DocReadi user guide

DocReadi turns invoices, receipts, delivery notes and your own custom document types into clean structured data and delivers it wherever your finance process lives — accounting-package CSV, API, webhooks, SFTP, S3, or email. You send documents in; DocReadi classifies, reads and checks them; you review anything it isn't sure about; approved data flows out. This guide explains every screen and setting as shipped today.

Chapters

• Getting started — Signing up (form or Sign up with Xero when available), workspaces and the tenant switcher, invites, logging in (incl. universal login and 2FA prompt), the onboarding checklist, your first upload, and reading the dashboard.
• Sending documents in — Every way a document reaches DocReadi: web upload, the per-workspace email-in address, WhatsApp, Inbox watched folders (SFTP/S3), and the REST API (pointer to /api/docs-guide; API keys). Accepted file types, including e-invoice XML.
• How processing works — The plain-language pipeline: classify, parse (native text vs scanned/OCR), extract, validate (cross-field checks), counterparty match, and the confidence gate. The shorter path an e-invoice XML takes. What confidence means, why a document lands in review, and what consumes credits.
• Reviewing documents — The review queue and fast review, the document viewer, editing fields, Re-extract, the split editor, bulk actions, approve/reject/unapprove, duplicate handling, and auto-approve thresholds with the required acknowledgement.
• Master data — Entities (your own legal entities, direction validation), counterparties (aliases, candidates, auto-confirm, anomaly baselines), GL accounts, tax codes, posting defaults and the postings review + learning loop, and registry import/export.
• Custom templates — Custom document types: the field designer, critical-field checkbox, the test-extract sandbox, the template gallery, and the templates API.
• Reports and exports — The report builder and saved reports, CSV/Excel/JSON export, the built-in reports (cost, LLM activity, credits, models), and the per-document export/download chooser.
• Delivering documents — Connections (shared credentials), Outbox destinations, bundle formats (Xero CSV, archive, PDF + sidecar and more), runs, schedules, on-approval delivery, and outbound HMAC-signed webhooks.
• Billing and credits — Plans and tiers, credits and top-ups, the ZAR (Paystack) and USD/EUR/GBP (Polar) rails, firm/pooled billing, the usage page, and SARS tax invoices.
• Workspace administration — Users, roles and invites, two-factor auth, active sessions, the activity log, number and date formats, filename/folder rules, retention, privacy rights requests, and closing a workspace (30-day reversible window).
• Troubleshooting — The pipeline errors page, common validation failures in plain language, duplicates, the awaiting-credits state, and how to contact support.
• Reference — The 13 built-in document types (including statements of account, and why a statement is not an invoice), the document status glossary, the roles/permissions matrix, and the tenant-visible feature flags (plus the platform-admin Models and Provider balances pages).

Looking for the developer/API reference? See /api/docs-guide. Can't find an answer? Email [email protected].