unthinkable
Case study · Pramerica Life Insurance, India · life insurer

Pramerica's customers' national ID numbers are masked in every stored document, ahead of the government's deadline.

Pramerica Life Insurance wanted every customer's national ID number protected in its records, as a new Government of India rule required, without the cost and errors of doing it by hand.

IndiaInsurance marketplaces & operations

What changed

500,000+documents masked (as reported)
10 secondsaverage time per document (as reported)
98%accuracy (as reported)
5 million+jobs run by the bots (as reported)
What stood in the way

What stood in the way.

What stood in the way
  • A new government data security rule required national ID numbers to be stored masked.
  • About 500,000 customer documents in IBM FileNet already held the full number.
  • Many files were scanned by hand, so the number was never in the same place on the image.
  • Doing it manually would have taken a great many hours and produced errors.
What changed
  • 500,000+documents masked (as reported)
  • 10 secondsaverage time per document (as reported)
  • 98%accuracy (as reported)
  • 5 million+jobs run by the bots (as reported)
What we built

What we built.

  • A proof of concept first, to test accuracy and feasibility.
  • UiPath bots that download each unmasked image from FileNet, open it, mask the number and save the masked copy back in place of the original.
  • A separate utility using the Google Vision API to find exactly where the 12-digit number sits on each image.
  • Custom scripts in Python, with the utility installed on desktops at Pramerica's own premises for secure access.
  • UiPath
  • Google Vision API
  • Python
  • IBM FileNet
What it shows

What this work shows about us.

We know insurers' compliance work

A rule change touches every stored policy document. We made the change across the whole archive, on the insurer's own machines, before the deadline.

Automation that handles messy scans

Hand-scanned images put the number anywhere; locating it first is what let the bots reach 98% accuracy (as reported).

Want to automate ID document masking with RPA like this? Tell us what you want to change. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with the right team.

Start your plan ↗
If we built it today

If we built it today, with AI.

This work came before today's AI. If Pramerica Life Insurance asked us now, this is what we would add, step by step.

Changes & renewal

An agent would read each scanned document as it arrives, find any ID number, mask it before it is stored and send unclear scans to a person. Compliance still owns the masking rules.

Questions & complaints

An agent would answer a customer asking why their ID shows masked and which documents hold it, from the document record. Staff still handle any request to see the original.

Ask Unthinkable

Want this for your business?

Tell us what you run or what you're building. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with a senior AI engineer.

Send us a note instead

Required

Thanks. A senior AI engineer will reply to you.
u
Ask UnthinkableClarity first, then the right team
Your conversation

An AI agent. This prototype uses prepared answers and sends nothing.