Getting started with Claude.
The chat window is where you type prompts. The left menubar runs down the side and holds navigation, tools and your profile menu.
- New chat, the plus icon near the top of the left menubar, starts a fresh conversation.
- Chats opens your conversation history.
- Your initial at the bottom-left opens the profile menu, including Settings (section 02).
To begin, click into the message box, type a request in plain English, and send. The conversation is saved in Chats.
To work from your own material, click the plus (+) button at the left of the message box and attach a file: a PDF, a document, a spreadsheet or an image. The file becomes part of the conversation's context, and Claude draws on it alongside your request when it answers (more on context in section 06).
Try one of these first
- "Search the web for [a company or topic]. Give me a half-page brief covering what it is, what has changed since [date] and three things worth watching. Cite the source for every recent factual claim. If search is unavailable, tell me rather than guessing."
- "Help me write a short, friendly email to reschedule a meeting."
- "Summarize the file I just attached in a few plain bullet points."
Treat Claude's first answer as a draft. Reply in the same chat with specific feedback on length, tone, format or audience.
Most controls sit around the message box.
Around the message box
- Plus menu (1): add files, connectors and other context.
- Model and effort menu (2): choose the model, then open Effort from the same menu.
- Dictation microphone (3): speak a prompt instead of typing, where available.
- Voice mode (4): have a spoken back-and-forth with Claude, where available.
- Prompt shortcuts (5): quick starts such as Learn, Write, From Drive, From Calendar and From Gmail, where available on your account.
Treat the chat box like a work brief. Include the source material, intended reader and a clear instruction.
Most account-level settings sit behind your profile menu.
- Click the circle with your initial at the bottom-left.
- Click Settings.
- Settings opens as an overlay on the General tab. The section that matters most is Instructions for Claude.
What's yours, what's your admin's
On Claude Enterprise, some switches belong to your IT admin, including which models, connectors and workspace features appear. If an option is grayed out or missing, that is usually policy, not a fault.
Tell Claude how you work
The most useful setting in Claude. In Settings → General, find Instructions for Claude, the box where Claude asks what it should consider in responses. What you write here applies to every chat.
- Open Settings.
- On the General tab, below your profile details, find Instructions for Claude.
- Click into the instructions box.
- Write a few lines on tone, length and format.
- Save your changes if prompted. Every new chat from here follows these.
Lines worth adding
- "Be concise. Lead with the answer, then the reasoning."
- "Write for a finance-literate reader. No need to define basic terms."
- "No Oxford commas. Spell out an acronym the first time you use it."
- "When you give a number or factual claim, cite or link the source. If you cannot verify it, label it unverified."
Keep out of the box
- Keep confidential or client-specific information out of this setting because it applies to every chat.
- One-off instructions for a single task. Say those in the chat itself.
- Long essays. If it runs past a short paragraph, it is doing too much.
A shortcut: let Claude interview you
Not sure what to write? Have Claude build it. Paste this into a chat, answer a few questions, then copy the block it produces into the instructions box above.
Paste this
Interview me to build my Instructions for Claude. Ask about the work I do, how I work, my preferences and pet peeves. Ask one question at a time. When you have enough, write it up as a short instructions block I can paste into Settings.
Review your privacy and memory settings
While you are in Settings, take two minutes before your first real conversation. Review two controls and learn one privacy habit.
- In Settings, open Privacy. Find Help improve our AI models, which controls whether eligible chats and coding sessions can be used to improve Claude, and choose deliberately.
- Open Capabilities, find Memory, then click View and manage memory. Claude curates your memory file from your chat history. Review it periodically and edit if necessary.
- Incognito chat is not saved to your history or memory. In Claude Enterprise, it is still logged. To activate, click on the ghost icon at the top right corner of the Claude window.
Claude comes in several model families and each is suited to different kinds of work. Switch between them in the model menu next to the send button. Reasoning models such as Fable and Opus are very powerful but run slower and use more tokens, so match the model to the task. Which models appear depends on your plan, region, feature rollout and, on Claude Enterprise, administrator settings.
| Model | Reach for it when | What it is good at |
|---|---|---|
| Fable 5 | A large assignment has several linked steps. | Planning and carrying a large assignment from start to finish. |
| Opus 4.8 | The work genuinely needs more time and judgment. | Detailed analysis and difficult judgment calls. |
| Sonnet 5 | You are unsure where to start. | Most writing, analysis, documents and spreadsheets. |
| Haiku 4.5 | You need a quick answer. | Summaries, quick lookups and simple extraction. |
Start with Sonnet. Use Opus for analysis that needs deeper reasoning and Fable for large assignments with several linked steps.
Model names will change. Use the descriptions in your account and keep the same basic rule: Sonnet for most work, Haiku for speed, Opus for deeper analysis and Fable for the largest assignments.
On models that support it, the same menu includes Effort. The model determines the capability tier; effort controls how much reasoning it applies. Higher effort can produce a more thorough answer, but it takes longer and uses more tokens. If Effort is missing for a model, use the default behavior or choose a model that exposes effort settings.
| Level | When to use it |
|---|---|
| Low / Medium | Routine asks where speed matters. Uses fewer tokens. |
| High | The everyday balance of quality and speed for harder work. |
| Extra | Deeper reasoning for long-running, complex work, without Max's full cost. |
| Max | The hardest, get-it-right work. Deepest reasoning and slower. |
The menu marks each model's recommended level with a Default badge. Trust the badge until you have a reason not to. Labels and availability can change; the options shown in your model menu are the source of truth for your account.
