Web and mobile: first 30 minutes

You already have a ChatGPT account, or you are about to make one. The first half hour is where people either leave with a usable draft, or bounce between app stores, plan banners, and a half-written prompt that never becomes something you can send.

This is Part 3 of Learn ChatGPT from scratch. Part 1 covered what ChatGPT is in plain English. Part 2 covered Free, Plus, Pro, Business, and Enterprise without the brochure voice. Here we stay off desktop modes, Codex installs, Custom GPTs, and API keys. The goal is boring on purpose: open ChatGPT on web or phone, run one real work task with a safe file or paste, rewrite until it is yours, and leave with a result you can edit and ship.

If you need LLM vocabulary first, read What are LLMs, ChatGPT, generative AI, and more. For vendor-neutral habits (context, verification, paste safety), the Practical AI series is the companion track. If you live in Anthropic’s stack the way this track lives in OpenAI’s, the sibling path is Learn Claude from scratch. Paths also sit on Learn.

What you will leave with

• A map of ChatGPT on web, iOS, and Android (and what each is good for on day one)
• Where to land: chatgpt.com, official mobile apps, and when to ignore the desktop download for now
• A timed first-session path: account, real question, safe file, rewrite
• A starter prompt pack you can paste without inventing a second career in prompt craft
• What to skip today so you do not derail into Work, Codex, connectors, or plan upgrades mid-draft

Before you open the tab

ChatGPT is OpenAI’s assistant product. The everyday chat surface for most people is the web app at chatgpt.com. OpenAI also ships official apps for iOS and Android, and a desktop app for Mac and Windows. Same account can move across those surfaces when you sign in with the same login.

For this half hour, pick one primary device. Web on a laptop is the easiest teaching surface: full keyboard, easy file attach, easy copy into email or docs. Mobile is fine for capture and short rewrites once the main draft exists. Desktop is useful later when you care about Chat, Work, and Codex living in one window. That is Part 4. Day one does not require all three.

Choose one job for the half hour

Vague goals create vague chats. Write a one-line job on paper or in a sticky before you type anything into ChatGPT:

• “Turn my messy notes into a 400-word project update for stakeholders.”
• “Outline a 20-minute teach-back on weekly revenue for non-analysts.”
• “Rewrite this cold email so it is shorter and less salesy.”
• “Build a study checklist from this PDF chapter (text I am allowed to use).”

If the job needs confidential customer data, payroll, unreleased strategy, or anything you would not put on a shared slide, stop. Use a redacted sample or synthetic numbers. Paste discipline gets more room later in this series. For now: day one is for learning the loop, not for stress-testing your company’s data policy.

Where ChatGPT lives: web, mobile, desktop

Think of surfaces as doors into the same account, not three different products with three different personalities. Features still vary by plan, by device, and by how new a feature is on each surface.

Sign in with the same account on each surface. History generally follows the account, not the laptop. That is why starting on web and polishing on the phone works when sync behaves. If a chat is missing, refresh, confirm you used the same login method, and check you did not accidentally open a work workspace when you meant a personal one (Business and Enterprise can look similar until you notice the org name).

What you are not installing today

ChatGPT’s product map has grown: ordinary Chat, agentic Work, coding-focused Codex, Custom GPTs, connectors into mail and calendars, and more. Those deserve their own series parts. For this half hour, treat anything that wants a multi-step office agent, a local coding project, or a full connector grant as out of scope. If the UI offers a shiny new product door, close it and stay in ordinary chat.

Same spirit as the Claude track: first half hour is about one deliverable, not a tour of every label OpenAI has shipped this year. If you already know Claude’s chat-first path, the muscle is the same. Only the labels change. See Learn Claude when you want the sibling product’s version of this lesson.

The first 30 minutes, timed

Use a kitchen timer if you like. The point is not speed-running. The point is finishing one loop: account ready → real question → safe context → draft → rewrite → usable file.

Minutes 0 to 5: account and land

1. Open chatgpt.com (or the official mobile app if that is your only device today).
2. Sign in or create the account. Prefer a login method you control long term (work SSO only if this is a work-approved account).
3. Confirm you can see a new chat and a text box. That is enough UI for day one.
4. Glance at plan indicators (Free versus paid). You do not need to upgrade mid-session. Limits exist; they are not a moral failure.
5. If a model name or mode switcher is visible near the composer, note it once. Leave the default for this pass unless your job already requires a specific paid feature.

