Your first half hour with Claude should end with one useful piece of writing you can edit and send. Open the chat, do one real task from your own week, and check the result before you share it.
Imagine you have just made an account. You do not want a tour of every feature. You want one real result, like a clean summary of a messy email thread or a first draft of a note to your team. This post walks you through exactly that, in the regular chat, with nothing to install.
It is the third post in Learn Claude from scratch, after what Claude is and how the plans compare. For the basic vocabulary, see our plain-English guide to AI chat tools.
Before you open the tab
Claude is the AI assistant made by Anthropic. You reach the everyday chat at claude.ai. Anthropic also ships a desktop app for Mac and Windows and mobile apps for iOS and Android, and the same account works in all of them.
Two rules trip people up late, and they usually blame “the model” when the real cause is one of these:
- Age: you must be at least 18 to use the services.
- Location: you need to be in a supported location. If login fails or features look missing, check Anthropic’s supported locations article before you rewrite your prompt for the fifth time.
Pick one device for this session. The web on a laptop is the easiest place to learn, because the layout shows the model controls and file attach options without a tiny keyboard getting in the way. Your phone is fine for follow-up edits and quick capture. The desktop app is useful when you want a dedicated window. You do not need all three on day one.
Choose one job for the half hour
Vague goals create vague chats. Before you type anything into Claude, write a one-line job on paper or a sticky note. Here are four examples:
- “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, stop and use a redacted sample or made-up numbers instead. A later post in this series covers what is safe to paste in full. For now, the test is simple: if you would not put it on a slide shared with the whole company, do not paste it into a free-form chat on day one.
Where Claude lives: three apps, one account
Think of the web, desktop, and mobile apps as three entrances to the same account, not three products with three personalities. Features still vary by plan and by how new a feature is on each device.
| Where | How you open it | Best day-one use | Watch-outs |
|---|---|---|---|
| Web | claude.ai in a browser | First real task, file upload, model and effort controls, Projects later | Browser extensions and password managers can clutter the screen; use a clean browser profile if something looks wrong |
| Desktop | Download from Claude’s download page for Mac or Windows | A dedicated window with less tab noise, and later Cowork-style work on paid plans | Skip the desktop-only features until chat feels boring and useful |
| Mobile | Claude app for iOS or Android | Capture notes, continue a thread, make short rewrites while you commute | Harder for long PDFs and careful editing; the model selector sits at the top of the screen |
Sign in with the same account everywhere. Chats generally follow the account, not the laptop, which is why you can start on the web and finish a polish pass on your phone later.
What you are not installing today
Claude’s product list has grown to include chat, Code, Cowork, Design, Science, browser connectors, and more. Each deserves its own post. For this half hour, treat anything that wants a terminal, the text window where you type commands instead of clicking, access to a folder on your computer, or an API (the connection programs use to send requests to a service) key (a private password that lets software talk to Claude) as out of scope. If the screen offers a shiny new product, close it and stay in ordinary chat.
The first 30 minutes, timed
Use a kitchen timer if you like. The point is not to rush. The point is to finish one full loop: you give input, Claude drafts, you edit, and you end up with a usable file.
Minutes 0 to 5: land and orient
- Open claude.ai and sign in (or create the account if you have not).
- Confirm you can see a blank chat and a text box.
- Find the model name. On web and desktop it usually sits near the text input. On mobile it sits toward the top. Click it once so you know the selector exists. Leave the default unless you already know you need a different model for a paid workflow. A workflow is the series of steps a task goes through.
- If you see effort or thinking controls, leave them at their defaults for this first pass. You can turn them up later when a task is actually hard.
- Glance at the plan indicator (Free or paid) so usage limits do not surprise you in the middle of a draft.
You do not need a tour of every menu. You only need to know which model is answering, and that you can raise the effort or thinking setting when a job deserves it.
Minutes 5 to 12: load context the honest way
Gather the raw material for your one job. Short and real works better than long and “complete.” Good choices include:
- 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 app accepts images)
On the web, use the + button next to the text box to attach files or photos when that is faster than pasting a wall of text. Claude commonly accepts PDF files, Word documents (DOCX, the usual Word file), spreadsheets saved as CSV, a plain text file where each line is a row and commas separate the columns, plain text (TXT), and web pages (HTML, the format a browser reads), among others. It accepts common image types, including JPEG and PNG photos, the usual photo files, animated GIF images, and WebP. Excel spreadsheets (XLSX, the normal Excel file) may need code execution and file creation turned on in settings. Limits change over time, so if an upload fails, shrink the file or paste a slice of it.
If you are moving from another assistant and care about long-term preferences, Free, Pro, and Max users can import memory from other AI providers on web and desktop. That is not a full chat history import. Anthropic’s own get-started help page says plainly that conversation history from another provider does not come over wholesale. Memory import is the supported path, and it is still experimental. You can ignore it for this half hour, because a good first result does not depend on imported preferences.
Minutes 12 to 22: one strong first prompt
Paste a structured request instead of a vague wish. This template 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, then read the whole answer once without editing anything. After that, do a second pass as a human editor:
- Delete anything you would not defend in a meeting.
