A cap rate, short for capitalization rate, is a property’s annual net operating income divided by its price, written as a percentage. A building that throws off $500,000 of net income and sells for $10 million has a 5% cap rate. That one number is how commercial real estate buyers compare deals at a glance and how sellers price them — which is exactly why it is also the most manipulated figure on any marketing flyer. This piece explains what a cap rate actually measures, what a quoted one quietly hides, and where AI genuinely helps a lean firm pressure-test the number before it moves a bid.
What a cap rate is, in plain terms
A cap rate is the yield a property produces if you bought it in cash, before any financing. Take the income the building generates after operating costs — its net operating income, or NOI — and divide it by the price. The result is the return the property earns on the money you put in, ignoring any loan.
The formula is simple arithmetic:
Cap rate = annual net operating income ÷ price
Run it forward or backward. A property earning $180,000 of NOI listed at $3 million carries a 6% cap rate ($180,000 ÷ $3,000,000). Flip it around and the cap rate prices the deal: if similar buildings in the market trade at a 6% cap and this one earns $180,000, the implied value is $3 million ($180,000 ÷ 0.06). Buyers use it both ways — to read how a listing is priced, and to translate an income figure into a value.
The reason the whole industry leans on one number is comparison. A broker with three retail centers of different sizes, prices, and incomes can line them up on a single axis by cap rate and see instantly which is priced richer or cheaper relative to what it earns. NOI is the foundation the cap rate sits on, and it is one of the handful of metrics that carry most of the weight in deal analysis — the full set is laid out in the guide to what underwriting actually means for a small firm.
What the cap rate tells you — and what it hides
The cap rate is a snapshot, and knowing what it deliberately leaves out is what keeps it from misleading you. Three things it does not account for.
It ignores debt. The cap rate is an unlevered figure — it describes the property, not your financing. Two buyers can pay the same price for the same building at the same cap rate and earn wildly different returns on their cash depending on their loans. That is a feature, not a flaw: stripping out financing lets you compare the asset itself, apples to apples. But it means the cap rate is not your return. Cash-on-cash and internal rate of return, which do fold in the loan, are separate numbers.
It ignores time. A cap rate is a single year frozen in place. It says nothing about where rents are heading, what the building will cost to maintain over your hold, or what it will sell for later. A 6% cap on a property with rising rents and a 6% cap on one bleeding tenants are the same number describing two different futures.
It ignores growth and capital. The cap rate uses today’s income against today’s price. It does not know that half the leases roll next year or that the roof is at the end of its life — those facts live in the assumptions behind a full model, not in the cap rate. The number is a starting read, not a verdict.
What counts as a good cap rate
There is no universal “good” cap rate — the number is only meaningful against the property type, the market, and the risk. As a rough frame, commercial cap rates in the US generally sit somewhere between 4% and 10%, and where a specific deal lands says more about risk than quality.
A lower cap rate means a higher price for the same income, which the market assigns to assets it sees as safer or more desirable: a well-leased industrial building in a supply-constrained market, a grocery-anchored center with a long lease. A higher cap rate means a cheaper price relative to income, which usually signals more risk — an older office building with near-term rollover, a property in a softening submarket, a tenant whose credit is shaky.
CBRE’s 2026 US Real Estate Market Outlook points to modest cap-rate compression across most property types after two years of expansion, with the spread staying wide for sectors carrying demand uncertainty like office and narrow for stable ones like industrial. The practical takeaway is not the exact figures, which move quarterly, but the principle: a cap rate that looks high for its asset class is the market pricing in a risk, and your job is to find that risk before you decide it is a bargain. Where those comparable cap rates come from is the subject of how AI is changing CRE market research.
Why a quoted cap rate is often too good to be true
Here is the part the textbook definitions skip: the cap rate on a marketing flyer is a claim, not a fact, and it is the easiest number in the whole package to inflate. Because value moves inversely to the cap rate, a seller has every incentive to make the cap rate look as high as possible — a higher advertised cap rate makes the same price look like a better deal. There are a handful of standard moves, all of them legal, all of them common.
Pro-forma income instead of actual income. The most frequent one. The flyer quotes a cap rate built on what the property could earn — market rents on every unit, full occupancy, projected rent bumps — rather than what it earns today. A “6.5% cap” that assumes the three vacant units are leased at rents no one is currently paying is a number about a future that may not arrive.
Understated or scrubbed expenses. NOI is income minus operating costs, so shrinking the cost side inflates NOI and the cap rate with it. Management fees left off because the owner self-manages, a property-tax line that ignores the reassessment a sale will trigger, deferred maintenance that never shows up as an expense — each one quietly lifts the quoted cap.
In-place rents presented as market rents, or the reverse. Whichever framing makes the number look better tends to be the one on the flyer. A cap rate is only as honest as the NOI underneath it, and the NOI is only as honest as the rent and expense assumptions underneath that.
None of this makes brokers dishonest — a pro-forma cap rate is a legitimate way to show upside, as long as everyone knows that is what it is. The danger is taking a quoted cap rate at face value and pricing a bid against a number that describes a hoped-for property rather than the real one. The fast filter that separates deals worth this scrutiny from the flood that is not has its own discipline, covered in how investment firms screen opportunities.
