Inheriting a model you cannot see into
A borrowed spreadsheet will happily produce a confident number from assumptions you never chose. If you cannot rebuild it, you cannot tell when it is lying to you.
Live class · Multifamily underwriting
Underwrite
Underwrite
with Claude.
A six-class series that builds multifamily underwriting from the ground up. Every line of the model, every assumption, and why it matters. Then it starts over, so you can join at any point and repeat anything that did not land.
The formatSix classes
- Live, with questions answered in the room
- Starts at the basics, assumes nothing
- Built on deals actually bought
- Bring a deal and we will use yours
The problem
A model that balances
A model that balances
can still be completely wrong.
Most people learn underwriting from a template somebody handed them. The formulas tie out, the summary page looks professional, and not one assumption underneath it has ever been tested.
Offering memoranda are marketing
Broker packages are written to be believed, not checked. The pro forma is a sales argument. Almost nobody rebuilds it from the rent roll and the T-12, which is exactly where the deal is won or lost.
Knowing the terms is not judgment
Knowing what a debt yield is and knowing which constraint actually binds on your deal are different skills. Videos teach the first. The second only comes from working through real numbers with someone who will push back.
Semi-seasoned is its own trap
You can fill in a model quickly, sound fluent in a meeting, and still not be able to defend a single line when an investor or a lender leans on it. That gap does not close on its own.
How it runs
Join at any class
6 classes
One full cycle,
then it restarts
then it restarts
2 weeks
Between live
sessions
sessions
From zero
No prior modelling
assumed
assumed
Your deal
Bring one and
we will use it
we will use it
The series runs on a loop. Class 06 is followed by Class 01, so there is no bad time to start and no penalty for sitting a class twice. Most people find the second pass is where it actually lands.
The syllabus
Six classes. Every part
Six classes. Every part
of an underwriting, in order.
Each class builds a section of the model and then tears it apart. You leave with the sheet you built, the reasoning behind every input, and a piece of homework against a real deal.
01Foundations
How a deal actually makes money
Before a single cell of a model. What underwriting is for, where returns genuinely come from, and the vocabulary that makes everything after this sound like plain English.
- What underwriting is, and what it is not
- The four sources of return: cash flow, amortisation, appreciation, tax
- NOI and cap rate, and why a dollar of NOI is worth fifteen to twenty of value
- Reading a rent roll for the first time
- Reading a T-12 for the first time
- GPR, loss to lease, vacancy, concessions, bad debt, EGI
- OpEx, NOI, DSCR, cash-on-cash, IRR, equity multiple
- Laying out the skeleton of your own model
- Homework: pull a live OM and label every number in it
02Revenue
Building a top line you can defend
The rent roll is where a deal is either real or imagined. We build revenue from unit-level data upward and keep what is in place strictly separate from what is hoped for.
- Rent roll forensics: unit mix, in-place versus market, lease expirations
- Month-to-month exposure and why it changes your risk
- Loss to lease and gain to lease
- Choosing comps: vintage, quality, amenities, location
- The comp adjustments nobody writes down
- Proving market rent instead of asserting it
- Other income: RUBS, pet, parking, laundry, fees, and what is really collectable
- Physical versus economic occupancy
- Vacancy, concessions, non-revenue units, bad debt
- Walking GPR down to EGI, line by line
- Homework: rebuild a broker's revenue page from the raw rent roll
03Expenses
Where most deals are actually lost
Revenue gets the attention, expenses do the damage. We normalise a real T-12 line by line and build the cost load a new owner will genuinely carry, not the one the seller enjoyed.
- Normalising a T-12: one-offs, owner add-backs, partial years
- Property taxes and reassessment on sale, the most common miss there is
- Insurance, and what the last few years have done to it
- Payroll, contract services, repairs and maintenance
- Turn costs and where they hide
- Utilities, and what RUBS does and does not recover
- Marketing, administration, and the management fee
- Replacement reserves, and what a lender will require
- Per-unit and per-square-foot benchmarks by vintage and market
- Why a flat two percent expense trend is a fiction
- Homework: normalise a T-12 and defend every adjustment you made
04Capital
Debt, equity and the capital stack
The same property underwrites to wildly different returns depending on how it is paid for. We size debt properly, then build the equity structure that sits on top of it.
