If the running balance on your Gazprombank 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 converted statement is verified against its own printed closing balance. The engine reads the running-balance column line by line, checks each figure against the previous one, and confirms the final line matches the statement’s declared total. A row that does not add up is flagged for review rather than silently accepted, so you can reconcile directly against the verified output.
If the Gazprombank statement you have is a scanned image or a phone photo instead of an exported file, that alone does not stop Convert·Into from reading the transaction table off it. Scanned or photographed statements are read with OCR and then put through the same verification step. The running-balance check catches digits the scan may have misread, a 3 mistaken for an 8 that would otherwise break the reconciliation. Because every row is checked arithmetically, the converted data stays as reliable as a clean file, even when the source is a photograph taken at a desk.
However your Gazprombank statement lays out its columns, that layout is only useful to you once it is mapped correctly into the export. Column mapping puts each field where it belongs in the export. The engine identifies the debit, credit, date, description and balance columns in the original layout and matches them to the corresponding columns in Excel, CSV or QBO. You can review the mapping before export, reorder columns with a drag, and confirm the closing balance carries through to the final file. The result is a clean ledger, structured for your own workflow, with no rekeying and no copy-paste drift.