One formula, endlessly repeated
Every flipping forum, podcast, and late-night guru eventually writes the same line: pay no more than 70% of the after repair value, minus repair costs. Maximum allowable offer equals ARV times 0.70, minus rehab. A house worth $300,000 fixed up, needing $45,000 of work, prices out at $210,000 minus $45,000: offer no more than $165,000.
The rule survives because it compresses a genuinely complicated cost structure into arithmetic you can do in a driveway. But a compression is not a law. The 30% it holds back is meant to cover two different things, your profit and a long list of costs, and whether that split actually works depends on the price tier, the timeline, and the market. So let me unpack where the 30% goes, run a deal where the rule saves you real money, and then run one where obeying it would have cost me a perfectly good payday.
What the 30% is actually for
Roughly speaking, the held-back 30% divides into two buckets. Costs: buying closing costs, financing (hard money interest and points), carrying costs while you own it (taxes, insurance, utilities), and selling costs (commissions, credits, transfer taxes), which together commonly absorb something like 12 to 15% of ARV on a typical flip timeline. Profit: whatever remains, usually landing near 15 to 18% of ARV if the estimate holds. Both figures are rules of thumb, and the worked example below shows where they came from.
Notice what this means: the rule is not '30% profit.' Half the margin is spoken for before you earn anything, which is why small violations of the formula do outsized damage to the half that is yours.
Deal one: the rule as a seatbelt
The house: ARV $300,000, needing $45,000 of work, so the rule says pay $165,000 at most. Buy it right there and here is the full ledger. Purchase $165,000, rehab $45,000, buying closing costs $3,000. Financing: a hard money loan covering 90% of purchase, $148,500, at 11% interest-only for a 5 month project is $6,806 of interest, plus 2 points at $2,970. Carrying costs, taxes, insurance, and utilities for those months, $2,500. Selling costs at 8% of the $300,000 sale, covering commissions, seller closing costs, and a modest buyer credit: $24,000. Total: $249,276. Profit: $50,724, which is 16.9% of ARV. The rule of thumb delivered almost exactly what it promised.
Now watch the seatbelt work. The seller holds firm at $185,000 and you talk yourself into it, because it is 'only' $20,000 on a $300,000 house. Financing scales up with the price, so the ledger becomes $270,461 all-in, and profit drops to $29,539. Then the market softens 5% while you renovate and the house sells at $285,000: profit $15,739. Then the project runs three months long, adding about $4,579 of interest and $1,500 of carry: profit is now roughly $9,660, on half a year of work and $70,000-plus of your cash exposed. One overpayment plus two ordinary misfortunes took $50,724 down to four figures. That cascade, not pessimism, is what the 70% rule is protecting you from: it prices in the certainty that something slips.
Deal two: the rule as a blindfold
Now move the same logic to a $650,000-ARV neighborhood, repairs $60,000. Seventy percent of $650,000 is $455,000, minus the $60,000 of repairs gives a maximum offer of $395,000. In most markets, nobody with a house worth $650,000 fixed is handing it over at $395,000 outside of genuine distress. Followed literally, the rule simply exits you from the entire price tier.
But run the actual ledger at a realistic $455,000 purchase. Rehab $60,000, closing $5,000. Hard money at 90% of purchase, $409,500, costs $18,769 of interest over 5 months at 11%, plus 2 points at $8,190. Carry $4,000. Selling at 7% of $650,000, $45,500. Total: $596,459. Profit: $53,541. That is 8.2% of ARV, well under the rule's blessing, and also more dollars than deal one, earned in the same five months. The rule rejected this deal not because the deal is bad but because percentage-based margins mislead across price tiers: costs scale with price, but a profit measured in dollars pays your bills, not one measured in ARV points. In expensive markets, experienced flippers quietly run 75 to 80% versions of the formula for exactly this reason, and in rough cheap markets some tighten it below 70. The percentage is a dial, not scripture.
Holding costs: the line most first flips blow
Here is my admission for this post. On my first flip I estimated the rehab almost to the dollar and was proud of it, and I still nearly lost money, because I had budgeted five months and the project took eight. Permit review, a plumber who vanished, then relisting after a failed buyer. Nobody warns you that the meter is the danger, not the toolbox. Every extra month on deal one's numbers is roughly $1,361 of hard money interest plus about $500 of taxes, insurance, and utilities: call it $1,900 a month of pure erosion, before price risk even stirs.
This is why experienced flippers obsess over days on market and contractor scheduling more than tile choices. When you model a flip, and the house flip calculator makes this explicit, run the timeline at your estimate plus three months and see if the deal survives. If it only works on the optimistic schedule, it does not work.
The two ends of the deal everyone underprices
First-time flippers model the purchase and the rehab and then wave vaguely at the transactions on either side. Both ends are computable. The entry costs, title, escrow, lender fees, prepaids, behave exactly like any purchase, and the closing costs estimator prices them in minutes. The exit is bigger: commissions, seller-side closing costs, transfer taxes, and almost always some buyer credit after inspection, which is why my examples carry 7 to 8% of the sale price. The seller net proceeds calculator turns your hoped-for sale price into the number that actually pays off the hard money, and I walked through every one of those seller lines in what it costs to sell a house.
If your flip model shows selling costs under 6% of ARV, you have not underpriced a line item. You have deleted one.
How to actually use the rule
Use it as a screen, never as underwriting. At the driveway stage, 70% of ARV minus repairs sorts a stack of leads into 'worth a full workup' and 'not close' faster than any spreadsheet. But before an offer, every surviving deal gets the full ledger: real financing quotes, a padded timeline, honest selling costs, and a profit target in dollars that respects the cash and months at risk. If BRRRR is your exit instead of a sale, the same discipline applies with a refinance where the closing table would be, and I ran that full version in the BRRRR honest guide.
And the standing note, which applies doubly to a strategy with this much leverage and timeline risk: everything here is arithmetic on stated assumptions, not financial advice. The formula fits in a text message. The judgment about whether your market, your contractor, and your cash reserves fit the formula does not.
Questions people ask
A screening formula: pay no more than 70% of the after repair value minus repair costs. On a house worth $300,000 fixed needing $45,000 of work, the maximum allowable offer is $210,000 minus $45,000, which is $165,000. It reserves roughly 30% of ARV for costs and profit combined.
Because on a typical mid-priced flip, transaction, financing, carrying, and selling costs absorb roughly 12 to 15% of ARV, leaving about 15 to 18% as profit, cushions that history has shown flips need. In higher price tiers those cost percentages shrink relative to dollars, which is why many flippers run 75 to 80% versions there.
Everything the property costs you per month of ownership: hard money interest, property taxes, insurance, and utilities. In the worked example they ran about $1,900 a month, which is why schedule overruns, not rehab overruns, quietly kill the most first flips.
Implicitly, yes: commissions, seller closing costs, and buyer credits are part of what the 30% holdback must cover, typically 7 to 8% of the sale price on their own. Modeling a flip without them overstates profit by tens of thousands of dollars.
