How to Build a Multifamily Pro Forma From Scratch

How to Build a Multifamily Pro Forma: Start With the Rent Roll, Not the Template

The offering memorandum lands Friday afternoon. Ninety pages, a glossy cap-rate summary on page four, and a rent roll buried in the appendix that does not quite match the broker's numbers on page four. You could open a template someone forwarded you two jobs ago and start typing over its formulas. That is how errors get inherited — a stale expense ratio here, a vacancy assumption from a different market there, formulas you did not write and cannot fully audit under a Monday deadline.

Building a multifamily pro forma from scratch is not about typing faster. It is about knowing, cell by cell, why each number is where it is — so when a lender or an LP asks "where does this rent number come from," you can answer without scrolling. This walkthrough builds one apartment pro forma from a blank workbook, tab by tab, using a single illustrative 24-unit property carried through every section. Every dollar figure below is a labelled example input, not a market data point.

By the end, you will be able to construct a unit-mix schedule, bridge in-place rent to market rent, build a defensible operating expense schedule, and lay a debt layer on top of net operating income — the full mechanical spine of an apartment underwriting model.

Unit Mix and the In-Place Rent Schedule

Every multifamily pro forma starts with the rent roll, because the rent roll is the only tab where the model touches reality — actual leases, actual tenants, actual dollars currently being collected. Everything else in the workbook is either a bridge from that reality (market rent, loss to lease) or a projection built on top of it (operating expenses, debt service, exit value).

Build the unit-mix tab first, before you touch a single revenue or expense formula elsewhere in the workbook. For the illustrative 24-unit property carried through this walkthrough, the unit mix might look like this as an example:

• 12 one-bedroom units at 650 square feet
• 12 two-bedroom units at 900 square feet

Each row on the unit-mix tab should carry, at minimum: unit number, floor plan, square footage, current tenant status (occupied or vacant), lease start and end date, and in-place rent — the actual rent currently being paid or contracted, not an assumption. This is the tab that eventually feeds a proper rent roll template, and it is worth building it as its own structured table (not a loose list) so that every downstream formula can reference it by row rather than by hardcoded value.

As an illustrative example only, suppose the 12 one-bedroom units carry an average in-place rent of $1,050/month, and the 12 two-bedroom units average $1,280/month. Multiply each unit type by its unit count and by twelve months to get an annualized in-place gross rent figure for that floor plan — then sum the floor plans for total in-place gross potential rent. This is a mechanical rollup, not a judgment call; the judgment happens in the next section.

In-Place vs. Market Rent: The Loss-to-Lease Bridge

This is the section where a multifamily pro forma either becomes an underwriting tool or stays a rent-roll transcription. In-place rent tells you what the seller is currently collecting. Market rent tells you what a comparable, freshly-leased unit in the same submarket would likely command today. The gap between the two — loss to lease — is very often the entire value-add thesis of the deal.

Add a market-rent column next to the in-place-rent column on the unit-mix tab, one row per unit or one row per floor plan if you are working at that level of aggregation. The market rent figure should come from comparable-lease research — recent leases signed at comparable nearby properties for a comparable floor plan and condition — never from the seller's own pro forma, which has an obvious incentive to lean optimistic. Confirm current comparable-lease data with your own market research or a local leasing broker; this article does not supply market rent figures as fact.

Build the bridge as its own explicit calculation rather than burying it inside a single "revenue" line:

1. In-place gross rent — the sum from the prior section.
2. Market gross rent — same unit count, at market rent per unit.
3. Loss to lease — market gross rent minus in-place gross rent, expressed both in dollars and as a percentage of market gross rent.

As an illustrative example, if the in-place one-bedroom average is $1,050/month against an illustrative market-rent assumption of $1,150/month, the loss to lease on that floor plan alone is $100/unit/month, or $1,200/unit/year — again, an example input, not a market figure. Multiply by unit count and repeat for the two-bedroom floor plan, then sum both lines to get a property-level loss-to-lease dollar figure and percentage.

The loss-to-lease percentage is the single number a value-add buyer should be able to defend in one sentence: this is the rent growth achievable simply by re-leasing at turnover, with no capital improvement required.

Keep two rent lines running in parallel through the rest of the model — an "in-place" scenario and a "market" or "stabilized" scenario — rather than collapsing to one number too early. A lender underwriting the loan today generally wants to see in-place income; an equity partner underwriting the business plan wants to see the stabilized, market-rent case and the timeline (typically tied to unit turnover and lease expirations) it takes to get there. This distinction, and how quickly a buyer can actually capture it, varies by market and by lease structure — treat it as a modeling convention to confirm against the specific deal's lease file, not a fixed rule. For a broader framework on separating a property's current-year numbers from its projected numbers, see the general walkthrough on how to build a real estate pro forma.

Vacancy, Credit Loss, and Other Income

Gross potential rent — whichever scenario you are running — is not the same as collected income. Two deductions and one addition sit between gross potential rent and effective gross income (EGI).

Vacancy loss accounts for units sitting empty between tenants. Build it as a percentage applied against gross potential rent, but calculate the percentage from the property's own trailing occupancy history where available, rather than typing in a round assumption. If the rent roll shows the actual number of vacant units on the survey date, that gives you a spot vacancy rate; a trailing twelve-month operating statement, if the seller provides one, gives you a more defensible average vacancy rate to underwrite against. Where you cannot get either, state the assumption plainly as an example input and flag it for the reader or your own credit committee to confirm against local leasing velocity, rather than presenting it as researched fact.

Credit loss (sometimes bad debt) accounts for rent billed but never collected — evictions, skips, unpaid balances written off. This is typically a smaller percentage than vacancy loss and is often estimated from the property's own historical write-offs if the seller discloses them.

