Expense Categorization

Bank export to categorized spreadsheet

The problem

You manually categorize hundreds of transactions every month from your bank export.

The outcome

A categorized spreadsheet with totals per category, flagged anomalies, and month-over-month trends.

What you need

Claude Code installed
A bank/credit card CSV export (columns: date, description, amount)

Step by step

Create your project folder

bash

mkdir ~/expense-categorization
cd ~/expense-categorization

Export your transactions

Download a CSV from your bank or credit card. Most banks have an "Export" or "Download" option in transaction history. Drop it in the folder.

Create the CLAUDE.md file

Create the CLAUDE.md file with the instructions below, or download it from the Downloads section.

Run it

bash

cd ~/expense-categorization
claude

Then tell Claude what to do:

Categorize all the expenses in the CSV file.

Review categories

Check the output. If Claude miscategorized something, mention it — "Uber should be Travel, not Food" — and it will learn for next time if you add that rule to your CLAUDE.md.

The CLAUDE.md file

Copy this into a file called CLAUDE.md in your project folder.

CLAUDE.md

# Expense Categorization

Read the CSV file in this folder (bank/credit card export).

## Categories
Categorize each transaction into one of these:
- Payroll
- Software & Subscriptions
- Travel
- Food & Meals
- Office Supplies
- Marketing & Advertising
- Professional Services
- Utilities
- Insurance
- Other

## Rules
- Amazon purchases: categorize as Office Supplies unless description mentions a specific category
- Uber/Lyft: Travel
- Restaurant names: Food & Meals
- Monthly recurring charges: Software & Subscriptions

## Output
Create two files:

### categorized_expenses.csv
The original data with an added "category" column.

### expense_summary.md
A markdown report with:
- Total spend by category (table, sorted highest to lowest)
- Month-over-month comparison if multiple months present
- Top 10 largest individual expenses
- Any flagged items (transactions over $500, or items categorized as "Other")

Format currency with $ and commas.

Customization ideas

Add your own category rules for vendors you use frequently
Split by team member if the CSV has a cardholder column
Add tax-deductible flags per category
Generate a summary formatted for your accountant

Related automations

Intermediate4 hrs/quarter

Customer Feedback Analysis

Survey responses to themes and insights

Hundreds of survey responses sitting in a spreadsheet. You skim them, miss patterns.

productcustomer successmarketing

Beginner45 min/file

CSV Cleanup & Standardization

Messy data to clean, importable format

You get CSVs with inconsistent formatting, duplicates, and missing fields.

operationssalesmarketing