For everyday work the default is usually right. Raise it for complex tasks: document analysis, multi-step calculations, anything where being right beats being fast.
Effort versus thinking
At the bottom of the same effort menu sits a Thinking toggle on models that support it. Effort controls how much reasoning Claude applies to each reply; thinking determines whether Claude presents an expandable summary of its reasoning approach before answering. Turn Thinking on when that added visibility is useful. The tradeoff is speed and token use on simple asks.
Give Claude the same brief you would give a capable colleague picking up the work cold: useful background and sources, the output, the reader, your preferences and what to avoid.
A prompt can cover six things: persona, context, output, audience, preferences and guardrails. Most tasks need only two or three.
| Cover | What it settles | Say something like |
|---|---|---|
| Persona | What role Claude should play. | "Act as an investment analyst with experience evaluating private-market transactions." |
| Context | What facts, source material or background Claude should use. | "Use only the attached CIM. Treat missing information as unknown." |
| Output | What should come back and in what form. | "Prepare a two-page first-pass investment memo by summarizing the attached CIM. Lead with the conclusion, then a metrics table, summary of the deal and key risks." |
| Audience | Who will read it and what they care about. | "Write for a CIO deciding whether to run a first pass." |
| Preferences | What tone, format and level of detail Claude should follow. | "Use direct, concise language. Use a table for metrics and short paragraphs for analysis." |
| Guardrails | What Claude should avoid, flag or ask about before drafting. | "Cite the page for every number. Flag unsupported claims and never invent a figure." |
Act as an investment analyst with experience evaluating private-market transactions.
Use only the attached CIM. Treat missing information as unknown.
Prepare a two-page first-pass investment memo by summarizing the attached CIM. Lead with the conclusion, then a metrics table, summary of the deal and key risks.
Write for a CIO deciding whether to run a first pass.
Use direct, concise language. Use a table for metrics and short paragraphs for analysis.
Cite the page for every number. Flag unsupported claims and never invent a figure. If material information is missing, ask up to three clarifying questions before starting. If the attachment is missing or unreadable, stop and tell me.
Put the lines together as one prompt.
Speak to Claude
When a request is easier to explain out loud, use the microphone or voice mode. Review the transcript before sending names, figures or technical terms.
Context is everything Claude can see when it answers: your message, the files you attach or paste, the chat so far and your instructions from section 02. Files, email and other material count only after you explicitly attach, paste or connect them, so the practical skill is choosing what Claude sees.
Claude answers better when it has the right source material and knows who the work is for. Put reusable background in Instructions or a project so Claude can use it in each new chat.
Without context
"Summarize this memo."
A competent summary written for no one in particular. Claude has to guess who is reading and why.
With context
"Act as an investment analyst with experience evaluating private-market transactions. Use only the attached CIM. Treat missing information as unknown. Prepare a two-page first-pass investment memo by summarizing the attached CIM. Lead with the conclusion, then a metrics table, summary of the deal and key risks. Write for a CIO deciding whether to run a first pass. Use direct, concise language. Use a table for metrics and short paragraphs for analysis. Cite the page for every number. Flag unsupported claims and never invent a figure. If material information is missing, ask up to three clarifying questions before starting. If the attachment is missing or unreadable, stop and tell me."
The brief made the difference; the memo, model and effort setting stayed the same.
When a chat runs long
Very long conversations consume more of the chat's context window and can lose the thread, so decide: if the answer is nearly there, keep going; if it is not, start over. To carry the work forward, ask Claude for a handoff prompt and paste it into the new chat. If you find yourself pasting the same background into every new chat, that is the cue for the next section.
A project keeps the same files and instructions available across its chats. Project instructions replace your global instructions inside that project.
- In the left menubar, click Projects, then New project. Give it a name and briefly describe what you are trying to achieve.
- Add project knowledge: templates, past examples, a style note. Loaded once and available to every chat inside the project.
- Set the project instructions: your preferences and instructions for this Project only.
- Start chatting. Every chat can use the project knowledge and instructions, so you skip the repeated setup.
Put reusable files in project knowledge
A file attached to one chat stays with that chat. If you keep uploading the same document, add it to the project once.
Try it on a real document
- Start a fresh chat and attach a non-sensitive document you understand well.
- Use the prompt levers the task needs: persona, context, output, audience, preferences and guardrails.
- Review the first answer and give one round of specific feedback.
- Check every name, number, date and citation against the source.
- If the work will recur, create a project and add the reusable files and instructions.
If you already use ChatGPT, this table shows the closest Claude equivalent.
| In ChatGPT | In Claude | Section |
|---|---|---|
| Custom instructions | Instructions for Claude, in Settings. | 02 |
| Memory | Memory, reviewed and edited in Settings. | 02 |
| Temporary chat | Incognito chat: not saved to history or memory, and still subject to data retention policies. | 02 |
| Projects | Projects, with knowledge files and standing instructions. | 07 |
| A custom GPT | The closest match is a project for persistent knowledge and instructions. A skill is a reusable set of instructions for a specific procedure that Claude can activate when needed. | 07 |
| Canvas (now writing and code blocks) | Artifacts, opening in a side panel you can edit in place. | 01 |
| Deep research | Research, in the plus menu by the message box. | 01 |
| The model picker | The model menu next to send: Haiku, Sonnet, Opus and Fable, sized by task. | 03 |
One habit before anything leaves your hands
A polished answer can still be wrong. Before anything leaves your hands, check every number, name, date and citation against the source. That is part of using Claude professionally.
A half hour of setup will save time in every chat that follows.