How it works
What you do takes about ten minutes. Here's what happens on both sides.
What you do
-
1
Create an account and pay
One flat fee, paid once, through Stripe. You'll have an account where every report you order stays available to download.
-
2
Enter the property details
The address, bedrooms, bathrooms, square footage, lot size, year built, and garage. We pre-fill what public records already know so you're mostly confirming rather than typing.
There's also a free-text box for anything a record won't tell us — a kitchen you gutted last year, a roof that's original, an addition that was never permitted. This genuinely changes the number, so it's worth a few sentences.
-
3
Upload photos
Front and back exterior, kitchen, bathrooms, main living areas, and anything unusual. Phone photos are completely fine — that's what nearly everyone sends.
More photos are better than fewer. The condition assessment is only as good as what it can see, and an unphotographed room is assumed to be average.
-
4
Submit and wait
We email you when the report is ready — usually within a few hours. It also appears in your dashboard.
What we do
The part that matters: how the number is actually built. This is the same sales comparison approach a licensed appraiser uses — automated, and shown to you in full.
1. Pull the market data
We query ATTOM's nationwide property database for your subject property's characteristics and for recent sales in the surrounding area — typically a few dozen candidates before filtering.
2. Select genuine comparables
Candidates are scored on distance, recency, living area, age, story count, and property type. Distressed and non-arm's-length sales are filtered out, since a foreclosure or a family transfer doesn't tell you what the open market would pay. Properties of a materially different type or era are set aside rather than force-fit.
3. Derive adjustment rates from the local market
Rather than applying national rules of thumb, we derive the dollar value of a square foot, a bedroom, a bathroom, a garage bay, and a year of age from the sales in your market using paired-sales and regression analysis. A bathroom is worth something very different in Youngstown than in Miami, and the adjustments reflect that.
4. Assess condition from your photos
Your photos are analyzed to grade the property's condition on the standard C1–C6 scale used in appraisal practice. The condition spread — what the market actually pays for a renovated home versus a tired one — is measured from the unexplained variation in your local comps, not guessed.
5. Adjust each comparable to your property
Every comp gets a line-by-line grid: what it sold for, then plus or minus for each way it differs from your property. Sale dates are time-adjusted for market movement since the sale. The adjusted values are then weighted, with the closest matches counting most.
6. Reconcile into a value and a range
You get a point value, a supported range, and a confidence level that reflects how much adjustment the comps required. Comps that needed heavy adjustment widen the range and lower the confidence — we'd rather tell you the number is soft than pretend it isn't.
7. Cross-check and deliver
The result is sanity-checked against a cost approach and against your property's own sale history before the PDF is generated and emailed to you.
Where this approach is weakest
Worth knowing before you pay. The method depends on having genuine comparable sales nearby, so it's least reliable for:
- Rural and large-acreage properties — comps may be miles away and not truly comparable.
- Unique or historic properties — if nothing like it has sold, there's nothing to compare it to.
- Very thin markets — few recent sales means a wide, low-confidence range.
- Properties with hidden problems — we can't see foundation issues, and photos won't reveal what's behind a wall.
If your property falls into one of these and we can't produce a credible value, we refund you in full rather than sell you a number we don't believe.
See it on your own property
Full report within hours.
Start Your ReportWant to see the output first? Look at a sample report.