You just got the OM on a new deal. The broker’s pro forma looks clean: tight expenses, growing revenue, a compelling NOI story. But you’ve been doing this long enough to know that “clean” sometimes means “curated.” The question isn’t whether the numbers add up. It’s whether they hold up against what the property actually did last year.
So you pull up the T12 and start mapping line items to the broker’s chart of accounts, category by category, trying to figure out where the broker’s projections diverge from trailing actuals and whether those gaps have any justification behind them. It’s not hard work. It’s just slow, tedious, and easy to lose focus on when you have six other deals in motion.
That’s exactly what this task is built to fix.
What This Task Does
You upload two things: the broker’s offering memorandum (or pro forma) and the property’s trailing 12-month operating statement. If the T12 is already summarized inside the OM, you only need the OM itself.
From there, the Real Estate Analyst (with Memory) takes over. It reads the OM, extracts the broker’s pro forma and assumptions, maps the T12 to the broker’s chart of accounts, and builds a comparison worksheet in Excel. Every line item gets a variance calculation, a note on the broker’s stated justification, and an analyst flag: Reasonable, Aggressive, Conservative, or Unsupported. It also calculates opex ratios for both the T12 and the broker’s numbers and tells you how much of the compression is actually supported.
The whole process takes roughly 10 minutes of your time. The AI does the rest.
Who This Task Is For
Every acquisition starts with a broker narrative. The numbers in the OM are the seller’s best case. Before you spend hours building your own underwriting model, you need to know where the broker’s story breaks down. This task gives you that answer fast.
This task is built for:
- Acquisitions analysts who need to scrub a broker pro forma against actuals before presenting a deal to their IC
- Principals and portfolio managers who want a quick read on whether a deal’s underwriting is grounded or aspirational
- Underwriters at lending shops who need to validate sponsor-submitted financials against trailing performance
- Asset managers evaluating dispositions who want to see how a broker is positioning the property’s financials relative to what actually happened
In short: if you already have an OM and a T12, this task gives you a line-by-line variance analysis with analyst-quality flags in minutes.
Why It Matters
Broker pro formas are marketing documents. They’re built to sell. That doesn’t make them wrong, but it means every line item deserves scrutiny, especially the ones where the broker projects savings or growth without explaining why.
You already know this. Anyone who’s underwritten more than a handful of deals has found the inflated rent bumps, the magically compressed expense ratios, the management fee that quietly dropped two points with no operational change to support it.
The problem isn’t that you can’t catch these things. It’s that doing it properly takes 20 minutes per deal, and when you’re screening ten OMs a week, that time adds up fast. Some deals slip through with less scrutiny than they deserve, not because you don’t care, but because the queue is relentless.
This task cuts that 20-minute scrub down to about 10 minutes. You upload the documents, the AI maps every line, flags every divergence, and hands you a formatted workbook. You spend your time reviewing the flags and deciding what matters, not manually lining up rows in a spreadsheet.
That’s the multiplier.
What the Output Looks Like
The Excel workbook generated by this task includes:
- A COA Mapping worksheet with every T12 line item mapped to the broker’s chart of accounts
- Broker Amount, Variance ($), and Variance (%) columns for every category
- Analyst flags (Reasonable, Aggressive, Conservative, Unsupported) on each line item
- Broker justification notes pulled directly from the OM, with analyst commentary engaging each claim
- An Opex Ratio Summary comparing the T12 and broker opex ratios with a characterization of supported vs. unsupported compression
The output is not a rough side-by-side paste job. It’s a formatted, flagged variance analysis, the kind you’d expect from a first-year analyst who actually read the OM before building the spreadsheet.
CRE Agents is a platform built for commercial real estate professionals who want to move faster without cutting corners. Task #[TASK_NUMBER] is just the beginning.
Frequently Asked Questions About Comparing Broker Pro Formas to Trailing Actuals With AI
Yes, and the task is designed with that expectation. The AI flags variances and provides analyst notes, but you are the one who decides what matters for a given deal. Spend a few minutes reviewing the flags, adjusting any notes that need more context, and confirming the mapping is accurate. The workbook is a first draft of the analysis, not a finished product. Your judgment is what turns it into something you’d put in front of your IC.
The output is a standard Excel workbook with clearly labeled columns, variance calculations, and analyst commentary. Nothing about it signals “AI-generated” unless you tell someone. It follows the same structure you’d build manually: T12 actuals vs. broker projections, dollar and percentage variances, and written notes on each line. The difference is that you got there in 10 minutes instead of 20, which means you have more time to refine the analysis before it reaches the committee.
The task works across property types. Whether you’re looking at multifamily, office, retail, or industrial, the logic is the same: map the T12 to the broker’s categories, calculate variances, and flag divergences. The AI references OER benchmarks relevant to the property type, so the analyst notes adapt to context. If you’re screening a pipeline of ten deals a week, you can run this on every one and spend your energy on the ones where the flags actually warrant deeper diligence.