Sampling measures the ordinary
Statistical sampling is built to estimate how common something is across a population. Forensic exposure is usually not common. It is concentrated, sustained, and structurally excluded from the frame the sample was drawn on.
An audit that finds nothing has often only established that the sample could not reach the problem.
Full population or nothing
Modern record volumes no longer require sampling for feasibility. Examining the full population is now the disciplined choice, and it converts an estimate into a set of specific, nameable records.
PERISCOPE briefings describe investigative principles. They do not disclose detection logic, thresholds, formulas, weighting or rules.