Most failed CSV imports come from the same four things: extra lines above the header row, dates in a format the tool does not expect, amounts with commas or a debit/credit split, and merged or blank cells. Open the file in a spreadsheet, fix these, save as CSV again, and try once more.
Gig workers hit this more often because they upload several different files: a bank statement, a payout report from a platform, and sometimes a wallet export. Each has its own odd layout.
What should the header row look like?
- The first row should name the columns, such as Date, Description, Debit, Credit and Balance.
Many net-banking exports begin with the account number, the period and a few blank lines. Delete everything above the header row. Remove any total or summary line at the bottom. One header, then rows.
How do I fix the date format?
Dates look right on screen but may be saved as text. Suppose cell A2 holds 05/03/2026 as text, in day/month/year form. In Excel or Google Sheets, this formula builds a true date.
- =DATE(RIGHT(A2,4), MID(A2,4,2), LEFT(A2,2))
- Copy it down the column, then paste the results as values over the original column.
- Format the column as yyyy-mm-dd, which is unambiguous.
If a date has a two-digit year or uses dashes, the positions change, so look at one cell first. Check that 05/03 means 5 March, not 3 May.
How do I clean amounts?
An amount stored as text, like 1,250.00 with a comma, can be rejected. If it sits in D2, use =VALUE(SUBSTITUTE(D2,",","")). Where one column holds both debits and credits with a Dr or Cr marker, split them in two columns, or put debits in negative numbers. Keep one number per cell and remove currency symbols.
| Problem in the file | What it looks like | Fix |
|---|---|---|
| Notes above the header | Account name and period in rows 1 to 5 | Delete rows above the header |
| Text dates | Left-aligned 05/03/2026 | Use the DATE formula above |
| Commas in numbers | 1,250.00 as text | SUBSTITUTE then VALUE |
| Debit and credit mixed | One column with Dr and Cr | Split into two columns |
Which gig-worker files cause the most trouble?
A platform payout report may list each trip with a gross fare, a commission and a net amount. Import the net payout lines as income, not every trip, unless you want trip-level detail. Hand-copied files often have merged cells; unmerge them and fill each row fully.
Does the app help with strange layouts?
SpendixAI accepts bank statements as PDF, CSV or Excel and shows the rows for review before anything is saved, so a misread date or amount can be caught first. An Import failed notification and an import history tell you which file to look at. It cannot promise that every layout will read perfectly, which is why the clean-up above is worth the five minutes.
What do I do when one file has thousands of rows?
A year of trip-level data can be heavy. Filter the sheet to the columns you need, keep the original untouched, and import in monthly chunks. Smaller files fail less and tell you precisely which month has the problem. Sort by date first, because some platform exports list the newest rows first, and importing in order makes checking easier.
- Make a copy of the file before touching it.
- Delete columns you do not need, such as internal IDs.
- Split by month if the file is large.
- Check the first and last row after each fix.
Before you try again
- Header on row one, data from row two.
- Dates as true dates in one format.
- Amounts as plain numbers.
- No blank or merged cells in the data.
- Save as CSV with the original file kept untouched.
Useful ideas start with a clearer picture of your own spending.
Explore your reports


