Every source row accounted for
Each row becomes a claim or a recorded exclusion with its reason. The report shows the reconciliation.
Bordereaux intelligence
TrueBind takes the bordereaux your partners actually send — any layout, any number of sheets — and turns them into validated, reconciled, explainable data your team can act on.
The engine
Upload any workbook — one sheet or twenty. File checks run first: type, size, structure. Macros never run.
Every sheet is inspected, header rows found below titles, and each source column matched to a Lloyd’s CRS field.
Required data, arithmetic to total incurred, dates, currency and status — row by row, with every source row reconciled.
Each finding says what happened, where it came from, the evidence and why it matters. Duplicates are told apart from development.
Decide, assign, follow up and export — every decision written to a hash-chained audit trail.
The product
Below is the report TrueBind produced for a synthetic 450-row, three-sheet test workbook we use to exercise the engine. The numbers are the engine’s real output for that file. They are not customer data.
Grade 5 / 5 · Excellent
450 claim rows across 3 of 3 sheets were assessed and every source row is accounted for. 6 exact resubmissions found. 6 claims show development from a previous period and are not treated as duplicates.
Measured in our test environment: that workbook went from upload to a rendered report in 3–5 seconds, including the mapping confirmation.
Trust
Each row becomes a claim or a recorded exclusion with its reason. The report shows the reconciliation.
Duplicates are flagged for a person to confirm, dismiss or send back. Source data is never altered.
Column mapping may ask a model about header text only. Cell values are never sent. Calls are capped and time-boxed.
Every mapping, decision and export is hash-chained. Anyone can verify the chain is intact.
Tenant isolation in the application and PostgreSQL row-level security underneath it.
CSV exports are guarded against formula injection; uploads are checked before anything reads them.
Automation
The same audited pipeline runs for every file. Web upload and API intake work today; e-mail intake is planned and marked as such.
Create an organisation, upload a workbook and walk through mapping, validation and the report yourself.