Organize Your First Claude Code Project

Organize Your First Claude Code Project

Here's the single most important thing to understand about Claude Code: it works on whatever folder you're in.

When you open your terminal and run claude, it can see the files in your current folder. It can't see files in other folders (unless you point it there). That's it. That's the whole mental model.

This means: a folder is a project. If you want to keep your sales reports separate from your expense tracking, you put them in separate folders. Different folder, different project.

Why folders matter for Claude Code

Without folders, here's what happens:

1. You clean a CSV and the output lands in your home directory
2. You do another task and now there are 6 random files mixed together
3. You ask Claude to "read the CSV" and it picks the wrong one because there are three CSVs sitting there
4. Everything is confusing

With folders:

1. Sales files live in ~/sales-reports/
2. Expense files live in ~/expense-tracking/
3. Marketing briefs live in ~/marketing/
4. HR docs live in ~/hr-onboarding/
5. When you cd ~/sales-reports and run claude, it only sees sales files
6. No confusion

How to set up a Claude Code project folder

Let's make this concrete. Say you want a place for your sales work.

On Mac or Linux:

mkdir ~/sales-reports

On Windows (PowerShell):

mkdir ~\sales-reports

That's it. You now have a project folder. Move your sales files into it:

mv ~/Downloads/sales_march.csv ~/sales-reports/

How the working directory affects Claude Code

Before you run Claude Code, navigate to the folder:

cd ~/sales-reports

Now run claude. Everything Claude does — reading files, creating outputs — happens in this folder. Your instructions can be simpler because there's no ambiguity about which files you mean.

Instead of:

Read the file at ~/Downloads/sales_march_2026.csv and clean it up,
save the result to ~/Documents/cleaned_sales_march.csv

You just say:

Clean up sales_march.csv and save it as cleaned_march.csv

Claude knows where it is. It knows what files are there. Less typing, fewer mistakes.

Separate projects into separate folders

If you have multiple things you use Claude for, give each one its own folder:

~/sales-reports/        ← weekly sales CSVs and reports
~/expense-tracking/     ← bank exports and categorized expenses
~/meeting-notes/        ← raw notes and extracted action items

Don't be afraid to make new folders. They're free. The benefit is that when you cd into one and run Claude, it's immediately scoped to that project. No cross-contamination.

Organize input and output files

A simple pattern that works: put your raw files and your outputs in the same project folder. Name outputs so you can tell them apart from inputs:

~/sales-reports/
  sales_march_2026.csv          ← input (raw export)
  sales_april_2026.csv          ← input (raw export)
  cleaned_march.csv             ← output
  report_2026-03-09.md          ← output

As your project grows, you might want subfolders — /input/ and /output/ — but don't overthink it early on. Start simple. Organize more when it starts feeling messy.

The key takeaway

Whenever you start a new type of work with Claude Code:

1. Make a folder for it
2. Put your files in that folder
3. `cd` into that folder before running claude

That's the whole system. It sounds simple because it is — but it's the foundation everything else builds on.

Next up

Now that you have a project folder, let's teach Claude to remember how you like things done. In the next episode, you'll create a CLAUDE.md file — a set of saved preferences that Claude reads automatically every time you work in that folder.