Sales comps

Sales comps are recent property sales used to judge value.

They are useful until someone uses them like a parts catalog: the dimensions look close, so surely this old bearing belongs in that new machine. Same metro and similar unit count are a start. They are not compatibility.

A comp earns relevance by showing how an actual buyer priced a genuinely comparable stream of income, physical condition, location, and risk at a known moment. Everything after “actual buyer” needs proof.

A comp has to earn relevance

Factor	Why it matters
Sale date	Old pricing may reflect a different capital market
Location	Submarket beats metro averages
Vintage and condition	Renovation needs change value
Unit mix	Studio-heavy is not family two-bed stock
NOI quality	Real NOI beats adjusted poetry
Debt market at sale	Buyer pricing changes with financing

Price per unit is easy to quote. Cap rate and NOI quality are harder to fake if you make people show the math.

You also need to know what transferred. Renovated interiors, deferred roofs, assumable debt, land, unusual other income, or a pending tax reset can make two sales look alike in a summary and behave nothing alike after closing.

The headline is the paint. The sale file is the maintenance record.

One table, three different assets

Sale	Units	Price	Price/unit	NOI	Cap rate
Subject offer	80	$12,000,000	$150,000	$720,000	6.00%
Comp A	76	$11,780,000	$155,000	$765,700	6.50%
Comp B	92	$13,340,000	$145,000	$733,700	5.50%

If the sponsor points to Comp A’s price per unit and Comp B’s cap rate, you have met buffet-style valuation: one metric from each tray, none of the property attached.

Comp A and Comp B may both be useful. The work is explaining which differences require an adjustment and why. You do not get to borrow the highest price per unit, the lowest cap rate, and the nicest renovation photos as if one buyer paid for all three in the same transaction.

A comp is a whole sale. Keep the parts together.

Ask which property you would buy instead

Ask which comp you would buy instead.

If the subject is older, has worse collections, needs more capex, or sits in a weaker pocket, it should not automatically get the better comp’s pricing. The subject needs an adjustment, evidence that the difference is already reflected in its NOI, or a real operating plan with costs and timing.

This question ruins lazy comparison because it forces the conclusion back onto the assets. If Comp A produces stronger NOI from better condition in a stronger location, its price per unit is not a gift certificate for the subject.

The market does not pay the subject for resembling a better-maintained property from the highway.

Keep an adjustment log

For every comp, write down:

• Sale date and the financing environment at closing.
• Distance and submarket relationship to the subject.
• Unit count, unit mix, vintage, and physical condition.
• Sale price, price per unit, reported NOI, and cap rate.
• Known renovation scope, deferred maintenance, and near-term capital needs.
• Material income, expense, tax, insurance, or debt differences.
• The adjustment you made and the evidence supporting it.

An adjustment log exposes opinions pretending to be measurements. “Superior location” is not useful until somebody names the boundary, demand evidence, or operating difference that makes it superior. “Similar condition” should survive property photos, inspection information, renovation history, and capital-needs records.

When the adjustment cannot be supported, lower the comp’s weight. Do not tighten the language around it.

Score relevance before using price

Score each comp:

Score	Meaning
1	Strong comp: recent, close, similar, NOI known
2	Usable comp: one or two adjustments needed
3	Weak comp: interesting, not controlling
Reject	Too different or too old

Then value the subject using only the 1s and 2s. If the valuation needs the rejected comps, the conclusion was probably shopping for evidence.

The score is not advanced math. It is a wear limit. Once a comp has too many differences, stop asking it to carry the valuation.

Red flags in a polished comp slide

Slow down when:

• The slide gives price per unit but omits NOI or cap rate.
• Sale dates are old enough to reflect a different debt market.
• Metro-level location language hides meaningful submarket differences.
• Renovated properties support the subject’s exit without renovation cost or timing.
• Reported NOI has no source or uses different expense treatment.
• Different comps supply whichever metric makes the subject look best.
• The broker summary is available but the underlying comp packet is not.

Ask who verified the sale, where the NOI came from, what condition the property was in, what financing market priced it, and why the comp belongs. Then ask what would make it a reject. A comparison with no rejection rule is a collection, not underwriting.

Ask for the comp packet

Ask for the actual sales comp packet, not just a slide. You want sale date, price, units, NOI, cap rate, condition, and why each comp belongs.

Put every candidate through the relevance score, keep the 1s and 2s, and record each adjustment. Then check whether the subject’s purchase or exit value still works without the weak comps.

If the value survives, the comps supported it. If it needs a mismatched sale to keep running, that was the part making the noise.