How to sanity-check a cap rate in three steps
Treating a quoted cap rate as a claim to verify, rather than a fact to accept, is the single habit that protects a lean firm from overpaying. The check is three steps.
-
Rebuild the NOI from actuals. Ignore the flyer’s income figure and reconstruct NOI yourself from the source documents — the rent roll for in-place rents and occupancy, the trailing twelve months of operating statements for real expenses. Add back the costs a seller conveniently dropped: management, a realistic reserve for capital, the property tax the sale will reset. The cap rate you get from your NOI is the one worth pricing against.
-
Compare against real comps. A cap rate means nothing in isolation. Line the deal up against what genuinely comparable properties — same type, same submarket, similar age and tenancy — have actually traded at recently. If the flyer’s cap rate is a full point above the comps, either the property carries a risk you have not spotted or the income is inflated. Both are worth knowing before you bid.
-
Check that the math ties out. Confirm the arithmetic is internally consistent: does the quoted cap rate actually equal the quoted NOI divided by the asking price? Does the NOI on the flyer match the NOI you can build from the rent roll and operating statements? Discrepancies between a summary number and the underlying documents are common, and they are exactly the kind of thing that gets missed when a deal moves fast.
Do this on every deal that survives screening, and a manipulated cap rate stops being a trap. The problem for a 4-to-20-person firm is not that the check is hard — it is that steps one and two are hours of mechanical document work, and a busy pipeline rarely leaves time for them on every deal.
Where AI helps sanity-check a cap rate
This is where AI earns its place: it collapses the hours the sanity check demands, without ever being trusted to make the judgment call. Every part of the mechanical grind above is work a model does quickly.
Point a general assistant — ChatGPT, Claude, or Microsoft Copilot — or a purpose-built deal tool at the rent roll, and it reads the tenants, in-place rents, and occupancy into a clean table in seconds instead of the twenty minutes it takes to key by hand. Feed it the trailing operating statements and it summarizes the real expense lines, flags the ones a pro-forma tends to omit, and drafts an NOI from actuals you can then correct. Hand it the flyer alongside the documents and it reconciles the two, surfacing exactly where the marketed NOI diverges from the numbers in the source files.
The arithmetic check is trivial for a model: does the quoted cap rate equal the quoted NOI over the price, and does that NOI match the documents? On the comps side, a tool wired into market data can pull recent comparable sales and their cap rates far faster than a lean team combing through subscriptions by hand. The pattern that lets a small shop run this depth of check on every deal, the way a larger firm with a full analyst bench would, is the argument of the small-firm playbook for out-operating institutional competitors.
Used this way, AI turns the cap-rate sanity check from a chore you skip under time pressure into a fast first pass you run on everything. The firm underwrites more deals, catches more inflated caps, and spends its scarce hours on the judgment rather than the transcription.
Where AI must not go
The line is firm, and the whole value of the tool depends on holding it: AI assembles and reconciles, but a human owns every assumption and the final number. A cap rate has two ingredients — the NOI and the value — and both rest on judgment a model cannot supply.
Ask a language model for a market cap rate, an exit cap rate, or a rent-growth assumption and it will produce a confident, specific figure it has no basis for. It has no genuine read on your submarket’s next eighteen months and no memory of what traded down the street. In a deal, a confidently wrong cap rate prices a confidently wrong offer. The model can pull the comps; you decide what they mean and what cap rate is defensible for this property.
The same holds for the NOI. AI can extract every figure from a rent roll, but a number read off the wrong row of a messy scanned document looks just as clean in the output table as a correct one. Every figure that will move money gets verified against its source before you trust it. The safe division of labor is the one that governs all AI in deal work: the model reads, extracts, and reconciles at speed, and a person forms the assumptions, checks the extractions, and sets the number.
One guardrail specific to a firm with no IT department: a rent roll and an operating statement are confidential financial data. Before you upload a deal’s documents to any AI tool, get three answers in writing — where the data is stored, whether your inputs train shared models, and how you delete your history. Major providers state that business-tier and API data is not used for training by default, but the contract is your safeguard, not the marketing copy. For how the cap-rate check fits alongside screening, comps, and underwriting in one workflow, the deal-analysis playbook for lean teams sequences the whole stack.
FAQ
What is a cap rate in simple terms?
A cap rate is a property’s annual net operating income divided by its price, shown as a percentage. It tells you the return the building would earn if you bought it in cash, before any loan. A property earning $180,000 a year and priced at $3 million has a 6% cap rate. Buyers use it to compare deals quickly and to translate an income figure into a value: at a 6% market cap rate, $180,000 of income implies a $3 million price. It is a snapshot of one year, not a full return.
How do you calculate a cap rate?
Divide the property’s annual net operating income by its price or value, then express the result as a percentage. NOI is the income after operating expenses — rent and other income minus vacancy, taxes, insurance, maintenance, and management — but before any mortgage payment. So a building with $250,000 of NOI selling for $5 million has a 5% cap rate ($250,000 ÷ $5,000,000). The calculation is easy; the accuracy lives entirely in whether the NOI is honest, which is why rebuilding NOI from actual documents matters more than the arithmetic.