You do not need a tour of every menu. You need three facts: you are signed in, you can type, and you know roughly which plan box you are in so rate limits do not feel like the model “broke.”

Minutes 5 to 12: one real question plus safe context

Gather the raw material for your one job. Prefer short and real over long and “complete.”

• Bullet notes from a meeting
• A draft email you hate
• A public-safe PDF or doc you are allowed to use
• A screenshot of a slide outline (if the surface accepts images)

On web, use the attach control near the composer when that is faster than pasting a wall of text. Common file types for everyday work include PDF, DOCX, CSV, TXT, and common image formats. Limits and supported types change; if an upload fails, shrink the file or paste a slice. On mobile, prefer short notes or one photo of a whiteboard you own rather than a 40-page PDF you will never skim carefully on a phone.

Safe file rule for day one: if the file has customer names, account numbers, unreleased revenue, passwords, or anything covered by your company’s AI policy, do not upload it on a personal Free account. Redact, invent a toy version, or wait until you know which workspace (personal vs company) is approved. The skill you are practicing is the workflow, not the thrill of pasting the real spreadsheet.

Minutes 12 to 22: draft, then rewrite

Paste a structured ask, not a vibe. Here is a template that works for almost any first deliverable:

Role: You are helping me draft for human review, not publishing for me.

Context:
- Audience: [who will read this]
- Goal: [decision or action I want]
- Tone: [plain, short, no hype]
- Constraints: [word limit, no invented stats, flag unknowns]

Source material:
[paste notes or describe the attached file]

Task:
1) Draft the deliverable in the format I asked for.
2) List assumptions you made in bullets.
3) List anything you could not verify from the source.

Refuse:
- Do not invent numbers, quotes, or sources.
- Do not claim the draft is final.

Run it. Read the whole answer once without editing. Then do a second pass as a human editor:

• Delete anything you would not defend in a meeting.
• Replace vague verbs (“optimize,” “enhance,” “leverage”) with the real action.
• Check every number against your notes or file.
• Ask a follow-up only for the broken pieces: “Shorten the second section to 80 words and keep my three risks.”

Two or three tight follow-ups beat one giant “make it better” message. The rewrite step is the product. The first draft is just clay.

Minutes 22 to 30: export and plant the next habit

Copy the final draft into the place it actually lives: email, doc, ticket, slide outline, LMS note. Title the ChatGPT chat with the deliverable name and date so you can find it later. If the product offers renaming, use it. “New chat” is how useful work disappears.

Optional five-minute mobile check: open the iOS or Android app, find the same chat if it synced, and try one small edit request (“make the subject line less corporate”). You are proving the account travels. You are not redoing the whole task on a phone.

Worked example: stakeholder update from messy notes

Suppose your notes look like this (toy example, not a real company dump):

- Q2 pilot: 3 stores, East region
- Dashboard v1 live Tuesday; still missing returns
- Support tickets up ~12% week over week (ops estimate, not finance)
- Next: freeze metric defs Friday with analytics + ops
- Risk: store managers using spreadsheet shadow metrics
- Ask: 2 hours from data eng for returns field in warehouse

A weak first prompt:

Write a project update about the pilot.

That invites invented timelines, fake “success,” and numbers that sound precise because language models like precision theater.

A stronger first prompt using the template:

Role: Draft for human review. I will edit before sending.

Audience: VP Retail Ops and analytics lead.
Goal: Clear status, risks, and one ask. No celebration copy.
Tone: Plain. Short paragraphs. No buzzwords.
Constraints: 350-450 words. Do not invent metrics beyond my notes.
If a number is approximate, keep it approximate.

Source notes:
[paste the bullets above]

Task:
1) Draft an email body with subject line options (3).
2) Sections: Status, Risks, Next steps, Ask.
3) Assumptions list.
4) Unknowns list.

Refuse: no invented ROI, no fake quotes from store managers.

What a solid draft should feel like after your edit: the ~12% ticket rise stays labeled as an ops estimate, returns stay missing, the ask is two hours of data eng time, and nothing claims the pilot “drove growth.” If ChatGPT softens a risk into a virtue, put the risk back. You own the email.

Web vs mobile for everyday jobs

You do not need a perfect multi-device setup. You need a default and a backup.

