Getting started

Budget app with CSV import: how to prepare the file so it loads cleanly

Using a budget app with CSV import? Prepare your file with the right columns, real dates and clean numbers, then check the totals before saving.

Editorial illustration for Budget app with CSV import: how to prepare the file so it loads cleanly

A CSV import goes smoothly when the file has one header row, one transaction per row, real dates, plain numbers and a clear split between money out and money in. Most failed imports come from a messy file, not a faulty app. Spend five minutes cleaning the CSV in Excel or Google Sheets and the import usually becomes a non-event.

What a clean CSV looks like

DateDescriptionDebitCredit
03/04/2026Grocery store1840.500
04/04/2026Salary052000
06/04/2026Cab ride3150
09/04/2026Refund, online order0699

Some banks give one Amount column and a Dr or Cr marker instead. Either layout is fine as long as it is consistent. Check what your bank's file looks like and what the app asks for before you edit anything.

The seven problems that break imports

  1. Extra lines above the header, such as the account holder's name or a statement period. Delete them so row 1 is the header.
  2. Totals or balance summaries at the bottom. Delete them; they are not transactions.
  3. Dates stored as text, or a mix of day-first and month-first. Make one format throughout.
  4. Amounts with commas or a rupee sign stored as text. Convert to plain numbers.
  5. Merged cells, which flatten unpredictably when saved as CSV.
  6. Odd characters in merchant names. Save as CSV UTF-8 to keep them.
  7. Blank rows in the middle. Delete them.

How do you fix text dates and numbers with formulas?

Suppose column A holds dates as text like 03/04/2026 in day-month-year order. In a helper column, enter this formula and fill down:

=DATE(RIGHT(A2,4), MID(A2,4,2), LEFT(A2,2))

It rebuilds a real date from the year, month and day parts. Then copy the helper column and paste it back as values. For an amount stored as text like 1,840.50, use:

=VALUE(SUBSTITUTE(C2, ",", ""))

SUBSTITUTE removes the comma and VALUE turns the result into a number. Both functions exist in Excel and Google Sheets.

A worked example: proving the file before you import

Hypothetical April file: 38 debit rows adding to ₹46,820 and 4 credit rows adding to ₹58,200. Before importing, enter =SUM(C2:C39) under the debit column and =SUM(D2:D43) under the credit column, adjusting the ranges to your file. Compare them with the totals on the bank's own statement. If they match, you know the file is complete. If the app later shows ₹46,320 of debits, the ₹500 gap points to one row lost in import, and you can search the file for an amount of 500.

What should you check after the import?

  • Row count: the number of rows imported equals the number in the file.
  • Account: every row sits against the right account. SpendixAI asks you to pick the account before importing.
  • Duplicates: an overlapping file should be flagged.
  • Categories: review the first twenty rows and correct them so later imports learn from you.

If something looks wrong, change one thing at a time and import again with a small slice of the file, say the first ten rows. Finding a bad row among ten is easy; finding it among four hundred is not. Keep an untouched copy of the bank's original download and do all your edits on a copy.

When to use a PDF instead

If your bank's CSV is awkward or unavailable, upload the PDF or an Excel file instead. They are accepted too, and the same review step applies. Keep the original download too; it is your proof if a row ever looks wrong.

SPENDIXAI JOURNAL

Useful ideas start with a clearer picture of your own spending.

Explore your reports