FAQ
No. Paper sits earlier in the cycle than your policy platform. Whichever platform you run stays the system of record, nothing is migrated, nothing is replaced. Where your platform offers an API, we use it. Where it doesn't, Paper works through the platform's system and automates the workflow end-to-end.
Email bodies, PDFs, scans, spreadsheets and the attachments inside forwarded chains. A submission that arrives as a covering email with a schedule attached three replies down is the normal case, not the edge case.
Every value carries the document and the page it was read from, and is re-checked against that source. Anything the source does not support is discarded rather than guessed, and anything uncertain is queued for a person before it is pushed. You can open any completed submission and see where each figure came from.
The gap is flagged rather than filled. Paper never invents a value to complete a record, and a submission that falls below the confidence threshold does not get pushed into your system unreviewed.
No. Models are ringfenced, client data is separated, and there is a full audit trail from every extracted value back to its source document. See Security for the certifications and the detail.
That depends on your platform and how many schemes you want live at once. The honest answer is that it is a conversation rather than a number, and the first step is small: we look at real submissions from the scheme giving you the most trouble.
Yes, and the renewal volume is usually the larger half. The data captured when a submission first arrives is what drives the renewal twelve months later.