- Replace vague verbs (“optimize,” “enhance”) 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 work better than one giant “make it better” message, because Claude can only fix what you point at.
Minutes 22 to 30: export and plant the next habit
Copy the final draft into the place where it will actually live, such as an email, a document, a ticket, a slide outline, or a note in your training system. Then title the Claude chat with the name of the deliverable and the date so you can find it later. If the product offers renaming, use it, because an untitled “New chat” is how useful work disappears.
You can add an optional five-minute phone check. Open the mobile app, find the same chat if it synced, and try one small edit request such as “make the subject line less corporate.” You are proving that the account travels with you. You are not redoing the whole task on a phone.
A worked example: a stakeholder update from messy notes
Suppose your notes look like this (a 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 warehouseA weak first prompt would be this:
Write a project update about the pilot.That invites made-up timelines, fake “success,” and numbers that sound precise, because language models tend to sound exact even when they are guessing.
A stronger first prompt uses 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.After your edit, a solid draft should keep the roughly 12% ticket rise labeled as an operations estimate. It should say that returns data is still missing, make the ask two hours of data engineering time, and never claim the pilot “drove growth.” If Claude turns a risk into a virtue, put the risk back, because you own the email.
Controls worth knowing on day one
Model selector
The screen shows which model you are talking to. Click it to switch among the models your plan offers. For a first session the default is fine. Switching models in the middle of a task can change the voice and thoroughness of the answers, so if you do switch, note it down so you are not comparing apples to mangoes later.
Effort and thinking
Anthropic offers settings for how much effort the model puts into a response and whether extended thinking is on. Use more effort or thinking when the problem needs several steps of reasoning but no web lookup, such as a tricky rewrite with many constraints or a careful comparison of two options you already pasted. Use ordinary chat for straightforward drafting. Use web search or Research (a paid feature, covered in the next post in this series) when the answer depends on current outside sources. Matching the control to the job saves both time and usage.
Usage limits
Free plans use session-style limits that reset on a schedule. Anthropic documents a five-hour style reset for free usage, and other fairness limits are possible when demand is high. Paid plans raise the ceiling. When you hit a wall, the product will tell you, and that message does not mean you prompted wrong. Save your draft somewhere else and continue when the window opens, or upgrade if this is daily work. The earlier post on plans in this series helps you choose.
Starter prompt pack for the first week
Copy these into a note app and swap the bracketed parts. Keep the lines that tell Claude what to refuse, because they stop it from inventing things.
1. Meeting notes to action list
Turn these notes into:
- Decisions (only if clearly decided)
- Open questions
- Action items with owner if named, else "owner missing"
Do not invent owners or deadlines.
Notes:
[paste]2. Explain like I have ten minutes
Explain [topic] for a smart coworker who is new to it.
Use: plain words, one analogy, three pitfalls, a 5-item checklist.
No hype. Flag where experts disagree if you are unsure.3. Rewrite shorter
Rewrite the text below to [N] words.
Keep my meaning and numbers.
Remove filler and corporate padding.
Show a short "what I cut" list after the rewrite.
Text:
[paste]4. Study sheet from allowed material
Using only the attached / pasted material:
- 10 quiz questions (answer key after)
- Glossary of 8 terms in one sentence each
- What I should re-read if short on time
If the material does not support an answer, say "not in source."5. Plan a small project
I need a plan for [goal] with about [hours] of work.
Audience: [me / team].
Deliver: phases, risks, definition of done, what to skip.
Do not invent tools I did not mention.
Context:
[paste]Web versus mobile in practice
People ask which app is “best,” but that is the wrong question for week one. Match the app to the friction you have right now.
| Situation | Prefer | Why |
|---|---|---|
| First structured prompt with a file | Web or desktop | Keyboard, multi-line edit, easier attach |
| Polish a paragraph between meetings | Mobile | Low setup cost |
| Long PDF with charts | Web, larger screen | Reading and citing pages is less painful |
| Dictated notes, then cleanup | Mobile then web | Capture fast, structure later |
| Anything involving codebases or local folders | Not day-one chat | That is later product territory |
If mobile feels cramped, you are not failing. Long drafting simply works better with a real keyboard, so use the phone as a remote control for chats you started on the web.
Common first-session mistakes
| Mistake | What happens | Fix |
|---|---|---|
| No single job | Five partial chats, zero sendable output | Write a one line goal before you open Claude |
| Prompt as wish | Pretty fiction | Give a role, source, format, and refuse list |
| Paste secrets “just this once” | Risk you cannot undo | Redact or use a fake sample |
| Switch models every reply | Inconsistent drafts, burned usage | Stay put until the draft stabilizes |
| Trust numbers without a source line | You become the error | Check against notes or file |
| Chase every product tile | No first win | Chat only for 30 minutes |
| Leave chat untitled | You cannot find the good one | Rename with deliverable and date |
| Expect full history import from another AI | Frustration | Memory import only; rebuild key chats as needed |
What comes after this half hour
The next post in the series covers chats versus Projects, files that stick across conversations, and Research for multi-source digs with citations. Later posts go into everyday writing jobs, memory and privacy, light workplace connectors, and when Claude is the wrong tool. The tour of Claude’s other products (Code, Cowork, Design, Science) waits until ordinary chat is already useful.
