To check the numbers in an offering memorandum (OM), tie every headline figure to the document that should support it: NOI to the trailing-12 (T-12) statement, occupancy and average rent to the rent roll, and returns to the underwriting model. In one finished review of a 120-unit multifamily package, the OM's NOI was USD 1,312,400 and the T-12 it cited totaled USD 1,029,089, the stated 96.0% occupancy was 91.7% on the rent roll, and the model reproduced none of the OM's returns. This page walks through six checks on that package, so you can run the same ones on your own deal.
Written by the Titleman team with AI assistance (Claude). Reviewed by the Titleman review desk. The worked example is one finished Deal Review run on a Titleman QA account, read on 15 October 2026. It reviewed a sample package of three documents (an offering memorandum, a T-12 with rent roll, and a buyer's underwriting model) with no client data and no address. Every figure below is from that run or is arithmetic on its figures. This is a method for checking a package, not an appraisal, investment advice or a recommendation to buy or not buy.
The question
An OM is a seller's document. Its headline numbers (NOI, occupancy, rent, expense ratio, returns) are usually repeated in the buyer's model and in the lender's file. If one of them is off, the error is carried into the price, the loan and the return. So the question before you sign is: do the numbers in the OM tie to the documents behind them, and do the buyer's own documents tie to each other?
The run checked a package of three files. It found 1,453 figures, formulas and statements. Code checked all 568 formulas and the cross-file consistency, and the model checked the remaining 879. The result was 7 red flags and 34 items to review.
Six checks, with what each found
| Check | What the OM or model said | What the documents behind it showed |
|---|---|---|
| 1. NOI to the T-12 | USD 1,312,400 (6.08% on the USD 21,600,000 guidance price) | USD 1,029,089 (4.76%) |
| 2. Occupancy to the rent roll | 96.0% | 91.7% (110 of 120 units, 30 September 2026) |
| 3. Average rent to the rent roll | USD 1,515, above the USD 1,495 market average | USD 1,458.23, below market |
| 4. Model returns to the OM | 14.8% and 1.92x | 8.87% and 1.50x recalculated |
| 5. Coverage on the T-12 NOI | 1.27x on the OM NOI | 0.99x on the T-12 NOI |
| 6. Growth and vacancy assumptions | 3% rent growth a year, 5.0% vacancy and credit loss | 0.5% to 2.3% a year in the market forecasts, and 10.8% lost on the T-12 |
1. Tie the NOI to the T-12
Find the T-12 the OM cites and sum its NOI. In the run, the OM's trailing-12 NOI was not the NOI of its own T-12. The USD 283,311 gap was exactly the gap between the T-12's operating expenses of USD 961,866 and the OM's USD 678,555. The expense ratio shows it: 34.1% in the OM against 48.3% on the T-12.
A fast test is to add each document's expenses back to its NOI. Both documents landed on the same total income of USD 1,990,955, so the income agreed and the whole gap sat in expenses. That narrows the question to one thing to ask the seller: what is in the USD 283,311?
2. Tie occupancy to the rent roll
Count occupied units on the dated rent roll and divide by total units. The OM said 96.0% and "occupancy above 95% for eight quarters". The 30 September 2026 rent roll showed 110 of 120 units occupied, which is 91.7%. The T-12 vacancy line was 8.0% of gross potential rent for the year. The period provided did not bear out the claim.
3. Tie average rent to the rent roll
Average the in-place rents on the rent roll. The OM said USD 1,515 in place against USD 1,495 market. The rent roll showed USD 1,458.23, below market, which is what page 3 of the same OM says. When a memorandum disagrees with itself, the rent roll decides.
4. Recompute the model, do not read it
Open the buyer's model and check that each saved value matches its formula. In the run, the loan cell was saved at USD 15,120,000, although its own formula (65% of the price) gives USD 14,040,000, a difference of USD 1,080,000. The exit loan balance was a typed USD 13,049,000, not a formula. As saved, the model returned 11.27% and 1.68x. Recalculated, it returned 8.87% and 1.50x, or 8.76% and 1.49x with the amortized balance of USD 13,104,549. None of these reproduced the OM's 14.8% and 1.92x.
The model's in-place NOI of USD 1,290,000, labeled T-12, was 25.4% above the actual T-12 and matched neither document. Three documents gave three NOI figures.
