Skip to main content
Case studies

The deal looked fine. Then someone ran the second number.

Four desks, four files, and the same move every time: run the deal twice and see whether the two answers agree. Names and figures are illustrative; the math is what the calculator actually runs.

Four worked examples · modeled, not customer outcomes
$0
Diligence cost these examples put at risk
$3,000–7,000 a file, committed before anyone reads the rent roll
0×
Throughput, same team
six minutes a file instead of twenty-five
0
Files that pass on payment, fail on cash
Track 1 clears 1.00x — Track 2 doesn't
The pattern behind every win

One number gets you the loan. The other tells you whether to take it.

Every scenario below turns on the same move: Greenstreet runs two DSCRs on one file. Track 1 is what the lender funds on. Track 2 prices in vacancy, management, and CapEx — whether the deal actually survives. When they disagree, Track 2 is the reason a deal got stopped before it cost a dollar.

Same property · same week · one application
Track 1 · Lender-qualifying DSCR
0.00xFunds
1.00 break-even

Rent covers the note. Most desks stop reading here.

Track 2 · Investor-survival DSCR
0.00xStops
1.00 break-even

Price in a 12% vacancy and the same deal goes underwater.

The deal, the catch, and the number that settled it.

These are constructed teaching examples, not customer histories. Names, quotes, figures, approvals, timelines, and outcomes are hypothetical. Individual results will vary.

40 doors · 4 marketsBlended DSCR1.11x$18M balance
04
Aurora·Portfolio operator · 40 doors

Forty doors, six years of one-off financing, one blended number: 1.11x.

1.11x blended on an $18M balance, with two of the four markets doing most of the lifting. It took about a week to assemble, and it is the sheet Aurora now brings to a lender instead of a folder.

$18M
Balance across 40 doors
1.11x
Blended coverage, whole book
1 wk
Spreadsheet stack to one view
Track 1Track 21.09x
01
Vela Capital·Broker desk · 12 agents

Two spreadsheets per file, 120 files a month. They deleted one of them.

A file that took twenty-five minutes to assemble now takes about six. Same team, roughly four times the files — that is arithmetic, not a promise, and it holds only as long as the rent roll going in is clean.

4×
Files per underwriter, same team
6 min
Per file, down from 25
120+
DSCR files a month
1.00 — break-even1.18xTrack 112% vacancy0.98xTrack 2
02
Northshore Non-QM·Multifamily investor

Track 1 said 1.18x. Track 2 said 0.98x. One vacancy assumption was the whole difference.

0.98x. Two hundredths under break-even, same building, same week. Track 1 would have funded it. Track 2 is the reason they walked before the deposit went hard.

1.18x
Lender-qualifying DSCR
0.98x
Same file, 12% vacancy priced in
12%
What the submarket actually runs
$14,800diligence not spent
03
Quintero & Co.·Foreign-national & ITIN files

Three deals died at the desk — before the appraisal invoice arrived.

All three cleared Track 1 and landed under 1.00x on Track 2. At $3,000–7,000 of diligence a file, that is $14,800 that stayed in the account. Each screen took about three minutes.

3
Files stopped pre-appraisal
$14,800
Diligence cost not spent
3 min
To screen each one

Ready to run your own deal through the same engine?