If your work is analytics-heavy, keep the Practical AI habits close: assist, do not replace; verify SQL and numbers (SQL is the language for asking a database questions); and treat pasting as a decision about who gets to see your data. Claude on the web does not cancel those rules. It just gives you a fast drafting partner.
What a finished first session looks like
When the timer ends, you should be able to point at four things:
- A deliverable outside Claude (an email draft in your mail app, a doc in Drive, a ticket comment, or an outline in your notes).
- A named chat you can reopen tomorrow without scrolling past mystery titles.
- A short note of what the model invented or softened, so you know where you spent your editing time.
- A saved prompt template that worked better than your first try.
If all you have is a long chat and nothing exported, the session is incomplete. The product invites you to stay inside it, but your real work lives elsewhere, in the inbox, the board deck, the training system, and the ticket queue. Claude is a stop along the way, not the destination.
A second sign that the session worked is being able to say in one sentence what you asked for and what you still distrust. For example: “I asked for a status email from my notes, and I still need finance to confirm the ticket spike.” That sentence is worth more than a confident paragraph you cannot defend.
Questions people ask
Do I need a paid plan for this half hour?
No. Free is enough to learn the loop of prompt, edit, and export. You may hit usage limits sooner on Free, which is annoying but not a sign you failed. Save the draft outside Claude and continue later, or upgrade when the limit keeps blocking real weekly work. The earlier post on plans covers that choice, so do not reopen the decision mid-timer.
Should I turn on every tool in the + menu?
No. Day one is ordinary chat plus an optional file attach. Research, connectors, and product-specific modes can wait until you have a reason, since extra tools add extra ways to burn usage and attention. The next post covers Research and Project knowledge for when you are ready.
The answer sounds smart. How do I know it is wrong?
Check the parts that can embarrass you: numbers, names, dates, claims about what “the team decided,” and anything that looks like a citation. Language models are good at smooth certainty, so your notes are the source of truth for this exercise. If a claim is not in the source, it is either an assumption (which should be listed) or fiction (which should be deleted).
Can I import all my old ChatGPT threads?
Not as full conversation history. Anthropic documents memory import for Free, Pro, and Max (with related plan notes for Team in the memory articles), not wholesale history migration. Memory is a summary of lasting preferences and facts, and the import flow is experimental. For day one, rebuild the one chat that matters by pasting the source material again, which is faster than fighting an import you do not need yet.
Web or mobile if I only have twenty minutes?
Use the web if you have a keyboard, and mobile if you only have a phone and a short rewrite. Twenty minutes on mobile with a long PDF is a recipe for frustration. Match the app to the size of the task, not to the marketing screenshot of people smiling at their phones.
Practice for the next two days
- Repeat the 30-minute loop on a second real task with a different format (email to outline, or outline to study sheet).
- Install the mobile app if you have not, open one existing chat, and make a single constrained edit.
- Write a personal “never paste” list of three items (customer IDs, credentials, health data, or whatever applies to you).
- Save your best first prompt as a template in a note app before you forget which wording worked.
Summary checklist
- Use the web for the first serious task, keep mobile for capture and light edits, and treat desktop as optional on day one.
- Confirm age and supported location before blaming the product.
- Know where the model, effort, and thinking settings live. The defaults are fine for the first win.
- Pick one job, write a structured prompt, edit by hand, and export the result out of the chat.
- Skip Code, the API, and deep product installs until chat is already paying for itself.
- Memory import is not full history import, so ignore it if it slows the first deliverable.
Sources
Product behavior and limits change. Confirm details in Anthropic’s help center when something on screen does not match this guide.
- Anthropic Help Center, Get started with Claude: https://support.claude.com/en/articles/8114491-get-started-with-claude (access surfaces, age 18+, model selector location, effort and thinking, memory vs history import)
- Anthropic Help Center, Where can I access Claude (supported locations): https://support.claude.com/en/articles/8461763-where-can-i-access-claude
- Anthropic Help Center, Upload files to Claude: https://support.claude.com/en/articles/8241126-upload-files-to-claude (types, limits, project vs chat files)
- Anthropic Help Center, Import and export your memory from Claude: https://support.claude.com/en/articles/12123587-import-and-export-your-memory-from-claude
- Anthropic Help Center, How do usage and length limits work?: https://support.claude.com/en/articles/11647753-how-do-usage-and-length-limits-work
- Claude product entry: https://claude.ai/ and desktop download: https://claude.ai/download
- Analytics Made Simple, What are LLMs, ChatGPT, generative AI, and more: https://analyticsmadesimple.com/analytics/what-are-llms-chatgpt-generative-ai-and-more/
- Analytics Made Simple, Practical AI series: https://analyticsmadesimple.com/series/practical-ai/
- Analytics Made Simple, Learn: https://analyticsmadesimple.com/learn/
Keep going
Same lessons in your feed
Short diagrams, hooks, and weekly tutorials on Substack, Instagram, X, and Facebook.
