The model ingests trailing twelve-month actuals through a mapping layer that tags each line item to a standardized revenue or expense category. Mapped figures flow automatically into the forward assumptions, eliminating manual re-entry and the transcription errors that come with it. Where the source data requires interpretation — partial periods, one-time items, or non-recurring expenses — manual override inputs allow line-level adjustments without breaking the automated flow. A built-in validation layer flags any discrepancy between the mapped T12 and the model's base-year figures, so the reconciliation can be confirmed before underwriting begins.