Getting started

Re-importing a corrected statement after a bad import

How to fix a wrong statement upload: check what the bad import added, remove those rows, correct the account or file, then re-import without doubling anything.

Editorial illustration for Re-importing a corrected statement after a bad import

To redo a bad import, first work out what the wrong upload added, remove those rows, fix the cause (usually the account chosen or the file used), and then import again. Doing the clean-up before the second upload is what prevents everything from showing up twice. Deleted transactions sit in a recycle bin for 30 days, so a careful clean-up can be undone if you remove too much.

Step 1: what exactly went wrong?

Different mistakes need different repairs, so name yours first.

  • Wrong account picked: the rows are right but sit under the wrong bank account or card.
  • Wrong file: you uploaded last month's statement, or another person's, or a different account.
  • Wrong format or garbled rows: the file was read but amounts or text are off.
  • Overlapping dates: the new file repeated rows that were already in.
  • Wrong base currency: you chose it before importing, and it cannot be changed once transactions exist.

That last one is the exception. If the base currency is wrong and transactions already exist, a fix by re-importing is not available, so check it before the first import of a new workspace.

Step 2: find the rows the bad import added

Open the import history. It lists each upload with a status, so you can see which batch is the one you want to undo. Then filter the transaction list to the account and dates that batch covers. Compare the count of rows with what the statement shows. If you added hand entries or made edits in the same period, take care not to lump those in with the bad batch.

Step 3: remove the bad rows, not the whole month

Select just the rows from the bad import and delete them. They go to the recycle bin for 30 days. This matters because it gives you a way back: if you later realise a few rows were genuinely right, restore those. Before you delete a large set, scan the first and last few rows so you are sure they belong to the bad batch.

Step 4: fix the cause, then import once more

CauseFix before re-importing
Wrong accountChoose the correct bank, card or wallet account before uploading
Wrong fileDownload the right statement and check the date range and account number on its first page
Garbled text or amountsDownload the same statement as CSV or Excel instead of PDF
Overlapping datesDownload only the dates not yet in the tracker

Say your savings statement for October has 86 rows and you accidentally uploaded it under your credit card. Delete those 86 rows, pick the savings account, upload again, and check that the review screen again shows 86 rows. A count that matches the statement is a good sign nothing is doubled.

What if I have already categorised the bad rows?

You will lose that effort on deleted rows, but not entirely. Corrections you made become part of your history, which later imports use to match similar merchants, and any rules you created stay. So the second import often categorises more quickly than the first. For a handful of rows that you carefully fixed by hand, consider keeping them and removing only the ones that are wrong, rather than wiping the batch.

After the re-import: a quick check

  1. Confirm the row count against the statement.
  2. Check the opening and closing dates are what you expected.
  3. Look at any possible duplicate flags and settle them using the bank's own list.
  4. Spot-check three large amounts, including one credit.
  5. Open the import history and confirm the new batch shows a finished status.

If the same file keeps failing, try a different format of the same statement, or finish the upload from the web on a laptop, which uses the same workspace.

SPENDIXAI JOURNAL

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

Explore your reports