When you import a Union Bank of India statement, the file format matters less than the checks afterwards. Download it from your own net banking or app, upload it against the right account, read the preview, and compare the number of debit and credit entries and their totals with what the statement says. If the counts match, you can save with confidence.
Which format works best: PDF, CSV or Excel?
| Format | Good for | Be careful of |
|---|---|---|
| CSV | Clean columns and quick checks | Dates may be written day first; a spreadsheet opening it first can rewrite them |
| Excel | Keeping the original column layout | Extra title rows or notes above the data |
| Official copy when nothing else is available | Wrapped narrations and page headers can be read as rows |
If the bank offers more than one, a spreadsheet-type file is usually the less error-prone choice. Do not open and re-save it in Excel first; that is when dates and long reference numbers can change.
How do I get the statement from Union Bank of India?
Look for the statement or e-statement option after you sign in; since the menus are redesigned now and then, go by the words rather than a position. Select the account if you hold more than one, set the period and download. Never share your login or a one-time code with anyone offering to fetch it for you; the bank does not ask for that. We have not seen every layout this bank produces, so the verification below is the safety net.
Does it matter which account I pick before uploading?
Yes. In SpendixAI each transaction keeps the account it was imported into. If you hold a savings account and a joint account with the same bank, importing both under one name makes the balances meaningless, so give each its own account and upload each statement separately.
How do I check the import matched the statement?
Count and add. Imagine a month in which the statement lists 38 debit entries totalling ₹29,460 and 6 credit entries totalling ₹52,000. Your preview should show:
- 38 outgoing rows adding to ₹29,460.
- 6 incoming rows adding to ₹52,000.
- A net movement of 52,000 − 29,460 = ₹22,540 for the month.
If the preview shows 37 debit rows, one is missing. If the total is right but the count is higher, a row may be a page subtotal or has been split in two. Work through the narrower difference first.
What if some rows look wrong after saving?
Edit them. Transactions can be corrected, and if a whole import was wrong you can delete the transactions and upload again; deleted entries go to a recycle bin for 30 days. Fixing the first month properly is worth it, because categorisation learns from your corrections and later months get easier.
What do I do with the original file afterwards?
Keep it. The statement is the record the bank stands behind, and you will want it if a total does not match later. Store it in a folder with a name that sorts, such as Union-2026-04, and keep one copy somewhere you trust.
Statements contain account numbers, balances and every payee, so treat the file with the same care as the account itself. On a shared computer or family tablet, delete the downloaded copy from the Downloads folder once it is imported and filed safely, and do not send it over chat apps just to move it between devices.
What next once the first month is in?
- Import the next month and confirm the opening balance equals the previous closing balance.
- Let the tracker categorise, then fix the lines it got wrong.
- Open the overview report and see whether the biggest category looks plausible.
If the file will not read at all, the Import failed notification and import history tell you which upload to look at. Try the other format before entering rows by hand.
Useful ideas start with a clearer picture of your own spending.
Explore your reports

