EBITDA Definition and Application in CRE Loan Analysis

Stripping out those four line items approximates the cash-generating engine that produces the business's cash flow, independent of the capital structure layered on top of it, though some lenders add back only D&A to NOI, arriving at a figure between the two.

There are two ways to arrive at the number, and CRE professionals should be comfortable moving between both. Start from net income and add back interest, taxes, depreciation, and amortization, or start from operating income (EBIT) and add back only depreciation and amortization, since interest and taxes have already been excluded at that stage.

EBITDA is a non-GAAP measure that companies may calculate in different ways. That's a detail sponsors gloss over at their own risk, and the covenant section further down explains why.

Lenders use it to estimate how much cash a borrower has available to service principal and interest. Investors use it to compare businesses that carry different debt loads and sit in different tax jurisdictions, on a basis that isn't distorted by financing decisions. Business owners lean on it as a proxy for operating profitability when raising capital or preparing to sell. Analysts use it as the foundation for enterprise-value multiples in M&A work.

The add-back of depreciation and amortization is what pulls EBITDA closer to actual operating cash generation than net income ever gets, since neither expense involves cash leaving the business in the current period. EBITDA shows up as the standard profitability yardstick across manufacturing, technology, healthcare services, and consumer products, and it functions as the common language of business lending and M&A everywhere outside of real estate. Real estate, as the next section explains, built something else. EBITDA = Earnings Before Interest, Taxes, Depreciation, and Amortization (each excluded item reflects financing choices or accounting conventions, not operating reality). In a worked example, $2M net income plus $400K interest plus $600K taxes plus $500K D&A equals $3.5M EBITDA, and the same business yields a very different number depending on where you start (Supermoney). EBITDA is not a GAAP measure and has no standardized definition, so companies may calculate it differently; per Thomson Reuters Practical Law, the definition is often negotiated in loan agreements to include carve-outs and add-backs.

Why real estate developed its own metric distinct from EBITDA

Net operating income is calculated by subtracting operating expenses from gross operating income, and it sits entirely at the property level: before financing costs, before taxes, before capital expenditure, before depreciation or amortization ever enter the picture. On the surface it looks like a real estate cousin of EBITDA. In practice, the two metrics diverge in ways that matter enormously once a lender starts sizing a loan.

The first divergence is structural. NOI builds vacancy loss into gross operating income from the very first line, because a property's income statement is built around occupancy and rent collection in a way a corporate income statement isn't designed to capture. A business income statement has no equivalent concept, since a company doesn't have "vacant units," it has customers it either sold to or didn't. The second divergence involves interest and taxes. Some lender formulations of NOI actually include interest and taxes rather than excluding them, making NOI, in those formulations, more conservative than EBITDA, which strips all four items out without exception. Some lenders split the difference by adding back only depreciation and amortization to NOI, landing on a number between the two conventional metrics entirely.

The logic behind NOI mirrors the logic behind EBITDA, even if the calibration differs. Two investors who own the identical property should show identical NOI regardless of how each financed the deal or what their personal tax situation is, because NOI isolates the performance of the asset rather than the investor. That's the same instinct that drives EBITDA in the corporate world, just recalibrated to property cash flows instead of business income statements. The valuation math runs in parallel too: NOI divided by a market cap rate produces a property value, in the same way that EBITDA multiplied by an industry-appropriate multiple produces business enterprise value. Willowdale Equity warns that the same framework with different inputs, when a sponsor treats the two metrics as interchangeable without adjustment, invites systematic errors into valuation and comparison work, a real underwriting risk that deserves close attention.

NOI remains the default for standard income-producing property: multifamily, office, industrial, the properties CRE professionals underwrite every day. EBITDA enters the conversation the moment an asset stops behaving like a passive collection of leases and starts behaving like an operating business that happens to sit inside a building. Where exactly that line falls is the subject of the next section.

The asset types where lenders reach for EBITDA instead of NOI

The dividing line comes down to where the revenue actually originates. If it comes from lease payments on a rent roll, NOI does the job. If it comes from an operating business embedded inside the real estate, one where staffing, service quality, and management decisions drive the top line as much as the address does, lenders start reaching for EBITDA instead.

Hotels sit at the center of this category. Hotel revenue moves with occupancy, average daily rate, and RevPAR, all of which swing seasonally and can't be captured by anything resembling a static rent roll. Senior housing and skilled nursing operate the same way: margins depend on census levels, staffing costs, and reimbursement rates, so a lender underwriting one of these deals is really underwriting a healthcare operation that happens to occupy a commercial real estate asset. Self-storage falls into the category too, since it's management-intensive and driven by active revenue management, meaning EBITDA margins tell a lender far more about operational quality than a lease abstract ever could. Retail properties anchored by operator-tenants on percentage-rent leases round out the list, since the tenant's own EBITDA health, strong or weak, dictates whether that lease survives the next downturn.

