The Complete Guide to Multifamily Deal Underwriting in 2026

Most multifamily deals don’t fail at closing. They fail at the underwriting desk — three months earlier — when an operator typed an aspirational rent into a spreadsheet they didn’t fully trust, and nobody caught it.

Underwriting is where multifamily money is made or lost. Yet for an asset class measured in the trillions, the day-to-day work of evaluating a deal is still done in fragile Excel files, passed by email, and re-keyed by hand. This guide is the framework we use at Del Val Investment Group — and the one that informs the technology we build through our affiliate Keptdo.

If you’re new to multifamily, this will give you a complete mental model. If you’re experienced, it’ll sharpen the checklist you already run.

What is multifamily underwriting?

Good underwriting answers three questions, in order:

1. What is this property earning today? (validated historicals)
2. What can it realistically earn under our ownership? (defensible pro forma)
3. Given the price we’re paying, what return does that produce — and what happens if we’re wrong? (returns + stress tests)

Everything else — broker decks, OMs, brochure photos — is noise.

The 7-step framework

Step 1 — Define your buy box before you look at a single deal

A working buy box has roughly seven dimensions:

• Markets: Specific MSAs (and the submarkets within them)
• Unit count: Minimum and maximum (e.g., 75–250 units)
• Vintage: Year-built range (1980+? 2000+?)
• Class: A, B, or C — and what condition each must be in
• Business plan: Stabilized, value-add (light/medium/heavy), or distressed
• Returns: Minimum stabilized cap rate, year-1 cash-on-cash, and 5-year IRR
• Hold period: Typically 3, 5, 7, or 10 years

Write this down. Share it with brokers. The discipline of saying no to the 90% of deals that don’t fit is the single biggest underwriting time-saver in the business.

Step 2 — Gather the T-12 and current rent roll

The two source-of-truth documents:

• T-12 (trailing twelve months): A line-item income statement for the past 12 months. This is your historical anchor — what the property actually earned, not what the broker projects.
• Current rent roll: A unit-by-unit snapshot of who lives where, what they pay, when their lease ends, and what’s vacant.

If either is missing, redacted, or “preliminary,” your underwriting is fiction. Ask for both before you spend more than 15 minutes on a deal.

Step 3 — Validate income line by line

Don’t trust the broker’s “Gross Potential Rent.” Build it yourself from the rent roll:

• Gross Potential Rent (GPR): Sum of every unit’s current asking rent × 12 months
• Loss to Lease: GPR minus actual in-place rent (the gap between asking and what tenants pay)
• Vacancy: What % of units are unoccupied right now, and what the trailing 12-month average is
• Concessions & bad debt: Discounts given and rent never collected
• Other income: Pet rent, parking, laundry, RUBS (utility reimbursement), late fees, application fees

The number you care about is Effective Gross Income (EGI) — gross potential less all the leakage. A T-12 will show you historical EGI; your job is to project forward EGI defensibly.

Underwriting truth: A deal that “works” at gross potential rent but breaks at actual collected rent is not a deal. It’s a broker’s pitch deck.

Step 4 — Validate expenses against industry benchmarks

This is where most underwriting goes wrong — by being optimistic on expenses.

Use these rough benchmarks (per unit per year, 2026 Sun Belt averages — adjust by region):

Expense category	Typical range
Property taxes	Variable — look up the millage rate, project tax reassessment at sale price
Insurance	$400–$1,200/unit (much higher in coastal/FL/TX)
Utilities (owner-paid)	$400–$900/unit
Repairs & maintenance	$400–$700/unit
Turnover costs	$500–$1,500 per turn × turnover rate
Payroll	$700–$1,400/unit
Management fee	3–5% of EGI
Marketing	$100–$250/unit
Replacement reserves	$250–$350/unit (lender will require this)
Total OpEx	Often 40–55% of EGI for stabilized B-class

If your underwriting shows a 32% expense ratio on a 1985-vintage B-class property, you’re wrong. Find the line items you underestimated and fix them now — not after closing.

Step 5 — Build the pro forma

The pro forma projects the property’s income and expenses across your hold period. For each year:

Effective Gross Income (EGI)
  - Operating Expenses (OpEx)
  = Net Operating Income (NOI)
  - Debt Service (P&I)
  - Capital Expenditures
  = Cash Flow Before Tax

For value-add deals, model the renovation explicitly: which units, what cost per unit, what rent lift, and over how many months. A “$15K interior renovation that lifts rent $250/month and turns over in month 18” is testable; a “value-add upside of 20%” is not.

