A proforma is not a forecast. It is an argument — a structured case that a building will produce a specific stream of cash at a specific level of risk, built from inputs you can defend line by line. Most proformas look complete on the surface. The income ties, the expenses footnote, the returns calculate. The problem is that the numbers that decide the deal are almost never the ones that get interrogated. This is a commercial real estate proforma explained as a decision tool: what every model should contain, what each field is actually stress-testing, and where the soft assumptions hide.
The four layers of a CRE proforma
Strip away the formatting and every proforma is the same four-layer stack. Income builds to a revenue line. Expenses subtract to net operating income. Debt sizes against that NOI and pulls out a levered cash flow. Returns translate that cash flow into the numbers an investor actually decides on. Get the order right and the model reads as one logical chain; get it wrong and you are reverse-engineering a return you already wanted.
The income build: what goes in a CRE proforma at the top line
The income side is where the deal looks its best, which is exactly why it deserves the most scrutiny. The mechanical fields here are cheap to verify and expensive to get wrong.
- Gross potential rent — in-place contract rent across every suite, tied to the rent roll, not a blended average. If you cannot reconcile it to the actual leases, you do not have a number, you have a hope.
- Square footage and lease term — rentable area, commencement and expiration dates, and renewal options. These are facts, pulled from a clean lease abstract, and they anchor everything downstream.
- Reimbursements — the recovery income that depends entirely on whether the lease is NNN, modified gross, or full-service. Misread the structure and you double-count or strand a major expense; the lease structure decision drives this line directly.
- Other income — parking, signage, storage, late fees. Small, but it should be sourced, not sprinkled in to round the number up.
- Vacancy and credit loss — the first true judgment call, deducted from gross potential to reach effective gross income. A 3% vacancy assumption on a market running 9% is not optimism, it is a thumb on the scale.
The expense stack: where NOI is quietly inflated
NOI is the number the whole valuation hangs on, so understating expenses is the most common way a proforma flatters a deal. The discipline here is to footnote every line against actuals and never assume away a real cost.
- Property taxes — reassessment on sale is the single most-missed line in CRE underwriting. Modeling trailing taxes when the basis is about to reset can overstate NOI by double digits.
- Insurance — quote it fresh. In hard markets premiums move faster than any other line, and last year's number is a liability.
- Utilities, repairs, and maintenance — driven by lease structure and asset condition. A clean roll-forward from operating statements beats a percentage-of-revenue guess every time.
- Management fee — charge it even if you self-manage. A proforma that omits management is not cheaper to run; it is mismarked.
- Replacement reserves and capital — TI, leasing commissions, and recurring capex belong in the model, not in a footnote. The deals that disappoint are usually the ones where capital was treated as someone else's problem.
Debt sizing and the returns layer
Once NOI is honest, debt and returns are mostly arithmetic — but the constraints are where lenders and equity actually live. The loan is sized to the more binding of two tests: loan-to-value and debt service coverage. Returns then translate the levered cash flow into the language of the investment committee.
- Loan amount, rate, amortization, and term — mechanical once you have a term sheet, and the levers a sensitivity table should flex.
- DSCR and debt yield — the coverage tests that decide how much leverage the NOI can carry. They double as a sanity check on the income build.
- Cap rate and stabilized value — entry economics, straight from NOI.
- Exit cap rate — the most consequential judgment call in the entire model. A 25-basis-point move at exit can swing IRR more than a full year of rent growth, and it is too often set equal to entry with no defense.
- IRR, cash-on-cash, and equity multiple — the return triad. None of them mean anything if the assumptions feeding them are not stress-tested.
Mechanical fields vs. the assumptions that hide the risk
Here is the line that separates a real proforma from a pretty one. Some fields are facts — they can be verified against a document and are either right or wrong. Others are judgment calls — they cannot be verified, only defended. Bad deals almost never die on the mechanical fields. They die on the assumptions nobody made the sponsor defend.
| Mechanical inputs (verifiable) | Judgment calls (defensible) |
|---|---|
| Contract rent and rent roll | Rent growth rate |
| Square footage and lease term | Vacancy and credit loss |
| In-place expenses from operating statements | Expense growth and reassessment |
| Loan rate and amortization (term sheet) | Exit cap rate |
| Reimbursement structure | TI/LC and downtime on rollover |
The mechanical column should be locked before anyone touches the model — that is the same fact-first sequence behind how to underwrite an industrial deal and a disciplined net-lease acquisition. The judgment column is where the analyst earns the work: every entry needs a stated basis — a comp, a market report, a lease clause — and a sensitivity range. An assumption you cannot range is an assumption you have not tested.
Proforma vs. actuals: the discipline that compounds
A proforma is a hypothesis. Actuals are the result. The teams that get sharper over time are the ones that close the loop — comparing what they modeled against what the asset delivered, line by line, after the fact. Proforma vs. actuals in commercial real estate is not a bookkeeping exercise; it is how you calibrate. If your vacancy assumption was 200 basis points light on the last three deals, that is a pattern, and the only way to see it is to have modeled every deal the same way.
The value of a canonical proforma schema is not the single deal — it is the portfolio. Run every acquisition through the same field set and you get comparability across deals, pattern recognition across cycles, and a defensible audit trail when an asset underperforms and an investor asks what you assumed and why.
That consistency is the whole point of a real underwriting standard. A one-off model in a borrowed spreadsheet answers one question once. A canonical schema — the same income build, the same expense stack, the same returns layer, every time — turns each deal into a data point in a larger system, and that system is what makes the next decision faster and the next argument harder to dispute. If you want the field-set checklist to run your own deals against, the underwriting toolkit is the place to start, and the same discipline carries straight into the offering memorandum and investment memo built on top of it.
Need this on a live deal? Capistrano produces underwriting, lease abstracts, investment memos, and capital-raise materials — AI-leveraged, principal-reviewed.