When your Bank Negara Indonesia statement arrives as a scan or a photo of the page rather than a clean export, the numbers on it are just pixels until they are extracted. A statement document that was scanned or photographed arrives as a picture of a page, not as a table of data. The engine reads that image with OCR, building a transaction list row by row, and then runs the same running-balance verification it applies to every other file. A digit that smudged to look like a different number will break the balance: the row is flagged so you can correct it, while the rest of the file exports cleanly.
If a transaction description on your Bank Negara Indonesia statement is long enough to wrap across two or three lines, it still needs to land in a single row once converted. When a transaction description is long enough to wrap onto a second or third printed line, most extraction tools split that single entry into several broken rows, each carrying its own phantom date and a fragment of text. Convert·Into reads the statement's column structure first, then recognises that a continuation line has no date or amount of its own, so it stitches the description back together into a single row before mapping the columns into your chosen export format. The output preserves every field: date, description, debit, credit and the verified running balance, delivered as Excel, CSV or QBO with no value shifted into the wrong column.