Rossum’s pricing has a complex structure, but a major part of Rossum’s subscription is the volume allowance for document processing. This article clarifies the exact cases of when and how Rossum charges for a document or a page.

The basic principle of Rossum’s billing is that we charge for documents or pages processed by our platform. That means every document or page imported into Rossum is counted, and data capture configuration determines the exact billing mode. Let’s explore some concrete examples.

Which documents or pages are billled

A document or page is billed when it is processed by Rossum’s AI Engine. That means when the document goes from importing state to review, confirmed, or exported state for the first time, depending on the configured automation level. (Documents that end up in the failed_import state are not billed.)

For per-page billing, you’re billed for all pages of a document (regardless of which pages contain captured data). However, this only applies up to the configured maximum number of pages per document processed by the AI Engine (32 pages by default).

Regarding the document splitting – you’re billed for both the original source document (or pages) and all produced child documents (or the sum of their pages) because the AI Engine currently re-reads these documents from scratch.

Billing for line items

Billing for document processing with line item extraction is applied differently than it is for document processing without line item extraction. Such a distinction is determined based on the extraction schema configuration. If a table (“multivalue tuple” in Extraction schema terminology) is configured as part of the schema, billing for the document or page applies as undergoing a line item extraction (regardless of whether line items are eventually found on the document or page). An exception is applied on “tax_details” tables , which do not count as line items.