01
Infer the structure
The model separates document fields, repeating rows, totals, and notes based on the file in front of it.
What is included in the invoice JSON export?
Dynamite Docs extracts invoice fields and item rows from a PDF, scan, or photo. Review the values beside the invoice, then export JSON containing column definitions, row data, and document metadata.
What to expect
The product export keeps inferred columns, repeating rows, and extraction metadata separate so a developer can map the reviewed values into an application-owned schema.
The full workflow
The output stays beside the evidence.
Check the proposed fields against the source, correct a value, then export the reviewed invoice JSON.
Dynamite Docs- 01Select the source file
- 02Review fields and rows
- 03Download invoice JSON
02
Correct against the source
The source remains visible while you edit cells, rename columns, inspect warnings, and mark the result reviewed.
03
Send useful data onward with Dynamite Docs
Download invoice JSON, or move a recurring workflow into Google Sheets, Drive, webhooks, or the API.
Dynamite Docs- supplier.name
- Northstar Office Supply
- buyer.name
- Riverbend Trading Ltd
- invoice_number
- INV-0842
Choose the AI route
Bring a free provider key. Keep the review workflow.
Open AI & Processing, choose Connect Provider, and pick a provider. The setup screen links to that provider's key page. Create a free key where the provider offers one, paste it into Dynamite Docs, verify it, then select a model for the run.
Your provider key is encrypted before it is stored.
Switch between hosted, own-key, privacy-restricted, and local models.
The same source-first review screen works whichever route you choose.
Choose the page for the job using Dynamite Docs
Use Invoice to JSON when engineering teams need a verified invoice JSON payload to design an integration, inspect document fields, or prototype an ERP ingestion mapper. Before processing, determine which fields belong in document metadata versus repeating line-item arrays, and establish required property constraints.
During review, cross-reference supplier identity, buyer details, invoice number, issue date, currency, tax amounts, and line items with the source document. Ensure numeric values maintain decimal precision, dates adhere to ISO 8601 formatting, and string fields preserve leading zeros in tax IDs or purchase order codes.
Use Free Invoice OCR when testing whether an obscure scan or layout can be parsed before working with JSON. Use PDF to JSON for non-invoice documents like operational logs or government forms. When moving to automated production pipelines requiring API authentication, webhooks, idempotency keys, and saved corrections, integrate via the REST API and invoice accounting workflow.
Invoice workflow
Choose the page that matches your next task.
Test invoice OCR firstCheck whether the source layout, scan, and line items can be read.Process recurring invoice batchesAdd stored files, saved corrections, reviewers, and repeatable exports.Map GST invoice JSONHandle GSTIN roles, tax components, null states, and nested line items.Use the document extraction APIAdd authentication, stable mapping, retries, idempotency, and delivery checks.
The 28-day Free trial includes 25 page-equivalents per month shared by hosted and own-key processing, 5 hosted pages per document, and 5 documents per day. Anonymous trials have separate daily limits.
Setup, review and export questions
Is the JSON schema fixed?
By default, the tool infers a schema from the invoice. In the full workspace you can define or save a reviewed schema when downstream code needs stable property names.
Are line items returned as an array?
The JSON download includes a rows array, column definitions and document metadata. Invoice line-item field names depend on the extracted schema; a line_items key is not guaranteed.
Can I use the result in an API?
Yes after review. Validate the JSON against your own schema and business rules before sending it to accounting, payment, or tax systems.
How should amounts and dates be represented?
Keep the reviewed source value available, then apply an explicit mapping rule. Use decimal-safe handling for money, preserve currency separately, and normalize a date only after confirming its source format. Avoid turning invoice numbers, tax identifiers or purchase-order references into numbers.
How do I keep integrations stable across supplier layouts?
Map each reviewed extraction into your own versioned contract rather than exposing inferred columns directly to production code. Validate required fields and types, reject unexpected changes, and keep a dead-letter or exception path for documents that do not meet the contract.
Choose the tool that matches the file.
Process the whole folder in one run.
Starter and above add stored files, batch jobs and connected exports. Email intake requires Pro or Scale. File and processing limits depend on your plan.
Compare batch plans