Skip to main content

Using OCR Autofill in New Joiner Form

Written by StaffAny

Contents of this article are applicable to the following users

Product: HRAny

Platform: Web

Access Level: Manager/Owner (setup), Any (new joiner form)

OCR (Optical Character Recognition) autofill on the New Joiner Form reduces manual data entry by automatically extracting information from documents uploaded by new hires. When OCR is enabled, new joiners upload their documents first, and the system suggests values for fields such as legal name, ID number, and date of birth — which the new joiner can review and apply before continuing.

This guide will cover how to:

Set Up Document Types for OCR

Before enabling OCR on a form, each document type that should be scanned must be assigned an OCR extraction type in Employee Document Types.

  1. Go to Settings > Team Management > Employee Document Types.

  2. Create or edit a document type.

  3. Under New Joiner OCR extraction, select the document kind this type represents:

    1. Do not scan — OCR is not applied to this document type.

    2. NRIC/FIN — extracts fields from Singapore IC or FIN card.

    3. Passport — extracts fields from a passport.

    4. Bank Statement — extracts bank-related fields.

  4. Click Save.

Each OCR extraction type (NRIC/FIN, Passport, Bank Statement) can only be assigned to one active document type at a time to avoid conflicting extractions.


Enable OCR Autofill on a New Joiner Form

OCR autofill is enabled per form when creating or editing a New Joiner Form.

  1. Go to HR > New Joiner Form.

  2. Click Create Form or Edit an existing form.

  3. Enter a Form Name.

  4. Check Enable OCR autofill. When enabled, new joiners will be prompted to review values extracted from their uploaded documents.

  5. In the Documents tab, select the document types to include in the form. Only document types with a configured OCR extraction type will contribute to autofill.

  6. Select any other fields to include.

  7. Click Save.


How New Joiners Experience OCR Autofill

When a new joiner opens an OCR-enabled form, the Documents step appears first.

  1. New joiner opens the form link and reaches the Documents step.

  2. New joiner uploads their documents (e.g. IC, passport).

  3. After upload, an OCR review screen appears showing extracted values. For each field, the new joiner can:

    • Check or uncheck the field to decide whether to apply the extracted value.

    • Select a different source document if multiple documents extracted a value for the same field.

    • Click View source to open the original document the value was extracted from.

  4. Click Apply to pre-fill the selected values, or Skip to fill in all fields manually.

  5. New joiner continues through the remaining steps with pre-filled values already in place. You can still edit the value afterwards if needed

  6. New joiner reviews and submits the form.


View Source Document for OCR-filled Fields

After submission, HR and managers can see which fields were filled by OCR and view the source document used.

In the Add Staff flow:

  1. Go to HR > Onboarding.

  2. Open the new joiner record.

  3. Fields filled by OCR show a View source button next to the value.

  4. Click View source to open the original document in a new tab.

In the View Submission flow:

  1. Go to HR > Onboarding.

  2. Click the submission to view.

  3. Fields filled by OCR show a View source button next to the value.


FAQ

Q: Which fields can OCR autofill?

OCR can extract fields such as legal name, ID number, date of birth, gender, race, religion, address, and bank account details — depending on the document type. The exact fields depend on what is readable from the uploaded document and which fields are included in the form.

Q: What happens if the new joiner skips the OCR review?

The form fields will not be pre-filled. The new joiner can fill in all fields manually as usual.

Q: Is OCR autofill available for all organizations?

OCR autofill is currently available for organizations in Singapore only.

Q: My scanning results were not accurate, what happened?

Currently users can expect lower accuracy when scanning blue NRIC and FIN as we are still improving the effectiveness of our OCR feature

Did this answer your question?