SoFi's 2026 guidance on this point is blunt: banks decide between NOI and EBITDA based on the borrower's industry before they ever approve a loan, and that decision is a matter of lender judgment applied deal by deal, not a fixed formula anyone can look up in advance. That variation appears across lender types too. Some banks add back depreciation and amortization to arrive at an inflated NOI figure, while others work directly in EBITDA terms, and a sponsor walking into a financing conversation should ask, explicitly and early, which metric the lender is actually underwriting to.

A leveraged-buyout parallel applies here as well. In classic leveraged transactions, debt financing gets sized as a ratio of debt to EBITDA, and CRE professionals acquiring operating-heavy assets through structured acquisitions will run straight into that exact framework, encountering a corporate-style debt sizing conversation regardless of expectation. The practical consequence for sponsors is that financing packages for these asset types need to include EBITDA presentation alongside, or instead of, the traditional rent roll, and that documentation burden is materially heavier than what a stabilized multifamily or office deal requires.

The three metrics lenders use to size a CRE loan, built from EBITDA

Once a lender settles on a normalized NOI or EBITDA figure, three sizing tests determine how large the loan actually gets. Loan-to-value divides the loan amount by the property's value, where value itself is derived by dividing NOI or EBITDA by a market cap rate and later confirmed by appraisal, so the quality of the EBITDA figure sets the denominator of the entire equation. Debt service coverage ratio divides NOI or EBITDA by annual debt service, functioning as the direct cash-flow test. Debt yield divides NOI or EBITDA by the loan amount itself, and its value lies in being rate-agnostic, stripped of any sensitivity to where interest rates happen to sit.

The DSCR benchmarks in active use vary by lender category. Conventional banks typically hold to a 1.20x to 1.25x minimum, and commercial real estate lenders generally run in the same 1.20x to 1.30x range, with added weight placed on NOI stability across the life of the loan Commercial Observer Clear Value Lending Supermoney.

Debt yield earns its place as a separate test because DSCR is hostage to interest rates in a way that has nothing to do with how well the property or business is actually performing. A deal underwritten at a 1.35x DSCR when rates are 6.25% can fall to 1.10x purely on a refinance at 7.50%, with zero change in the underlying operating performance DDEE.ai. Debt yield exists precisely to give lenders a rate-neutral read on that same cash flow.

None of these three tests function correctly if the input feeding them is wrong. Sponsors entering leveraged-transaction territory should also know, ahead of the conversation, what leverage multiple is typical for their asset class, since Thomson Reuters Practical Law notes that debt in these deals is often expressed directly as a multiple of EBITDA. DSCR benchmarks in effect (per ClearValue Lending, sourcing SBA SOP 50 10 8). SBA 7(a) and 504 loans carry a 1.15x minimum (Clear Value Lending). SBA 7(a) Small Loans of $350,000 or less carry a 1.10x floor (DDEE.ai). EBITDA/NOI is the input feeding all three tests. Get that number wrong or present it without normalization and all three sizing tests cascade to the wrong answer, and the loan shrinks or disappears.

EBITDA in loan covenants: the ongoing obligation most borrowers underestimate

EBITDA's job doesn't end at closing. According to eCapital, it's an integral component of many loan covenants that banks use to monitor borrower risk for the full life of the loan, not just at the moment the loan gets sized.

That breach carries teeth. A lender can demand full repayment of the outstanding balance, seize agreed-upon collateral, or raise the interest rate on the loan, and the standard first move once a covenant trips is for the account to move into what's called workout, sometimes labeled special assets, troubled credit, or corporate recovery depending on the institution. None of that requires a missed payment. It only requires EBITDA to deteriorate past the line drawn in the credit agreement.

That's the asymmetry that catches sponsors off guard: a property or business can be current on debt service, every payment made, and still trigger a covenant breach if EBITDA deteriorates, because operating performance and payment history are separate tests. Compounding the risk, the EBITDA definition governing the covenant may not match the EBITDA figure presented at underwriting at all. Thomson Reuters Practical Law notes that these definitions get negotiated into the loan agreement with specific add-backs and carve-outs, so sponsors need to read the defined term itself rather than assume the textbook formula applies.

EBITDA needs tracking between reporting periods, and the annual covenant test date alone is too infrequent to catch problems as they develop. Treat it as an ongoing asset management discipline rather than a box checked once at origination: the early warning that prevents a workout is visible only when someone is actually watching the number move. In a common covenant structure, the lender sets a maximum Debt/EBITDA ratio (a ratio of less than 3x is a commonly cited threshold), and if the borrower's ratio rises above it, covenant breach occurs even if every payment has been made on time.

The lending environment that makes accurate EBITDA presentation decisive right now

Lending volume is accelerating fast. CBRE's Lending Momentum Index rose 112% year-over-year in the third quarter of 2025, marking the highest level of lending activity recorded since 2018. Lenders are processing more deals, faster, which cuts against the instinct to assume a slower market means more time to get the EBITDA presentation right.