What is a good cap rate for commercial real estate?
There is no single good cap rate — it depends on the property type, market, and risk. As a rough range, US commercial cap rates generally fall between 4% and 10%. A lower cap rate means a higher price for the income, which the market assigns to safer, more desirable assets. A higher cap rate means a cheaper price, usually because the property carries more risk. A “good” cap rate is one that fairly compensates you for the specific risk of that deal, not simply the highest number you can find.
Is a higher or lower cap rate better?
Neither is universally better — it depends on whether you are buying and what risk you are taking. For a buyer, a higher cap rate means paying less for each dollar of income, which looks attractive but usually signals more risk: older buildings, weaker tenants, softer markets. A lower cap rate means paying more, typically for stability. A high cap rate is only a bargain if the risk the market is pricing in is one you understand and are comfortable with. Chasing cap rate alone, without asking why it is high, is how buyers overpay for problems.
Why is the cap rate on a listing often too good to be true?
Because a higher advertised cap rate makes the price look like a better deal, and sellers have several legal ways to inflate it. The most common is quoting a pro-forma cap rate built on projected market rents and full occupancy rather than actual in-place income. Others include leaving out management fees, ignoring the property-tax reassessment a sale triggers, or omitting a realistic capital reserve — each shrinks expenses and lifts the NOI. The cap rate is only as honest as the NOI underneath it, so rebuild the NOI from actual documents before you trust the number.
Can AI calculate a cap rate for me?
Yes, and more usefully it can rebuild the NOI the cap rate depends on. A general assistant or a deal tool can read a rent roll and trailing operating statements, pull the real rents and expenses into a table, draft an NOI from actuals, and compute the cap rate from your numbers rather than the flyer’s. It can also reconcile the marketed cap rate against the documents and flag where they diverge. The arithmetic is trivial for a model; the value is speed on the document work. Just verify every extracted figure against its source before the number moves a bid.
Should I let AI set the cap rate or the assumptions?
No. Setting a market cap rate, an exit cap rate, or a rent-growth assumption is judgment about a specific submarket at a specific time, and a model has no genuine basis for it. Asked for one, a language model will produce a confident, specific number with nothing behind it — and a wrong cap rate prices a wrong offer. Let AI pull the comparable sales and their cap rates, then decide yourself what cap rate is defensible for the property in front of you. The model supplies the raw material; you own the assumption and the final number.
What is the difference between a cap rate and cash-on-cash return?
A cap rate ignores financing; cash-on-cash return is built on it. The cap rate is NOI divided by price — an unlevered read on the property itself, useful for comparing deals apples to apples. Cash-on-cash is annual pre-tax cash flow, after the mortgage payment, divided by the cash you actually invested — a read on what your money earns given your specific loan. Two buyers can purchase the same building at the same cap rate and see very different cash-on-cash returns depending on their debt. The cap rate prices the asset; cash-on-cash measures your return on it.
Is it safe to upload a rent roll to an AI tool to check a cap rate?
Only under the right terms. A rent roll and operating statement are confidential financial data, so before uploading them to any AI tool, get three answers in writing: where the data is stored, whether your inputs train shared models, and how you delete and export your history. Major providers state that business-tier and API data is not used for training by default, but for a firm with no IT department the contract is the safeguard, not the marketing copy. Treat a vague answer as a reason to keep proprietary deal data out of that tool.
What is the difference between a going-in cap rate and an exit cap rate?
A going-in cap rate is what you pay today — the first-year NOI divided by your purchase price. An exit cap rate is what you assume a future buyer will pay when you sell, applied to the NOI at that point to estimate a sale value. The exit cap rate is an assumption about a market years away, and it swings the projected return heavily. It is also exactly the kind of number a model will invent if you ask; set it yourself from a defensible view of where the market is heading, and treat any AI-supplied exit cap as a placeholder to replace, not a figure to trust.
Key takeaways
- A cap rate is annual net operating income divided by price — an unlevered snapshot of a property’s yield, used to compare deals and translate income into value.
- The cap rate deliberately ignores debt, time, and growth. It is a starting read, not your return and not a verdict on the deal.
- Because value moves inversely to the cap rate, a quoted cap rate is the most inflated number on a flyer — pro-forma income, scrubbed expenses, and optimistic rents all push it up.
- Sanity-check every quoted cap rate in three steps: rebuild NOI from actual documents, compare against real comps, and confirm the math ties out.
- AI is genuinely useful for the mechanical side of that check — reading rent rolls, drafting NOI from actuals, reconciling the flyer, pulling comps — and dangerous the moment it is asked to supply the cap rate or the assumptions. Let it assemble; keep every number with a human.
Want to know where your firm’s hours actually go in deal analysis, and which parts AI can safely take off your plate? A short assessment maps that against your deal flow, your documents, and where AI belongs without introducing risk. Book your free AI-readiness assessment →
Arthur Wandzel
