Direct answer: Prepare trust-account CSV files by exporting the complete period from each authoritative system, preserving the original files unchanged, and validating account, cutoff date, opening and closing balances, required columns, transaction counts, unique identifiers, and client-ledger completeness. Normalize copies for calculation, not the raw evidence, and document every mapping or correction.
Key takeaways
- Export all three sources using the same cutoff date.
- Never edit the only copy of a source export.
- Use stable transaction and client or matter identifiers.
- Reject incomplete or ambiguous files before calculating.
Minimum useful fields
Bank and journal transaction exports usually need a date, amount, description, reference, and account identifier. The client-ledger source needs a stable client or matter identifier, ending balance, and enough detail to support the period activity or the firm's review process.
Column names can vary. Create a documented mapping rather than renaming or deleting evidence in place. Preserve original signs and formats alongside normalized values.
Validation before import
Confirm that the file is for the expected trust account and period. Check missing headers, blank identifiers, malformed dates, duplicated transaction IDs, impossible amounts, inconsistent currency, and unexpected row-count changes from the prior month.
For client ledgers, prove that the export includes zero-balance and inactive matters when the firm's rules require them, rather than assuming a filtered screen represents the full population.
Corrections and versions
If the source system contains an error, correct it in the authoritative system and create a new export. Do not patch the CSV to make the reconciliation work unless the firm's controlled process explicitly treats that change as a documented adjustment rather than source truth.
Name and retain versions so the packet can state exactly which exports produced the result. A source change after approval should trigger a new run and review.
Frequently asked questions
Can TriBalance import any CSV format?
The current workflow depends on supported schemas and documented mappings. Validate a sample export before relying on a new source system or custom report.
Should we remove client names before upload?
Data minimization is valuable, but the correct approach depends on the identifiers needed for reconciliation and the firm's security process. Use stable pseudonymous matter IDs when they support accurate review.
Sources and further reading
This article is operational education, not legal advice. Trust-account rules and retention requirements vary by jurisdiction. Confirm the requirements that apply to your firm and accounts.