100 Simple Claims: PDF Extraction and Reporting Delays

The challenge

A claims team needs a searchable list of 100 forms. Extract the records, then break down the workload by claim type, status, and time taken to report each incident.

Start here

Open the SharePoint folder linked above. claim_0001.pdf through claim_0100.pdf are the inputs; ins_claims_simple.zip packages those same 100 forms. Use either the loose files or the archive. claims.csv and claims.json are answer files.

Preview claim_0001.pdf and claim_0002.pdf. Each is a compact one-page form with selectable text and a single total amount. This collection does not contain itemized expenses.

Try it

1. Recover all 15 fields in the reference claims table from the PDFs.

2. Calculate the total claimed amount and counts for each status and incident type.

3. Find the largest claim and calculate the average incident-to-report delay.

4. Prove that your archive handling has not doubled the number of claims.

What to build

Build a 100-row claims CSV or JSON array, a per-file validation log, and answers to the summary questions. Use the clean variants for checking after you have extracted a pilot batch.