- Agency, bank, bridge and debt fund, and when each is the right tool
- Sizing on LTV, LTC, DSCR and debt yield
- Working out which constraint actually binds
- Fixed versus floating, interest-only periods
- Rate caps, and what they cost now
- Amortisation, prepayment, yield maintenance, defeasance, assumption
- Sources and uses: everything that has to be funded at close
- LP and GP economics: preferred return, splits, catch-up, promote
- Acquisition, asset management and disposition fees
- Modelling a waterfall without getting lost in it
- Homework: size the debt three ways and find the binding constraint
05Business plan
CapEx and the value-add case
A value-add deal is a construction plan with a spreadsheet attached. We scope the work, price it from real numbers, prove the premium, and model how long it honestly takes.
- Scoping the work: interior, exterior, amenity, deferred maintenance
- Building a cost per unit from quotes, not rules of thumb
- Proving the renovation premium with comps
- Turn pace: how many units per month you can really deliver
- Downtime, turn cost, and the drag on revenue while you work
- Modelling the stabilisation curve month by month
- Sequencing capital: what gets spent when
- Contingency, and why five percent is rarely enough
- What happens to the plan when the premium does not appear
- Homework: build a renovation schedule and layer it into your model
06The decision
Returns, risk, and defending your number
The number at the bottom is the easy part. This class is about what that number hides, what has to stay true for it to hold, and how to defend it to somebody who disagrees with you.
- Choosing an exit cap, and why you expand it
- Hold period, and what changes when you are wrong about it
- IRR, equity multiple, cash-on-cash and average annual return
- What each of those metrics quietly hides
- Levered versus unlevered, and what the gap tells you
- Sensitivity tables: exit cap, rent growth, renovation premium, rate
- Break-even occupancy and break-even rent
- Stress testing: what has to go wrong before you lose money
- Writing the investment memo
- Presenting numbers you can stand behind
- Homework: underwrite a live deal end to end and defend it to the room
Class 06 is followed by Class 01, and the series begins again.
One to one
Bring an underwriting
Bring an underwriting
you have nearly finished.
This is a review, not a lesson. You bring a model you have already built and taken close to the finish line, and Claude goes through it with you: whether the underwriting holds up, which assumptions do not survive a second look, what is missing, and the things that are easy to miss on a deal like yours.
It is not a session for building a model from scratch, and it is not a shortcut around the class. If you have not underwritten a deal before, start at Class 01 and come back when you have something to test.
- 01
Send it ahead
Your model and the offering memorandum, a few days before the session. Far enough along that there is something real to test.
- 02
Reviewed before you meet
Claude goes through it in advance, so the two hours are spent on findings rather than orientation.
- 03
Two hours, line by line
Where the underwriting holds and where it does not, what is missing, and what would have to be true for the deal to work. Ask anything.
- 04
Leave with a list
A marked-up model and a specific list of what to fix, in the order it matters.
Who it is for
New to underwriting,
New to underwriting,
or stuck halfway.
If you have never opened a T-12, start at Class 01 and nothing will be assumed of you. If you already build models but cannot always say why an input is what it is, this is the part that usually gets skipped.
- Aspiring and new underwriters
- Analysts who inherited a model and want to understand it
- Operators moving from single-family into multifamily
- Passive investors who want to check a sponsor's numbers themselves
- Anyone who can fill in a template but cannot yet defend it
Questions first?
Infos@chazakinvestment.com +1 (347) 471-1540Tell us where you are with underwriting today and which class you were thinking of joining. We will tell you honestly whether it is the right starting point.
Underwrite with Claude is an educational programme. It teaches a method for analysing multifamily real estate and does not constitute investment, tax or legal advice, a recommendation to buy or sell any security, or an offer of any investment. Deals discussed in class are used as teaching material. Any figures shown are illustrative. You remain responsible for your own analysis and your own decisions.