Insights
Case studies and perspectives from the CALMO team on legacy data, evidence and deadlines.
Featured case study
About 230 million records and 1 million tax documents consolidated into a single database, with zero discrepancies against the source and assessments delivered in days.
Perspectives
A calculation table may run up to a given year while actual payroll ended earlier, leaving only later residues. Looking only at the database structure suggests the full period is there; measuring reveals the gaps.
That is why, during validation, we measure the first and last period with real data in each source and report gaps to the partner before the assessment. The difference between the two coverages often decides whether a legal theory is viable.
Computing the hash only at the destination proves the file did not change after arrival — not that it arrived identical to what existed on the client's server. A SHA-256 computed at source, recorded in a manifest and checked at destination closes that gap.
Together with the collection record, the custody log, timestamps and immutable retention, hashing at source allows a third party to verify, years later, that a figure in the report comes from exactly that data.
Under Brazilian law, the period to claim a refund of overpaid taxes is, as a rule, five years. Every month without the data, older periods become unrecoverable — and tax assessments are upheld for lack of workpapers.
Old servers do not wait either: disks and memory at their limit are a signal to prioritize collection. Starting with a read-only diagnostic is the safest way to know what can still be proven.
How can we help?
Planning is built backwards from the due date.