When your Affinity Credit Union 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. Even a paper statement that was scanned or photographed on a phone can be read. The engine applies optical character recognition first, then treats the result exactly like a born-digital file. The same transaction table is built, and the same running-balance verification runs against it, so a misread digit is flagged rather than silently exported.
Whichever order your Affinity Credit Union statement prints its columns in, each one still has to land in the right place once it reaches a spreadsheet. When the transaction table is ready, every column is mapped to its export counterpart. Date, description, debit amount, credit amount and balance are each routed into the correct field, preserving the decimal precision from the original. You review the mapping once; after that the same column layout is reused for every statement with the same structure, so a bookkeeper processing a dozen statements a month does not remap fields each time.
If the running balance on your Affinity Credit Union statement does not add up row by row, that is exactly the kind of mismatch Convert·Into is built to catch rather than pass through silently. Every statement passes through a line-by-line arithmetic check before any export file is produced. The engine reads the closing balance printed on the original document, recalculates the running balance from the extracted transactions, and flags any row where the two do not agree. The rest of the file remains intact; you correct the one flagged entry instead of redoing the whole conversion, which keeps a single scan artefact from turning into a full reconciliation problem.