5. Re-run coverage on the T-12 NOI
Debt service coverage is NOI divided by annual debt service. On the OM's NOI, a USD 14,040,000 loan at 6.25% (annual debt service USD 1,037,360) covered 1.27x. On the T-12 NOI it covered 0.99x, and the debt yield was 7.33%. A rate sheet of 1 September 2026 asked 1.35x at 65% loan-to-value, so the coverage test was not met. The 6.25% rate itself was cautious against the quotes of 5.70% to 6.00%. The gap came from the NOI, not the rate. For the full coverage worked example, see the NOI gap and DSCR page.
6. Test the assumptions against the market
Growth, vacancy and exit cap are the inputs the OM controls. In the run, 3% rent growth a year compared with market forecasts of 0.5% (2026), 1.0% (2027) and 2.3% (2028). Vacancy and credit loss of 5.0% compared with the 10.8% lost on the T-12. Both lift the projected NOI. The 5.75% exit cap sat inside the surveyed range of 5.18% to 6.86%, but it was 33 basis points below the OM's 6.08% and 22 basis points below the model's own 5.97% going-in cap, so it assumed compression. It was 99 basis points above the 4.76% yield on the T-12 NOI.
Smaller items that still matter
Not every finding is a headline. The run also found that the T-12's repairs total left out September's USD 5,796. Gross potential rent stepped up to USD 183,000 a month from April, while the 30 September rent roll priced the units at USD 179,400. September's T-12 rent after loss to lease and vacancy, USD 163,183, exceeded the rent roll's in-place rent of USD 160,405 by USD 2,778 a month, or USD 33,336 at an annual rate. A monthly non-revenue unit charge of USD 1,595 had no unit on the rent roll. These are the items a reviewer marks to review, not red flags, and they are why the T-12 is the better figure in the file but still not a final answer.
What to write down as not checked
A review is only as good as its stated limits. In this run no address was given in any file. So the records on the property (title, liens, court cases, permits, taxes and flood), the year built and site area, recorded sales, competitors and new supply could not be checked, and the comps map and competitor map could not be drawn. Lease files, the general ledger, the tax bill and the loan documents would settle the remaining open items. Write down which parts of your own package you tested and which you did not.
What this does not tell you
- That the seller acted in bad faith. A gap shows that two documents disagree. The explanation may be innocent, such as a one-time cost in the T-12.
- The right price. The checks tell you which numbers hold up, not what the property is worth.
- A forecast. The figures are trailing, and the market forecasts quoted are the run's inputs, not a prediction.
- A decision. The buyer, the lender and the committee decide.
For the full list of what a review flags, see what a Deal Review catches before you sign. For the document itself, see what an offering memorandum is, what a T-12 is and how to read a rent roll.
To have every material figure in your own OM, rent roll, T-12 and model marked verified, to review or red flag on a copy of your files, send one deal’s documents to Deal Review.
Frequently asked questions
How do you check the numbers in an offering memorandum? Tie each headline figure to the document that should support it. Rebuild the NOI from the T-12, the occupancy and average rent from the rent roll, and the returns from the underwriting model, then compare each to what the memorandum states. In the worked run above, the memorandum's NOI was USD 1,312,400 and the T-12 it cited totaled USD 1,029,089, and the stated 96.0% occupancy was 91.7% on the rent roll.
What should you compare the offering memorandum NOI to? Compare it to the T-12 operating statement the memorandum cites, and then to the buyer's own model. Add each document's operating expenses back to its NOI. If total income matches and expenses do not, the whole gap is in expenses. In the worked run both documents showed USD 1,990,955 of income, and the USD 283,311 NOI gap equaled the expense gap exactly.
Why do underwriting model returns not match the offering memorandum? Common causes are a saved input that no longer matches its formula, a typed number where a formula should be, and an NOI that matches neither the memorandum nor the T-12. In the worked run the model's loan cell was saved at USD 15,120,000 while its formula gave USD 14,040,000, and none of its outputs reproduced the memorandum's 14.8% return and 1.92x multiple.
What does a full check of an offering memorandum package cover? A full check covers every figure, formula and statement in every file, because the figures that matter are stated in more than one place. The worked run covered 1,453 figures, formulas and statements in three files, and code recomputed all 568 formulas. It reported 633 claims checked in three documents, with 7 red flags and 34 items to review.
Written by the Titleman team with AI assistance (Claude). Reviewed by the Titleman review desk. Figures come from one finished Deal Review run of a sample 120-unit multifamily package on a Titleman QA account, read on 15 October 2026. The 6.25% rate, the USD 14,040,000 loan and the 1.35x requirement are taken from the package and the run's rate sheet of 1 September 2026. Debt service, the 1,080,000 loan difference and the basis-point gaps are arithmetic on those figures. A review is not an appraisal, investment advice or a recommendation.