A pattern that works for a lot of analysts and ops people: start the real draft on web after the meeting, then use mobile only for “shorten this” or “three subject lines” once the structure exists. That keeps the heavy context on the surface that can hold it.

Starter prompt pack (copy freely)

Use these as first messages. Replace bracketed fields. Keep the refuse lines when numbers matter.

Meeting notes to update

Turn these notes into a status email for [audience].
Length: [N] words. Sections: status, risks, next steps, ask.
Flag anything that looks like an invented metric.
Notes:
[paste]

Email rewrite

Rewrite this email. Keep my facts. Cut fluff. Tone: [plain / firm / friendly].
Offer 2 subject lines. List any claims that need a human fact-check.
Draft:
[paste]

Explain a concept simply

Explain [topic] for [audience who already knows X but not Y].
Use one concrete workplace example. No hype. End with a 5-bullet checklist
I can use Monday morning.

File-assisted summary

I attached [filename]. Summarize only what is in the file.
Structure: (1) plain summary, (2) open questions, (3) what I should verify
before I act. If the file does not support a claim, say so.

Common mistakes in the first session

What “good enough” looks like after 30 minutes

You are done when all of these are true:

• You can open ChatGPT on web or mobile without hunting for the official link or a random third-party app.
• You have one chat that produced a deliverable you would actually edit and send.
• You checked numbers and claims against your notes or file.
• The draft lives outside ChatGPT (doc, email, ticket), not only inside the chat thread.
• You know which shiny doors you deliberately skipped for later (desktop modes, Work, Codex, connectors).

If you only got a pretty first draft and never rewrote it, you are not done. The rewrite is where ChatGPT becomes a coworker instead of a slot machine.

FAQ

Do I need Plus for the first half hour?

Usually no. Free is enough to learn the account → question → safe file → rewrite loop. Paid plans raise limits and unlock more features. Part 2 of this series covered plan shape without hype. Upgrade when you hit real limits, not because a banner looked urgent.

Should I start on desktop?

Not required. Web teaches the same chat skill. Desktop matters more once you care about Chat, Work, and Codex in one app. That is the next part. If you already installed desktop, use Chat only for this half hour.

Is mobile “real” ChatGPT?

Yes. Official iOS and Android apps are first-class surfaces for many people. They are just a weaker place for long documents and careful multi-section edits. Use them for capture and short rewrites, then finish important work on a larger screen when you can.

How is this different from Learn Claude’s first half hour?

Same teaching idea, different vendor doors and labels. Claude’s first session lands on claude.ai; ChatGPT’s lands on chatgpt.com. The sibling path is Learn Claude from scratch if you need both stacks for work.

How to practice this week

1. Run the timed half hour once with a real, low-risk work artifact.
2. Save your best rewrite as a personal template (notes app is fine).
3. Repeat twice more this week with different job types: one rewrite, one explain, one file summary.
4. When ordinary chat feels boring, open Part 4 and meet the desktop app’s modes without abandoning the rewrite habit.

Quick recap

• Land on chatgpt.com or the official mobile apps; same account can travel later.
• Pick one job line before you type.
• Account → real question → safe file or paste → rewrite → export.
• Skip Work, Codex, and deep desktop setup until chat is useful on autopilot.
• Rename the chat; own every number before you send.

Next: Part 4, Desktop app: Chat, Work, and Codex in one place, maps the desktop modes so you stop treating every button like the same chat window. For the Anthropic parallel track, stay oriented with Learn Claude. More paths: Learn hub.

Sources

Product surfaces and download pages change. Re-check official pages when a button moves.

• ChatGPT web app (everyday chat entry at chatgpt.com)
• Download ChatGPT (desktop and mobile download hub)
• ChatGPT on the App Store (official iOS app)
• ChatGPT on Google Play (official Android app)
• OpenAI: ChatGPT for your most ambitious work (desktop Chat, Work, Codex announcement context; details deferred to Part 4)
• OpenAI Help Center (account, apps, and plan help articles)
• Analytics Made Simple: Learn Claude from scratch (sibling product path)
• Analytics Made Simple: Practical AI (vendor-neutral verification and workflow habits)
• Analytics Made Simple: Learn (learning hub)

More definitions in Key Terms →

Merch on Etsy

Open store →

Sigmoid Function Shirt View on Etsy → Imbalanced Classes Ceramic Coffee Mug View on Etsy → Mean Squared Error Shirt View on Etsy → You're On Mute Shirt View on Etsy →