Other income flows the opposite direction — it adds back to gross potential rent before you land on EGI. Common multifamily other-income line items include:

• Parking fees
• Pet fees and pet rent
• Laundry income
• Utility reimbursement (RUBS — ratio utility billing system)
• Application and administrative fees
• Storage unit rental

Build other income as its own itemized schedule rather than a single lump "other income" line, even if some rows are small. An LP or lender reviewing the model will want to see what is driving that number, and a single unlabelled line reads as a plug.

The EGI formula, once all three pieces are built, is straightforward:

EGI = Gross Potential Rent − Vacancy Loss − Credit Loss + Other Income

Run this formula for both the in-place scenario and the market/stabilized scenario, keeping the two columns side by side so the loss-to-lease bridge carries all the way through to EGI, not just to gross rent.

Building the Operating Expense Schedule

The operating expense tab is where a from-scratch pro forma either earns trust or loses it. A single "opex ratio" line — expenses as a flat percentage of EGI — is fast to build and impossible to defend to a lender's underwriter, who will ask about specific line items, not a blended ratio.

Build the expense schedule line by line instead, with each major category as its own row:

• Property taxes
• Property insurance
• Utilities (whatever portion the owner pays, net of RUBS reimbursement)
• Repairs and maintenance
• Property management fee (commonly a percentage of EGI — build it as a formula against the EGI cell, not a hardcoded dollar amount, so it flexes correctly between the in-place and market scenarios)
• Payroll (on-site staff, if any)
• Marketing and leasing costs
• General and administrative
• Landscaping and grounds
• Pest control
• Trash removal

For each line, calculate a per-unit-per-year figure — total expense divided by unit count — and sanity-check it against the property's own trailing operating statement if one exists, or against comparable properties you have underwritten before. Per-unit expense benchmarks vary meaningfully by market, property vintage, and whether utilities are separately metered; treat any per-unit figure you use as a working assumption to confirm with the seller's operating history, a property manager, or your own portfolio data — not as a fixed industry rule. Property tax specifically deserves its own line of scrutiny: many jurisdictions reassess on sale, so the seller's trailing tax expense may understate what the buyer will actually pay in year one. Confirm the post-sale reassessment treatment with the local taxing authority or a property tax consultant before finalizing the line.

Sum all line items to total operating expenses, then calculate the ratio against EGI as a check figure — not as the input, but as the output you review for reasonableness once the line-by-line build is complete.

From EGI to NOI, Reserves, and Debt Service

NOI = EGI − Total Operating Expenses. This is the formula that everything else in commercial real estate underwriting is built on top of — cap rate, DSCR, debt yield, valuation. It deliberately excludes debt service and capital expenditures; those sit below the NOI line, not inside it, because NOI is meant to describe the property's unlevered operating performance independent of how any particular buyer chooses to finance it.

Add a capital reserves line just below NOI, before debt service — commonly built as a fixed per-unit-per-year figure or a percentage of EGI, set aside for non-recurring capital items like roof and mechanical replacement. Lenders frequently require a minimum reserve contribution as a loan condition, though the required amount and the calculation method vary by lender and by loan program — confirm the specific reserve requirement with your lender rather than assuming a standard figure.

With NOI and reserves built, add the debt layer:

1. Loan amount — driven by whichever is more restrictive between a maximum loan-to-value (LTV) constraint and a minimum debt service coverage ratio (DSCR) constraint. Build both tests as separate formulas and let the model flag which one binds.
2. Debt service — principal and interest on the loan amount, at the interest rate, amortization period, and term you are underwriting. Build this from an amortization formula referencing the loan terms as their own labelled input cells, not a hardcoded annual number, so the model recalculates cleanly if the rate or loan amount changes.
3. DSCR = NOI ÷ Annual Debt Service. Run this for both the in-place and market/stabilized NOI scenarios — a deal that clears a lender's minimum DSCR on stabilized income but not on in-place income tells you something specific about the transition risk between closing and stabilization.
4. Cash flow after debt service = NOI − Reserves − Debt Service. This is the levered cash flow line that ultimately feeds equity returns.

Interest rates, required DSCR minimums, and maximum LTV all vary by lender, loan program, and prevailing credit conditions at the time of the loan — treat every rate and coverage threshold in your model as a confirmed quote from an actual lender before you rely on it, not as a general assumption.

What the Finished Pro Forma Tells You

A multifamily pro forma built this way — rent roll first, in-place-to-market bridge explicit, expenses itemized, debt layered last — does more than produce a single NOI number. It produces a model you can defend line by line, in a lender call or an LP meeting, because every figure traces back to either an actual lease, a comparable-rent input you sourced yourself, or a labelled assumption you can point to and revise.

That traceability is also what separates a from-scratch build from a template with the formulas locked or hidden: if you cannot see or edit the loss-to-lease formula, you cannot defend it, and you cannot adapt it when the next deal's rent roll looks different from this one. For a deeper look at the rent-roll structure itself, see the rent roll template walkthrough, and for the fully built-out version of this exact tab structure — unit mix, loss-to-lease bridge, itemized opex, and the debt layer, unlocked and documented — see the apartment pro forma template and multifamily pro forma Excel resources.

Building this from a blank workbook is the right exercise once, so you understand every formula. For live deals under a Friday-afternoon deadline, the Multifamily Acquisition Pro Forma carries this exact structure — in-place versus market rent, itemized operating expenses, reserves, and the debt layer — fully unlocked and documented, so you can underwrite the next rent roll without rebuilding the spine from scratch.