Sort mixed bookkeeping documents automatically
Yes. untxt. reads each document in a mixed batch and works out what it is. Receipts, invoices, credit notes, and bills are classified and prepared for review; a bank statement, or anything that is not a document untxt. extracts, is recognized and set aside rather than forced in. You tag nothing and pick no categories.
What "mixed" means in a real batch
One upload often holds several document types at once: supplier invoices, card receipts, a bank statement, a credit note, sometimes a contract that is not a bookkeeping document at all. Separating that by type is the slow first step most people do by hand before any data entry begins.
How untxt. sorts a batch by document type
untxt. reads and understands each document, decides what it is, and classifies it as an invoice, receipt, credit note, or bill. It groups multi-page documents together and sets aside anything it does not extract — a bank statement, a contract — rather than forcing it into the books. The decision comes from reading the document itself, so a receipt is recognized as a receipt no matter how it arrived or what it was named.
Sorting inside a client's books
You work inside a business in untxt., so a batch is sorted within that client's Cost and Spending. Each document lands classified and ready, and you move to the next business for the next client.
What the sorted result looks like
Every document arrives in the review queue already classified, with a confidence score, the original on the left and the extracted data on the right. If a type ever looks off you correct it in one click and move on.
Why rule-based sorting misfiles things
Old tools sort by rules you maintain: this sender is always this type, this inbox is receipts. The moment a document arrives by a different route, the rule files it wrong and you find out later. untxt. decides by reading the content, so where a document came from does not change what it is recognized as.