DOCUMENT AI API · DOCUMENT ANALYSIS

Rule-based intelligent document review.

Configure rules per use case, NDA gap analysis, contract compliance checks, policy adherence validation, regulatory filing review. The Document Analysis API runs your rules against incoming documents and returns structured findings through the hosted API.

Try now Read the docs →
Rule-based document review Structured findings, not just fields Configured per use case
WHEN TO USE IT

Beyond extraction, into judgement

Extraction APIs get you data. Document Analysis gets you a verdict, "this NDA is missing 4 standard clauses", "this contract violates your master services agreement template in 7 places", "this insurance claim doesn't match the prior auth on file".

NDA gap analysis

Receive a counterparty's NDA → compare against your gold-standard template → identify missing/modified clauses. Rules-driven and fast. Used by legal ops + procurement.

Contract compliance review

Check vendor contracts for non-standard liability, indemnification, IP, or termination terms. Configurable per your contract standards.

Regulatory filing validation

Pre-submission check of regulatory filings (RBI returns, GST returns, insurance filings) against the latest requirement set.

Policy adherence checks

Insurance: does claim match policy coverage? Lending: does borrower meet credit policy? Reads docs, checks rules, flags exceptions.

HOW IT WORKS

From your domain rules to live API

1
Discovery call
Talk through your use case + show us sample documents + describe the rules / decisions you'd ask a human reviewer to make.
2
Rule configuration
We codify your rules into our analysis engine. Iterate on your sample document set. Validate accuracy.
3
Sandbox + integration
Dedicated sandbox endpoint with your ruleset. Integrate with your application. Test in production-like conditions.
4
Go live + ongoing tuning
Production API endpoint with your rules. Ongoing rule tuning. Monitor findings, refine over time.
PRICING

Scoped to your use case

Document Analysis is rule-based and configured per use case, so there isn't a fixed per-page rate. Pricing is scoped once we've seen the documents and rules involved, taking into account:

  • ✓ Rule complexity, number of rules, dependencies between them, cross-document checks
  • ✓ Document volume, pages processed per month
  • ✓ Integration scope, sandbox + production + ongoing tuning
  • ✓ Support level, periodic rule reviews + updates

Talk to us about pricing

Pricing for Document Analysis isn't published yet, it's scoped per use case. Share your sample documents and we'll propose a scope and pricing. Contact us →

FAQ

Common questions

What does the Document Analysis API do?
It runs your configured rules against incoming documents and returns structured findings via the hosted API. Unlike extraction, which returns data fields, analysis returns a verdict, for example which standard clauses an NDA is missing or where a contract deviates from your template. The ruleset is configured per use case.
How is document analysis different from data extraction or OCR?
Extraction and OCR turn a document into structured data or text; document analysis evaluates that document against your rules and returns judgements. It answers questions such as whether a filing meets the latest requirement set or whether a claim matches the policy on file, flagging exceptions rather than only reading fields.
What types of documents can it analyse?
It is used for NDAs, vendor contracts, regulatory filings such as RBI returns, GST returns and insurance filings, as well as insurance claims and lending documents. Because the ruleset is configured per use case, the exact document types and checks are defined during scoping rather than fixed in advance.
Can I configure my own rules and templates?
Yes. Rules are codified per use case, for example comparing a counterparty NDA against your gold-standard template, or checking vendor contracts for non-standard liability, indemnification, IP or termination terms. Rules are iterated on your sample document set and validated for accuracy before go-live.
How fast is an analysis?
NDA gap analysis is fast, and turnaround for other use cases depends on the ruleset and the document — it is validated during the sandbox phase before it goes into production.
How do I get started and integrate it?
Onboarding starts with a discovery call to review sample documents and the decisions a human reviewer would make, followed by rule configuration, a dedicated sandbox endpoint for integration and testing, then a production API endpoint. Findings are returned in a structured form your application can consume.
How much does the Document Analysis API cost?
Pricing for Document Analysis is not published yet and is not a fixed per-page rate. It is scoped per use case, because the ruleset is configured for your documents. Contact us to talk through your use case and get pricing.
Are the rules kept up to date after go-live?
Yes. Rules are tuned periodically and refined over time as findings are monitored. For use cases such as regulatory filing validation, checks are run against the latest requirement set so filings can be validated before submission.
THE FULL TOOLKIT

Explore the rest of the Document AI API

One wallet, every capability — take this and you also get all of these.

Data Extraction OCR Data Masking Document Analysis

Talk to us about your analysis use case

Share your sample documents and we'll propose a scope, in under a week.

Try now Contact us