the defect

Be specific. “Too long” is weak. “Cut to 200 words and remove the strategy paragraph I never asked for” is strong. Common defect labels:

• Invented facts or numbers
• Wrong audience (too technical / too vague)
• Wrong length
• Wrong structure (essay instead of bullets)
• Wrong tone (hype, hedge, lecture)
• Missing ask or missing decision
• Scope creep (answered three extra questions)

Step 2: patch the prompt, not only the reply

You can edit the last answer (“delete paragraph 2”), and you should. Also add the missing constraint so the next turn does not reintroduce the bug. If it invented a metric, add: “You invented X. Remove all numbers I did not provide. List any [UNKNOWN] fields instead.”

Step 3: re-run with a tight instruction

Revise your previous draft with these patches only:
1) Delete every number not in my bullet list.
2) Cut to 200 words.
3) Keep the four-part format.
4) Replace hype words with plain verbs.
Do not add new sections. Show the full revised draft.

Step 4: know when to start a new chat

If the thread is polluted with three bad strategies and a pile of invented metrics, a clean chat with the strong prompt is faster than ten patches. Copy your facts and constraints into a new thread. Leave the mess behind. Part 1’s “one chat per deliverable” still applies.

Worked iteration story

Alex pastes the strong Q3 prompt. Reply looks good, but invents “NPS at 72” and adds a “strategic outlook” section. Alex’s patch:

Defects:
- You invented NPS 72. Remove it.
- You added a strategic outlook section. Delete it.
- Blocker section has 5 bullets; max 2.

Revise the full draft. Same facts only. 200 words. Four parts + two questions.

Second reply is clean enough for a ten-minute human edit. That is success. Not perfection on turn one.

Pre-send checklist

Run this before important prompts. Ten seconds. Saves thirty minutes of cleanup.

Rebuild two weak prompts from Part 1

If you kept a weak-prompt graveyard from last week, rebuild two of them now. If not, use these stand-ins.

Graveyard item A

Weak: “Help me plan my week.”

Strong rebuild: role (realistic planner), goal (one-week plan with three outcomes), constraints (18 focused hours, existing meetings, no work past 6pm), example (one good “done means” line), format (outcomes, day blocks, parking lot, refuse list, Friday review). That is Tuesday’s prompt from Part 1, which was already structured on purpose. Steal that shape for any planning job.

Graveyard item B

Weak: “Make this email better” + paste.

Role: editor for my workplace voice (direct, short, no hype).
Goal: rewrite the email below so I can send it after a quick human pass.
Constraints:
- Keep every commitment, date, and name exactly.
- Cut about 40% length.
- One clear ask in the final paragraph.
- No new offers or apologies I did not write.
- If something is ambiguous, mark [CLARIFY] instead of guessing.
Format: subject line option + body.
Example of bad tone to avoid: "Just circling back to touch base when you have a moment."
Email:
[PASTE]

“Make it better” is not a constraint. Better how, for whom, with what risk if you change a commitment? Spell it.

Common structure mistakes

What structure does not fix

Structure will not make ChatGPT a reliable source of live market stats, legal advice, or your warehouse truth. It will not replace reading the PDF you care about. It will not make Work or Codex unnecessary for agentic multi-step jobs; those are different modes on the product map. Structure also will not fix a bad paste policy. A perfect prompt with a full payroll CSV is still a bad idea on the wrong account.

If you need tools that act across many files with approvals, that is Work territory later. If you need repo edits, that is Codex later. If you need the same recipe every day for a team, that might become a Custom GPT after you know the recipe works in plain Chat. This part makes the recipe legible.

How this fits the AMS path

You are in the ChatGPT everyday tutorial. Part 1 built a useful week. Part 2 made prompts inspectable. Part 3 will cover long docs and multi-file work (chunking, indexing what is in a file, verifying quotes). Foundations remain Learn ChatGPT from scratch. Mode and product orientation: ChatGPT product map. Broader hub: Learn. Sibling habits on Anthropic’s stack: Learn Claude from scratch.

Practice

1. Take one real task for tomorrow. Write the weak one-liner you would have typed six months ago.
2. Rebuild it with the five-slot template. Time yourself. Aim under three minutes.
3. Run it in Chat. Label defects on the first reply with the defect list above.
4. Patch once. If still messy, new chat with the cleaned strong prompt only.
5. Save the strong prompt in a notes file titled with the job type (prompt-ops-update, prompt-agenda-30m). Reuse next time.

Quick recap

• Stack: role, goal, constraints, example, format.
• Constraints and facts do most of the quality work.
• Weak prompts leave slots empty; strong prompts fill them.
• One micro-example beats a pile of tone adjectives.
• When wrong: name the defect, patch, re-run; new chat if the thread is polluted.
• Stay in Chat. Edit under your name. Next: long docs and multi-file work.

Sources

Prompt patterns here are teaching practice for everyday Chat, not vendor secrets. Product UIs and help articles change; re-check official pages when you train a team.

• ChatGPT app (Chat surface used throughout this tutorial)
• OpenAI: ChatGPT product home (product overview)
• OpenAI Help Center (feature and plan articles; search for current prompting and data guidance)
• OpenAI Platform: Prompt engineering guide (builder-oriented techniques; many principles still help everyday Chat even if you never use the API)
• OpenAI: Usage policies (allowed use boundaries)
• Analytics Made Simple: Learn ChatGPT from scratch (foundation series)
• Analytics Made Simple: ChatGPT product map (modes and product orientation)
• Analytics Made Simple: Learn (learning paths hub)

Merch on Etsy

Open store →

Imbalanced Classes Mouse Pad, Data Science Mouse Pad, Analytics Mouse Pad, Imbalanced Mouse Pad View on Etsy → Y Hat Mouse Pad, Data Science Mouse Pad, Analytics Mouse Pad, Y Pred Mouse Pad View on Etsy → You're On Mute White Ceramic Coffee Mug, Data Science Mug, Data Mug, Analytics Mug, Statistics Mug, Programming Mug, Gift Mug, Coffee Mug View on Etsy → Mute Mouse Pad, Data Science Mouse Pad, Analytics Mouse Pad, You're On Mute View on Etsy →