How to build a real estate buy box
A buy box is the written set of criteria a property must meet before you'll spend time on it: geography, property type, price band, condition, and the return floors underneath them. A useful one is specific enough to reject most of the market — if your criteria don't eliminate 95% of listings on sight, they aren't criteria, they're preferences. Write it down, apply it mechanically, and revisit it on a schedule rather than whenever a tempting deal fails it.
The five layers
Order matters. The cheap filters go first so the expensive judgment is only spent on what's left:
| Layer | Decides | Cost to check |
|---|---|---|
| 1. Geography | Metro, submarkets, ZIP codes, or a drive-time radius | Free — a map filter |
| 2. Property | Type, beds, baths, square footage, year built, lot | Free — listing fields |
| 3. Price and capital | Price band, maximum cash in, maximum rehab | Free — arithmetic on your own balance sheet |
| 4. Condition and situation | Scope you'll take on, occupancy, title you'll tolerate | Cheap — photos, remarks, a preliminary search |
| 5. Return floors | Cash-on-cash, DSCR, cap rate, cash flow per door | Expensive — a real underwrite |
Run in this order, a market of 400 listings usually reaches layer 5 with fewer than 20 candidates. Run in the reverse order, you underwrite 400 properties to reject 380 of them for reasons visible in the first screenshot.
A starter buy box
A single-family rental box for a mid-market metro. Copy the shape rather than the numbers — the numbers are only defensible against a specific market's rents and prices:
| Criterion | Value | Why this and not something looser |
|---|---|---|
| Submarkets | 6 named ZIP codes | Rent and turnover are local. "The metro" is not a market you can know. |
| Type | Single-family and duplex | Financing and exit are ordinary; the buyer pool at sale is the largest |
| Beds / baths | 3+ bed, 1.5+ bath | Two-bed rentals attract shorter tenancies in most family markets |
| Size | 1,100 – 2,000 ft² | Above 2,000, rent stops keeping pace with price |
| Year built | 1955 or later | Avoids the systems-replacement cliff and most lead-paint scope |
| Price | $150,000 – $260,000 | Matches the capital available for 3–4 acquisitions, not 1 |
| Rehab | Under $25,000, cosmetic only | Structural scope is a different business with a different skill set |
| Occupancy | Vacant, or lease ending within 90 days | A below-market lease with 14 months left is a 14-month problem |
| Cash-on-cash | ≥ 7% at 25% down | The return on the money actually invested |
| DSCR | ≥ 1.25 | What a lender will require, and what survives a vacancy |
| Cash flow per door | ≥ $150/month after all reserves | An absolute floor so a percentage can't flatter a small deal |
Eleven criteria, all checkable, and the last three require a real underwrite. That's the shape to aim for: most of the box is free to apply, and the expensive part is reached by few properties.
Check the return floors
Cash-on-cash, DSCR and the full expense breakdown from one set of inputs — the layer-5 test, on a property that's already passed the cheap filters.
Open the cash-on-cash calculatorThree criteria that quietly reject nothing
- "Good neighborhoods." Unfalsifiable, and it drifts to fit whatever you're looking at. Use something a stranger could check: named ZIP codes, a school rating threshold, a rent-to-price band, a vacancy rate. Anything you can point at.
- "Positive cash flow." Positive by a dollar is positive, and it's also one bad month from negative. It also invites the trick of quietly dropping the maintenance and vacancy reserves until the number turns green. Set a floor per door, after reserves, and name the reserves.
- "Below market value." Below whose market value? Without a stated method this criterion is satisfied by any listing you can talk yourself into. Replace it with a discount to a comp-set median, or drop it — see how to find underpriced homes.
The common thread: a criterion that can't be failed by a property you like isn't a criterion. The test of a buy box is that it occasionally rejects something you wanted.
Making it operational
Write it in one place, in one page, and give the same page to anyone sending you deals — an agent, a wholesaler, a partner. A wholesaler who knows your box sends you three relevant deals a month instead of thirty irrelevant ones, and that alone tends to justify the hour spent writing it down.
Then hold a standing review. Track what you rejected and why; if one criterion is responsible for most of your rejections, it's either doing the most work or it's miscalibrated, and only the record tells you which. For what to do once a property passes, see how to analyze a deal end to end and how to run comps on a property.
Common questions
How specific should a buy box be?
Specific enough that someone else could apply it without asking you a question. "Cash-flowing rentals in good areas" is not a buy box — it contains no test anyone could fail. "3+ bed single-family, 1,100–2,000 ft², built after 1955, $150k–$260k, in these six ZIP codes, minimum 7% cash-on-cash at 25% down" is.
Should I have more than one buy box?
Only if you genuinely run more than one strategy, and then keep them completely separate. A merged rental-and-flip box has the union of both sets of criteria and the discipline of neither, so it accepts deals that are mediocre at both.
How often should I change it?
On a schedule — quarterly is reasonable — and never in the middle of evaluating a deal that just failed it. That is the moment the change is least likely to be about the market and most likely to be about the deal. If a criterion keeps blocking things you later wish you'd bought, that's real evidence; note it, and act on it at the review.
What if nothing meets my buy box?
First check the return floors against what the market can currently deliver — an 8% cash-on-cash floor in a market where the best properties yield 5.5% will reject everything forever, which is information about the market, not a filter that needs loosening. Then decide deliberately: widen the geography, change strategy, or wait. Quietly relaxing the floor one property at a time is the failure mode.
Related guides and calculators
How to find underpriced homes
The signals worth screening on, and how to run them across a market rather than a listing at a time.
How to analyze a wholesale deal
End-to-end on one property, including the buyer's-side check most wholesalers never run.
How to calculate MAO
The maximum allowable offer formula, both versions, and how to pick the percentage honestly.
How to find real estate comps
Where the sold data actually lives, and the six filters that separate a comp from a nearby house.
Point a buy box at a market.
Faro scans a city against your criteria, underwrites every listing that fits, and ranks what's left by return.
Estimates for analysis and educational use only — not financial, investment, tax, or legal advice. Verify every number independently before making a purchase decision.