Chapter 3: Build the Trusted Data Set

Build the Trusted Data Set

Reconcile the rent roll, T12, and any recent run-rate view into one input set, and log the unresolved issues that still block decision-ready underwriting.

Decision output: Reconciled Input Set And Unresolved Issues Log. Case requirement: state when file is not decision-ready.

Inputs

Inputs

  • Rent roll and T12 as the primary underwriting documents.
  • T3 annualized only when recent operations better reflect the current run-rate.
  • OM and broker notes for context, never as the source of truth.

Example card

Reconciling rent roll, T12, and recent run-rate

The rent roll annualizes to $3.42M of gross potential rent, but the T12 only supports $3.16M of collected rent because occupancy trailed and concessions spiked in the last two quarters.

  • Keep the T12 as the base proof of collections.
  • Use the rent roll to explain the upside path rather than overwrite the operating history.
  • Log unresolved mismatches so the file is not labeled decision-ready too early.

Analysis

How to analyze it

Reconciliation is the discipline of explaining why the operating files agree or disagree. The analyst should be able to point to concessions, occupancy movement, bad debt, or nonrecurring events instead of hand-waving around them.

  • Foot the rent roll and annualize it carefully.
  • Trace collections and concessions through the T12 and T3.
  • Create a visible unresolved-issues log if key lines do not reconcile.

Output

What output you should produce

Output artifact: Reconciled Input Set And Unresolved Issues Log

  • A reconciled underwriting input set with the preferred value for each key line.
  • A short issues log naming any mismatch, stale file, or missing support.
  • A plain statement about whether the file is decision-ready or still provisional.

Common mistakes

Common mistakes

  • Letting the OM overwrite the source documents.
  • Ignoring mismatches because the difference seems 'small enough.'
  • Dropping the unresolved issues log once the model starts to work.

Price meaning

What this means for price

If the trusted data set is weak, the price conclusion must be weak too. Unreconciled inputs are a reason to slow down the bid, not a reason to round into comfort.