Platform

One connected workspace, from records to a defensible conclusion.

ForensicTrail turns bank and credit-card statements into a source-linked ledger, reads paystubs and tax records alongside them, routes anything it isn't confident about to a human reviewer, and carries the reviewed work into analysis and court-ready work product.

How it works

Four jobs, one connected workspace.

  1. Intake and organize

    Upload bank and credit-card statements, paystubs, tax returns, and other supporting records to a case. Extraction reads tables and text, splits combined statements into separate accounts, and only accepts a file once its transactions reconcile against the statement's printed balances — a mismatch holds it for review instead of passing it on to classification.

  2. Investigate money movement

    Trace transactions across accounts, follow transfers between them, and see forensic flags on structuring, round amounts, and other risk indicators — each transaction extracted from a statement stays linked back to the page it came from.

  3. Review and substantiate

    Transactions the system isn't confident classifying wait in one case-wide human review queue, where a reviewer approves, corrects, or excludes each one and can turn a correction into a rule that applies retroactively. Every decision is recorded.

  4. Report with evidence

    A reviewed claim, its supporting evidence, and any counter-evidence are saved as a source-linked Finding; transactions extracted from a statement trace back to the page and Bates label they came from.

Follow the sample case from flag to Finding

Where AI works

The same four jobs, and what AI does in each.

  1. Intake and organize

    AI
    Works out how to read unfamiliar statement layouts and itemizes paystub lines; a paystub itemization is kept only when it matches the printed totals.
    Your expert
    You resolve any statement whose transactions don't add up to its printed balances; it's held instead of used.
  2. Investigate money movement

    AI
    Sorts transactions into categories where your firm's rules and known merchants leave off, and the case assistant answers questions with citations.
    Your expert
    You follow flagged transactions to their source pages and decide what they mean.
  3. Review and substantiate

    AI
    Takes a second look at large, flagged transactions it categorized; a disagreement goes to your review queue.
    Your expert
    You review anything the AI isn't sure about and approve, correct, or exclude it; an exclusion carries a reason.
  4. Report with evidence

    AI
    A case-assistant answer you pin arrives with its citations and counter-evidence attached, as a draft for you to edit.
    Your expert
    You edit the claim, save the Finding, and own the conclusions.

Bring a matter. See the trail.

Walk through the workflow with our team, or explore the fictional sample case first.