AI System for Classifying Documents by Type and Retention Rule

The problem

In day-to-day operations, documents, sender details and filing plans arrive in separate records and files, preventing the team from reviewing the same case in a single view.

As a result, the link to source evidence can be lost while preparing a document type and target repository recommendation, leading to repeated checks and uncertainty about the basis for the decision.

Incomplete automation may overlook cases involving personal data or an unclear document type, while poorly controlled automation risks inappropriate access and retention.

What we set out to improve

• Time required to prepare a document type and target repository recommendation for review
• Correct routing of records involving personal data or an unclear document type to the appropriate specialist queue
• Recording of recommendations corrected or rejected by the records manager

How the system works

1. 01 Retrieve document, sender and filing plan data from an authorised source
2. 02 Verify the source system, record identifier and recency information
3. 03 Transform the fields into the target process schema
4. 04 Prepare a document type and target repository recommendation draft
5. 05 Compare the draft with business rules and source records
6. 06 Check for exceptions involving personal data or an unclear document type
7. 07 Obtain the records manager's approval, correction or rejection
8. 08 Write the input, recommendation, changes and final decision to the audit log

Methods we used

• Source-traceable data extraction from documents, sender details and filing plans
• Structured output that restricts document type and target repository recommendation fields through business rules
• An exception gate that separates cases involving personal data or an unclear document type
• A queue that shows the source and recommendation together for the records manager
• Duplicate-processing controls and rollback records to mitigate the risk of inappropriate access and retention

Where people stay involved

Data and security

Only the fields required for the task are processed from the document, sender details and filing plan. Source-system permissions remain in force, and personal or commercially sensitive data is minimised before being provided to the model. Every read, recommendation, human amendment and write to the target system is logged.

Who this suits

Frequently asked questions