Layer onto that a maturity wall few sponsors can ignore. Every sponsor refinancing one of those loans has to re-underwrite EBITDA in a rate environment that looks nothing like the one the original loan was priced in.

Underwriting is also getting tighter, not looser, even as volume climbs. Over 900 banks currently carry CRE exposure above 300% of capital, and that concentration has pushed institutions toward stricter underwriting standards rather than toward pulling back on volume altogether Agora Real Estate / Commercial Real Estate Lending Trends. Scrutiny of EBITDA quality is sharpening as a direct consequence, not softening.

Private credit has stepped further into the gap. Private lenders accounted for roughly 34% of all construction financing in 2025, up from about 20% in 2023 and around 9% in the years right after the financial crisis, and these lenders frequently bring EBITDA-based underwriting frameworks imported straight from corporate lending, particularly on operating-business assets. Northmarq's 2026 outlook frames the current constraint accurately: capital access itself isn't the bottleneck anymore, timing, structure, and borrower confidence are, and how a sponsor presents and defends EBITDA, add-backs included, now shapes execution speed and cost of capital as much as it shapes whether the deal closes at all. CMBS issuance adds one more layer of pressure toward precision, with year-to-date issuance reaching $115.2 billion through November 2025, the highest level since 2007, and CMBS pools covering operating-business assets require standardized EBITDA disclosure to pass rating agency review Agora Real Estate / Commercial Real Estate Lending Trends. The maturity wall adds urgency: per the MBA, approximately $875 billion in commercial mortgages are scheduled to mature in 2026, part of an estimated $1.3 trillion maturing across 2025–2026 combined (sponsors refinancing these loans must re-underwrite their EBITDA in a higher-rate environment) (Commercial Observer, LBC Capital) LBC Capital / 2026 Real Estate Lending Market Outlook.

Common EBITDA adjustments and lender pushback

Raw EBITDA pulled straight from a financial statement is rarely the number a lender actually underwrites to. It tends to carry one-time items, owner-specific expenses, or non-recurring costs that don't reflect ongoing operating performance, and the figure lenders actually rely on is the normalized, or "adjusted," version that strips those distortions out.

The most common add-backs include one-time legal costs or litigation settlements, owner compensation set above market rate (a frequent issue in owner-operated hospitality and senior housing), non-recurring capital expenditures that got expensed through the income statement rather than capitalized, and management fee normalization, where the actual fee paid gets compared against an arm's-length market rate. Lenders push back hardest in three specific spots: recurring items dressed up as one-time, revenue booked before cash actually gets collected, and cost savings that only hold up if the current owner sticks around after closing, an assumption no lender wants to underwrite to.

Thomson Reuters Practical Law confirms that these definitions get negotiated directly into the loan agreement, and lenders routinely exclude items like working capital movement and capital expenditure from the approved add-back list, even when borrowers push hard for their inclusion. Sponsors need to know, ahead of the conversation, which add-backs are standard practice and which will require real documentation to justify.

Capital expenditure deserves its own callout, because EBITDA never subtracts it. eCapital is direct on this point: the money used to fund equipment replacement or other capex sits outside the EBITDA calculation entirely, neither added nor subtracted. That's precisely why lenders underwriting operating-business CRE assets often run a second, more conservative test alongside headline EBITDA, sometimes called EBITDA less capex or unlevered free cash flow, that accounts for the maintenance spending headline EBITDA simply ignores. Sponsors preparing for one of these conversations should build a normalized EBITDA bridge that starts from the reported figures, walks through each adjustment individually, backs each one with documentation, and arrives at the number actually being underwritten. Lenders who can't follow that logic step by step will apply their own haircut, and it's rarely a generous one.

Using EBITDA in CRE deal analysis and lender conversations

Which metric to lead with depends entirely on the asset. Stabilized multifamily, office, industrial, and net-lease retail should lead with NOI, since that's what lenders expect to see and what actually drives valuation in those categories. Hotels, senior housing, self-storage, and other operating-business-anchored assets call for EBITDA presented alongside, or in place of, NOI, matching whatever underwriting framework the lender is actually going to apply. Any refinance or acquisition sitting inside a covenant-heavy loan structure demands a close read of the EBITDA term as defined in the credit agreement itself, which can differ from the metric used at origination.

The documentation lenders expect on these deals differs meaningfully from a standard CRE package. Two to three years of full operating financials belong in the package, profit-and-loss statements, not simply a rent roll. A normalized EBITDA bridge with line-item documentation for every adjustment needs to accompany it. Trailing twelve-month figures, plus an interim period where available, round out what a lender actually wants to underwrite against.

None of this is optional paperwork. In a lending environment defined by faster deal flow, a looming maturity wall, and tighter underwriting standards even as volume climbs, the sponsor who can present a clean, well-documented EBITDA figure, and defend every add-back on it, moves faster and pays less for capital than the one who can't. The metric was never a real estate invention. But real estate, particularly at the operating-business end of the asset spectrum, now runs on it just the same. Documentation requirements differ from standard CRE packages.