# Document OCR for KYC and customer onboarding

Source: https://idscanly.com/kyc-onboarding/
Language: en

DOCUMENT OCR FOR YOUR WORKFLOW

Turn document images into fields your onboarding system can review. IdScanly provides an extraction layer for teams building their own identity and compliance workflow.

[Discuss this integration](https://idscanly.com/contact/?useCase=kyc-onboarding) [Try the live demo](https://idscanly.com/live-demo/)

![Original illustration of the document workflow for KYC & onboarding](https://idscanly.com/images/use-cases/kyc-onboarding-1200.webp)

KYC & onboarding

## An extraction boundary in your identity stack

Customer journeys often need the same printed details in several systems. OCR gives your application a structured starting point, while allowing rules, human review and specialist checks to remain explicit.

## Map source fields into your own schema

Names, document identifiers, dates, issuing country and supported MRZ or barcode data, depending on the document.

[Read the API documentation](https://idscanly.com/developers/)

## Document views and asynchronous results

1. ### Open a scan session
   
   Capture representative front and back views required by your document policy.
2. ### Submit the required views
   
   Normalise and map returned fields, preserving source values and confidence where available.
3. ### Resolve output and uncertainty
   
   Compare against the application and run the independent identity checks required before approval.

## Keep extraction data out of application logs

Redact tokens, document numbers and OCR payloads from general-purpose logs. Agree transient processing and retention responsibilities before connecting the extraction layer to production identity workflows.

Do not describe an OCR-only flow as full KYC. Define the capabilities of each component and an escalation path for uncertain cases.

[Privacy Policy](https://idscanly.com/privacy/)

## Two integration patterns to evaluate

### Front-and-back document flow

Your application submits the views required for the selected document within its scan-session workflow. Map the returned merged fields to your schema and leave unavailable fields explicitly unresolved.

### Handoff to an external verification service

Pass only the fields that the next system needs, with their source and uncertainty where available. Correlate results using your own workflow identifier, not an assumed identity-verification verdict.

## Developer questions about onboarding OCR

What is the difference between OCR and identity verification?

OCR reads information from an image. Identity verification also needs evidence that the document is genuine and belongs to the person, plus any regulatory checks relevant to the use case.

Is a missing field the same as an empty value?

No. Preserve the distinction in your schema and ask for correction or review instead of inventing a value.

Does polling create another extraction?

Result checks retrieve processing state. Follow the documented session workflow instead of submitting duplicate document uploads.

Can high confidence bypass independent verification?

No. Extraction confidence concerns reading the image; authenticity and holder verification remain separate capabilities.

KYC & onboarding

## Discuss this integration

[Discuss this integration](https://idscanly.com/contact/?useCase=kyc-onboarding)

## Related use cases

[Hotels & hospitality](https://idscanly.com/hospitality/) [Fintech](https://idscanly.com/fintech/) [Car & mobility rental](https://idscanly.com/car-rental/) [Property rental](https://idscanly.com/property-rental/) [Access control](https://idscanly.com/access-control/)