Verified multifamily underwriting

Underwriting numbers
you can defend.

PropLens extracts your T-12s and rent rolls, then checks every extraction against the totals printed in the document itself — and shows you exactly what matched and what didn’t.

Free during beta · No credit card · Your data never trains AI models

Validated across 280+ real PMS documents

Rent Roll

rent-roll.pdf · 234 units

234/234 units — matches document summary
UnitTypeSq FtMarketCurrent
1011BR/1BA7421,2951,245
1022BR/2BA1,0581,6501,650
1031BR/1BA7421,2951,180
104Studio521995975
+ 230 more units
  • UnitsExtracted234Document234
  • Sq FtExtracted198,450Document198,450
  • Market RentExtracted312,480Document318,240

PropLens checks every extraction against the totals printed in the document itself.

Upload. Verify. Underwrite.

From raw operating statement to a defensible worksheet in three steps.

Step 01

Upload

Drop in your T-12 and rent roll exactly as the PMS exported them — Excel, CSV, or PDF. Excel files process in under 30 seconds; PDFs take 1–3 minutes.

Step 02

Verified extraction

PropLens extracts every line item and unit, then compares its own totals against the totals printed in the document. Matches are marked verified; differences are flagged in amber for review.

Step 03

Editable worksheet

Review NOI, occupancy, loss-to-lease, and lease expirations. Click any cell to adjust it — your edits layer on top of the extracted data and can always be reverted. Export to Excel when you're done.

Extraction you don’t have to take on faith

Most documents already state their own totals — the unit count on a rent roll summary, the income and expense totals on a T-12. PropLens reads those too, and reconciles its extraction against them on every run.

  • Rent rolls: unit count, total square footage, and market rent checked against the document's summary rows.
  • T-12s: total income, total expense, and NOI checked against the statement's own grand totals.
  • Anything that doesn't reconcile is flagged in amber — on the dashboard and on the affected worksheet totals.
  • When no comparable total can be found in the document, PropLens says exactly that instead of implying a check it didn't make.

The spec sheet

Straight answers to the questions analysts actually ask.

Which property management systems are supported?

Yardi, RealPage, OneSite, Entrata, MRI, AppFolio — and ad-hoc spreadsheets that didn't come from any PMS at all. If the columns are recognizable, PropLens will find them.

What file types can I upload?

.xlsx, .xls, .csv, and .pdf — including scanned PDFs. Excel and CSV files usually process in under 30 seconds; PDFs take 1–3 minutes because we extract the tables page by page.

Can I edit what the AI extracted?

Yes — every cell in the worksheet is click-to-edit. Your edits are saved as overrides layered on top of the extracted data, so nothing is ever overwritten: you can revert any single cell, or all of them, back to the original extracted value at any time.

How do I know the numbers are right?

PropLens checks its own work. Every extraction is compared against the totals printed in the document itself — unit count, square footage, and rent totals for rent rolls; income, expense, and NOI totals for T-12s. Matches are marked verified; any difference is flagged in amber for your review instead of being hidden. When we can't find a comparable total in the document, we say so plainly instead of implying a check we didn't make.

Is my data used to train AI models?

No. Never.

Run your next deal through PropLens

Upload the T-12 and rent roll you’re working on right now and see every number checked against the document itself.

Free during beta · No credit card · Your data never trains AI models