Statement guide · Wallet Analysis
How to analyze credit card spending without linking your bank account
You don’t need to connect a bank account to get a clearer picture of your credit card spending. Download transaction files from your card accounts, review them locally, and use the results to ask better questions about where your money goes.
1. Start with the right statement files
Download a transaction export from each credit card account. CSV is usually the simplest starting point for Wallet Analysis; Excel (.xlsx or .xls) and compatible text-based PDFs are also accepted. Scanned PDFs are not supported, and a PDF with selectable text can still have a layout the reader cannot recognize.
For spreadsheet files, look for a date, merchant or description, and amount—or separate debit and credit columns. Name files clearly, such as Citi.csv or Chase.csv, and use the last two years if you want a year-over-year comparison. Compatibility depends on the export layout, so review the result rather than assuming every file from a particular bank will work.
2. Review purchases, payments, and credits
In Settings, import the files and check the assigned card names and statement signs. Different statements use different conventions for positive and negative amounts. A purchase should be spending; paying your card bill should not become a second purchase.
Check a familiar purchase and a card payment in Transactions before relying on totals. Wallet Analysis separates payments from spending. Its spending summaries also exclude nonpositive amounts, so they are not a net-spending calculation after refunds. Don’t expect the spending total to equal your statement balance, which can include earlier balances and payments.
3. Check categories before drawing conclusions
Automatic categories are a starting point, not a perfect description of every purchase. A large store may sell groceries, clothing, and household items in the same transaction. Search for a merchant or review uncategorized charges, then correct an individual transaction or apply a category to that merchant.
Begin with your largest categories and a few large charges. That makes it easier to spot a misplaced purchase that could otherwise make one category appear unusually high.
4. Compare like-for-like periods
A partial month is not directly comparable to a complete month. Check the dates shown in Overview and the partial-month note before interpreting a drop. For year-over-year comparisons, make sure you have statements covering the same period in both years.
For example, an illustrative $1,200 travel purchase can make one month look expensive without indicating a recurring habit. Open the month’s transactions to see whether a change comes from a one-off charge, several smaller purchases, or incomplete imports.
5. Review recurring charges, not just totals
Use Recurring to examine possible subscriptions and bills, then check the underlying transactions. A repeating merchant is not necessarily a subscription you should cancel, and detection can miss charges. Verify the service and billing history yourself before taking action.
Pick one useful question: “Am I still using this service?” or “What explains this category’s increase?” The goal is a clearer understanding of your statement, not an automatic judgment about every purchase.
6. Keep a backup you control
Wallet Analysis reads your statements in your browser and keeps your cards, transactions, and edits in local browser storage. It does not upload financial data or use AI for spending analysis. Typed feedback and anonymous visitor counting are separate from financial records.
Local storage is not a cross-device backup. In Settings, use Download my data to save a backup file, and keep it somewhere safe: the file contains your financial records. Load from file can restore it in another browser after confirmation. Clearing browser storage can remove your local data.
Your first review can be simple
Import one compatible file, check a purchase and payment, review the largest category, then save a backup. Add more statements once you are confident the data looks right.
Import your statements