Loss runs to structured data in one business day.
LossRunDesk turns carrier loss runs and submission documents into clean Excel or CSV files for wholesale brokers and MGAs. Totals are checked against the carrier's reported totals, and anything uncertain is flagged for you to check.
Send us a sampleOr start with a free pilot of 5 loss runs.
| Claim # | Date of loss | Status | Paid | Reserved | Incurred |
|---|---|---|---|---|---|
| GL-1041 | 03/14/2023 | Closed | 12,500 | 0 | 12,500 |
| GL-1187 | 09/02/2023 | Open | 18,200 | 12,500 | 30,700 |
| GL-1223 | 01/?/2024 | Open | 0 | 5,000 | 5,000 |
| Total | 30,700 | 17,500 | 48,200 | ||
Retyping loss runs is still someone's job.
Every carrier formats loss runs differently. Someone on your team reads each PDF, finds the policy periods, claim counts, and paid, reserved, and incurred figures, then retypes them into a spreadsheet or submission.
The work is slow, a single transposed number can go unnoticed, and it takes skilled staff away from placing business.
How it works
-
Email us the PDFs
Scanned or digital, from any carrier.
-
We extract and validate
Data is extracted automatically. Claim-level amounts are reconciled against the document's reported totals, and policy periods and dates are checked for consistency.
-
You get your file
Excel or CSV in the layout your team or rating sheet needs, within one business day. Anything we couldn't confirm is flagged in the file.
What we handle
Loss runs first
For each policy and line of coverage, we capture:
- Carrier
- Policy number
- Policy period
- Valuation date
- Claim number
- Date of loss
- Claim status
- Paid, reserved, incurred
Other submission documents
ACORD applications, statements of values (SOVs), supplemental applications, and similar documents.
If you're not sure whether we can handle a document, send it and we'll tell you.
How your files are handled
- Automated validation
- Where a document reports totals, claim-level paid, reserved, and incurred amounts are reconciled against them. Policy periods and dates are checked for consistency.
- Flagged, never guessed
- If the system isn't confident about a value, it's marked in your output so you can check it against the source. Nothing is silently filled in.
- Confidential handling
- Your documents are used only to produce your deliverable. They are never used to train AI models.
- Deleted after delivery
- Source documents and outputs are deleted within 7 days of delivery.
- NDA on request
- We're happy to sign yours before you send anything.
Send us 5 loss runs. We'll return them as structured data at no cost.
You choose the output format. There's no contract and no commitment, so you can judge the results on your own files.
Start a free pilot