DocParse Docs
Dashboard

Overview

A high-level picture of how DocParse works end-to-end. If you're new here, this gives you the mental model in 5 minutes.

The flow

plaintext
┌──────────┐    ┌───────────────────────┐    ┌─────────────────┐    ┌───────────┐
│ Your app │ →  │ POST /api/v1/create…  │ →  │ DocParse engine │ →  │ JSON back │
└──────────┘    └───────────────────────┘    └─────────────────┘    └───────────┘
                           ↑                                               ↓
                           │                                               │
                           └──── poll getBatchResults OR webhook ──────────┘
  1. Create an Extraction — call POST /api/v1/createExtraction with your field schema and files in a single multipart request.
  2. Upload more documents via POST /api/v1/uploadFiles — each call creates a new batch automatically.
  3. Get results back by polling POST /api/v1/getBatchResults or via the webhook you've configured.

Two completion patterns

  • Polling — call POST /api/v1/getBatchResults until all files reach a terminal status. Good for ad-hoc scripts and CLI tools.
  • Webhook — register a URL in the dashboard's API page. We POST the results to your endpoint when each file finishes. Good for production integrations.

See Polling vs Webhook for the detailed comparison.

Two products

DocParse has two API surfaces:

  • Data Extraction API — schema-driven field extraction. You define what fields to pull from invoices/receipts/forms; we return them as JSON. Endpoints →
  • Document Classification API — given a folder of mixed documents, sort each one into a category (invoice / contract / resume / etc.) and optionally chain into an extraction template per category. Endpoints →

Beyond extraction

Extraction is the start. DocParse also makes the data trustworthy, reviewable, automatically delivered, and measurable:

  • Validation Rules — business-rule checks that run after every document and flag bad data for review.
  • Review Queue — one cross-extraction inbox of documents needing a human; confirm or correct, and they clear.
  • Email Ingestion — a forwarding address per extraction; emailed attachments import and extract automatically.
  • Export Destinations — push results (JSON/CSV, signed) to a URL or automation tool on processed or on confirmed.
  • Analytics — straight-through-processing rate, exception rate, validation pass rate, and per-extraction breakdowns.

Free tier

Every account gets 100 pages every month, free, forever. Once you're past 100, see pricing.

Where to next