Loss Run Data Request and Normalization Kit

Get clean loss runs on a predictable cadence: request templates, standard schema, QA checks, and a monthly refresh workflow.

Run the OperationCoreStarter Kit60 minFinance, Procurement and Ops

You cannot control claim cost if your claim data is stale, inconsistent, or incomplete.

This kit creates one standard way to:

  • request loss runs from broker and TPA
  • normalize the data into a consistent schema
  • run basic QA checks
  • refresh monthly so open claims do not drift unnoticed

What you will produce

  1. Loss run request template
  2. Standard loss run schema and data dictionary
  3. QA checklist and gap log
  4. Monthly refresh workflow

Step-by-step implementation

Step 1: Set the loss run request cadence

Default recommendation:

  • monthly refresh for open claims
  • quarterly deeper review for patterns and controls

Step 2: Send a standard request to broker and TPA

Use one template so you do not renegotiate data each time.

Loss run request email (copy and paste)

Copyable template (TEXT)

Subject: Loss run request (monthly refresh)

Hi [Name],

Please provide the latest loss runs for the following lines of coverage and policy periods:
- Lines of coverage:
- Policy periods:
- As of date:

Please include:
- paid, reserve, incurred
- claim status (open or closed)
- claim description and cause
- location and business unit identifiers if available
- claimant and vendor fields if available

Preferred format:
- CSV or Excel with one row per claim and consistent column names
- Include a data dictionary if your columns differ from prior runs

Thank you,
[Name]

Step 3: Normalize to a standard schema

Do not aim for perfection. Aim for consistency.

Standard loss run schema (copy and paste)

Copyable template (TEXT)

Standard Loss Run Schema

Required columns:
- claim_id
- policy_number
- carrier
- line_of_coverage
- policy_effective_date
- policy_expiration_date
- claim_date_of_loss
- claim_reported_date
- claim_status
- claim_cause
- claim_description
- paid_amount
- reserve_amount
- incurred_amount
- location
- business_unit
- adjuster_name
- tpa_name

Optional but useful:
- claimant_type
- vendor_name
- litigation_flag
- subrogation_flag
- deductible_amount

Step 4: QA checks and gap logging

QA checklist (copy and paste)

Copyable template (TEXT)

Loss Run QA Checklist

- File date and as-of date captured
- Claim IDs present and unique
- Paid, reserve, incurred are numeric and non-negative
- Incurred equals paid plus reserve (or variance explained)
- Claim status present for all rows
- Line of coverage present for all rows
- Missing location or business unit percent estimated
- Subrogation flag present or derived where possible

Gap log (copy and paste)

Copyable template (TEXT)

Loss Run Gap Log

| Date | Source | Gap | Impact | Owner | Resolution target date | Notes |
| ---- | ------ | --- | ------ | ----- | ---------------------- | ----- |

Step 5: Set a monthly refresh workflow

Monthly refresh checklist (copy and paste)

Copyable template (TEXT)

Monthly Loss Run Refresh

1) Request loss runs by day 3 of the month
2) Normalize to standard schema within 2 business days
3) Run QA checks and update the gap log
4) Publish the open claims list for triage and reserve review
5) Confirm subrogation candidates are flagged

Common failure modes

  • data arrives late, so open claims drift for months
  • columns change every month, so comparisons break
  • no QA, so incurred math errors go unnoticed
  • no ownership, so gaps persist indefinitely

Change log

v1.0 (2026-01): Latest release