Step 6 — Calculate the metrics that actually matter

The five:

1. Cap Rate = NOI ÷ Purchase Price. Tells you the unlevered yield. Compare to market cap rates for the submarket.
2. Cash-on-Cash (CoC) Return = Year-1 Cash Flow ÷ Equity Invested. Tells you your annual yield on equity.
3. DSCR (Debt Service Coverage Ratio) = NOI ÷ Annual Debt Service. Lenders typically require ≥ 1.20–1.25.
4. IRR (Internal Rate of Return) = the discount rate that makes the NPV of all cash flows zero. Time-weighted; accounts for the full hold + sale.
5. Equity Multiple = Total Cash Distributions ÷ Equity Invested. A 2.0x means you doubled your money over the hold.

A typical institutional-quality multifamily deal in 2026 targets: 5.5–6.5% going-in cap, 7–10% Year-1 CoC, 14–18% IRR over a 5-year hold, 1.8–2.2x equity multiple.

Step 7 — Stress test before you submit the LOI

If your model only works under one set of assumptions, you don’t have a model — you have a hope. At minimum, run three downside cases:

• Rent growth at 0% instead of your projected 3–4%
• Exit cap rate 75 bps higher than going-in (cap rate expansion)
• Vacancy 200 bps worse than current

If the deal still hits acceptable returns in at least two of those three cases, it’s a real deal. If all three break it, you’re paying too much.

The four mistakes that kill returns

After looking at hundreds of underwriting models, the same mistakes show up over and over:

1. Aspirational rents. Underwriting to “market” rents that the property has never achieved. Test: is there a comp within a half mile, same vintage, same condition, hitting that number?
2. Expense ratios that don’t match reality. A 30% ratio on a 1980s C-class property is fiction. Use real benchmarks.
3. Ignoring the tax reassessment. When you buy at a higher basis, the county will reassess. Model the new tax bill, not the seller’s.
4. No exit cap discipline. Most modelers use the same cap rate at exit as at entry. In a rising-rate environment, that’s wishful thinking. Default to 25–50 bps of cap expansion over the hold and let the deal earn its way out.

Why spreadsheets are the wrong tool for this

A multifamily underwriting model in Excel is a 12-tab, 2,500-formula house of cards. Every operator has stories: the broken VLOOKUP that hid an extra $200K of expenses, the assumption tab someone overwrote, the model that “worked” until it was opened on a different version of Excel.

Spreadsheets were not designed for this. They have no validation, no version control, no audit trail, no shared assumption library. They scale poorly with team size, and they get re-built from scratch on every deal because nobody trusts the last one.

This is exactly the gap that Multifamily Deal Analyzer Pro was built to close. Keptdo’s MFDA Pro takes the seven-step framework above and bakes it into an underwriting workflow that’s fast, opinionated, and auditable — so the same operator can underwrite five deals in the time it used to take to underwrite one, with more confidence in every number.

That’s the thesis behind Del Val Investment Group: the multifamily industry deserves operator-grade technology, and we’re building it.

Frequently asked questions

How long should it take to underwrite a multifamily deal?

A first-pass screen should take 15–30 minutes. A full underwriting — once you’ve decided the deal is worth the work — typically takes 4–8 hours for an experienced operator using a good template, longer if you’re building the model from scratch.

What’s a “good” cap rate for multifamily in 2026?

In most Sun Belt markets, B-class stabilized multifamily is trading at 5.5%–6.5% going-in cap rates. A-class core trades tighter (5.0%–5.75%); C-class value-add can trade at 6.5%–7.5%+ depending on condition and basis. Cap rates are local — always anchor to recent submarket comps, not national averages.

What’s the difference between a T-12 and a pro forma?

The T-12 is the trailing 12 months of actual operating results — historical fact. The pro forma is your forward-looking projection of what the property will do under your ownership — a model. Lenders and partners care about both: the T-12 tells them what the property has done; the pro forma tells them what you think it will do, and why.

What software do multifamily syndicators use to underwrite?

Most operators still use Excel templates passed down from prior firms. Purpose-built tools include Multifamily Deal Analyzer Pro from Keptdo, Archer, RealData, Spreadsheet Solutions templates, and the underwriting modules inside platforms like Yardi and AppFolio Investment Management. The right choice depends on deal volume, team size, and how much you value speed vs. customization.

What does “value-add” actually mean?

A value-add multifamily deal is one where the buyer plans to physically improve the property (unit interiors, common areas, exterior, amenities), raise rents to a renovated comp set, and exit at a higher value. The “value” being “added” is operational, not financial engineering. Light value-add typically means cosmetic upgrades ($5–10K/unit); heavy value-add can mean $25K+/unit and meaningful downtime.

What to do next

If you’re underwriting your first multifamily deal, start with the seven-step framework above — even a paper-and-pencil pass through it will catch most of the mistakes that sink real money.

If you’re underwriting your hundredth deal and tired of spreadsheets, take MFDA Pro for a spin: mfda.keptdo.com.

And if you’re building something interesting in this space — operator, founder, or investor — we’d like to talk.