Adjusting Comps for Condition Without Fooling Yourself

Two houses on the same street, same footprint, same year: one sold for $312,000 last spring, the other for $228,000. Nothing is wrong with either number. One had been renovated and one had not.

Condition is usually the largest adjustment you will make to a comp, and it is the one investors most often invent. Here is how to derive it instead.

Derive the adjustment, do not assume it

The wrong method is picking a number that feels right — "renovated is worth about 20% more" — and applying it everywhere. The right method is a paired sale: two sales that differ mainly in the one variable you are pricing.

Worked example (hypothetical). Your subject is a 1,450 sq ft 3/2 that you will renovate fully. In the same submarket:

• Comp A — renovated, 1,500 sq ft, sold $312,000
• Comp B — original condition, 1,420 sq ft, sold $228,000

Before you can compare them, remove the size difference. Use a marginal rate for additional finished square footage — what the market pays for the next square foot, not the average price per square foot of the whole house. Say that rate is $75 in this submarket:

• Comp B size-adjusted to Comp A: $228,000 + (1,500 − 1,420) × $75 = $228,000 + $6,000 = $234,000
• Condition premium: $312,000 − $234,000 = $78,000

That $78,000 is a derived figure for this submarket at this price point at this time. It is not a rule, and it does not travel to the next ZIP code.

Apply it toward your subject

Adjust each comp toward the subject — up if the comp is inferior, down if superior:

• From Comp A (already renovated, 50 sq ft larger): $312,000 + (1,450 − 1,500) × $75 = $308,250
• From Comp B (original condition, 30 sq ft smaller): $228,000 + (1,450 − 1,420) × $75 + $78,000 = $308,250

Those two agreeing looks like confirmation. It is not. Comps A and B are the pair the adjustment was derived from, so applying it back to them is circular — the arithmetic cannot disagree with itself. This is the single most common way an investor fools themselves with a comp grid, and it feels like rigor while it happens.

You need a comp from outside the pair to learn anything new.

• Comp C — kitchen updated only, everything else original, 1,480 sq ft, sold $270,000
• Size-adjusted: $270,000 + (1,450 − 1,480) × $75 = $267,750

Now you have a genuine question: how much of the $78,000 renovated premium is the kitchen? You do not know. You can bracket it — the subject should be worth more than $267,750 and the full-reno comps say around $308,000 — but you cannot fabricate the split. Comp C belongs in the file as support, not as a driver of the number.

Three ways this goes wrong

Adjusting the wrong direction. The rule is unforgiving and easy to invert: adjust the comp, toward the subject. A comp superior to your subject gets adjusted down. Half of bad grids are a sign error.

Stacking overlapping adjustments. If you take a condition adjustment for "renovated" and then separately adjust for the new kitchen, the new bathroom and the new flooring, you have counted the renovation two or three times. Pick one level of granularity and stay there.

Adjusting until you get the answer you wanted. If your ARV moved every time you added a comp and it only ever moved up, you were not analyzing. Write your estimate down before the last two comps go in, and see whether they change it.

Document it so a second reader can follow

The test is not whether you believe the number. It is whether a lender, a partner or a buyer's agent can read your grid in five minutes and reconstruct how you got there. That means, for each comp: the raw sale price, each adjustment as a separate line with its sign, the source of each adjustment rate, and the adjusted figure.

A grid where every comp lands within a few thousand dollars of the others after honest adjustments is a real ARV. A grid where they scatter $40,000 apart is telling you something true — that your submarket is not homogeneous, or that you do not have the right comps yet. The scatter is the finding. Do not average it away.

Frequently asked questions

What is a paired sales analysis?

Comparing two sales that are similar in every meaningful way except one variable, so the price difference between them prices that variable. It is how appraisers derive adjustment amounts rather than assuming them.

Which direction do you adjust comps?

Always adjust the comparable toward the subject. If the comp is better than your subject, adjust its price down. If it is worse, adjust up. The adjusted figure answers "what would this comp have sold for if it were my house?"

How much is a full renovation worth on a comp?

There is no portable number — it depends on the submarket, the price band and the buyer pool. Derive it from paired sales in the same area and the same period, and re-derive it when you move to a different submarket.

Should I use price per square foot to adjust for size?

Not the average price per square foot of the whole house. Use a marginal rate — what buyers pay for the next square foot of finished space — which is typically well below the average and is itself derived from paired sales.

How many adjustments is too many?

When the total adjustment on a comp gets large relative to its sale price, the comp is telling you it is not really comparable. At that point the honest move is to find a better comp, not to adjust harder.