The 70% rule for flips

The 70% rule is a screening convention, not a formula for value: a fixed percentage of the after-repair value, less the rehab budget, taken as the ceiling the convention places on an offer before any deeper analysis. The Fix & Flip Calculator computes that ceiling and, separately, the profit a specific purchase price would leave after the rehab, holding, and selling costs entered alongside it. This guide works the convention through on one deal, set up differently from the calculator page's example: the purchase price sits below the ceiling rather than on it, so the headroom between the two can be followed into every result; the rehab figure is built line by line in a budget before it enters the rule; and the rule percentage is run as a series — down, up, and out to the limiting case where the convention holds nothing back at all. Every figure is computed at build time by the engine behind the calculator, the inputs are stated in the sentence around each one, and no computed figure is typed in.

The convention and its arithmetic

The ceiling is rule% × after-repair value − rehab costs. The engine applies 70% when the rule field is left blank, and this page reads that default back from the engine's own result rather than typing it. The percentage sets the share of the ARV that acquisition and renovation together may consume; what it leaves aside — 30% of ARV at the default — is the margin. The rehab budget is then taken out of the share, not out of the margin, so the ceiling is the acquisition price that still fits inside the share once the renovation is paid for. Nothing else enters the formula: not the purchase price, not holding costs, not selling costs, and not the time the project takes.

The deal this guide uses throughout is entered as an after-repair value of $350,000, a rehab budget of $42,000, a purchase price of $190,000, holding costs of $12,000, and selling costs of 5% of ARV, with the rule field left blank. The ceiling is 70% × $350,000 − $42,000 = $203,000. The margin is $350,000 − ($203,000 + $42,000) = $105,000, which is 30% of the ARV. The entered purchase price sits $13,000 under the ceiling; this guide calls that gap the headroom, and it is the reason the deal was set up this way.

The projected profit is ARV − purchase − rehab − holding − selling costs, and the rule percentage is nowhere in it. Selling costs are the entered percentage of the ARV — the price the sale is assumed to close at — so 5% of $350,000 is $17,500. The engine adds the $190,000 purchase price, the $42,000 rehab, the $12,000 of holding costs, and the selling costs into a total invested of $261,500, nets it against the ARV for a projected profit of $88,500, and divides the profit by the total for a return on cost of 33.8%.

Because the entered price sits under the ceiling, the profit can be read a second way, as margin + headroom − holding costs − selling costs: $105,000 + $13,000 − $12,000 − $17,500 = $88,500, an identity the build checks. It shows what each piece of the convention is for. The margin is the room the percentage reserves; the headroom is the room the entered price adds to it; holding and selling costs are what the two have to pay for; and the profit is what is left. Every example that follows moves one of those pieces and shows which output moves with it.

The ceiling series

The rule percentage is a field, and the clearest way to see what it does is to step it. On the base deal, entering 60% gives a ceiling of $168,000; the default gives $203,000; 72% gives $210,000; 80% gives $238,000; and 100% gives $308,000. The 2-point step from the default to 72% raises the ceiling by $7,000, which is 2% of the ARV; the 10-point steps down to 60% and up to 80% move it by $35,000 each way, and the build checks that the two steps are equal. The rehab subtraction is the same in every run, so the whole difference comes from the first term. One consequence is worth noticing: at 60% the ceiling of $168,000 is below the entered $190,000 purchase price, so the same price is outside the convention at that percentage and inside it at every other percentage in the series. The engine does not say which; it reports the ceiling and leaves the comparison to the reader.

The projected profit at the entered $190,000 purchase price does not move at all across the series. It is $88,500 at 60%, at the default, at 72%, at 80%, and at 100%, and the build checks every run against the base. The percentage is not in the profit formula, so it cannot be.

What the percentage really sets is how much of the ARV is held back, which shows when the purchase price is set to each ceiling in turn, with the $42,000 rehab, $12,000 holding, and 5% selling inputs left as entered. At the 60% ceiling of $168,000, total invested is $239,500 and the projected profit is $110,500, a 46.1% return on cost. At the default ceiling of $203,000, it is $75,500 on $274,500 (27.5%) — the margin of $105,000 less the $12,000 of holding costs and $17,500 of selling costs, with the headroom now gone. At the 72% ceiling of $210,000, it is $68,500 (24.3%); at the 80% ceiling of $238,000, $40,500 (13.1%). Each step up in the percentage moves the ceiling up and the profit-at-ceiling down by the same amount, because every dollar the convention lets into the purchase price is a dollar taken out of the margin.

The limiting case makes the mechanism plain. At 100% the ceiling is simply the ARV less the rehab budget, $308,000: nothing is held back. Set the purchase price to that figure and the projected profit is -$29,500, a loss of exactly the $12,000 of holding costs plus the $17,500 of selling costs — the build checks that identity — at -7.8% on cost. The engine accepts the entry without complaint. It is not an error; it is the convention with its margin set to nothing, and the costs the margin exists to absorb have nowhere to go but the profit line.

The percentage is bounded on both sides. Enter a negative percentage, -5% for instance, and instead of a result the engine returns the message "The rule percentage cannot be negative." The upper bound is stated the same way, in the engine's own words, when it is crossed. Inside the bounds any percentage is accepted, 72% as readily as the default: the engine keeps no list of permitted percentages, and which one fits a given project is a judgment this guide does not make.

Where the rehab figure comes from

The rehab figure is the one input with its own tool, and the $42,000 in the base deal was not chosen at random: it is the grand total of a 5-line budget from the Rehab Budget Calculator, which totals line items priced by the user — materials and labor per line — and applies a contingency percentage the user chooses. The lines are a roof at $6,500 of materials and $4,000 of labor, a kitchen at $7,000 and $5,000, a bathroom at $3,200 and $2,800, flooring at $3,600 and $2,400, and interior paint at $1,000 and $2,000, every figure an illustrative entry. Materials total $21,300 and labor $16,200, for a subtotal of $37,500; a 12% contingency adds $4,500 for a grand total of $42,000. The build checks that this grand total is the base deal's rehab figure, so the ceiling of $203,000 in the section above is the rule fed directly from the budget tool's bottom line.

The budget tool reports two totals, and the rule takes only one number, so which total is carried matters. Carry the $37,500 subtotal instead of the grand total, with the $350,000 ARV, $190,000 purchase price, $12,000 holding costs, and 5% selling costs unchanged, and the ceiling rises to $207,500 — higher by $4,500, the contingency amount to the dollar, which the build also checks — while the projected profit at the entered price rises from $88,500 to $93,000 by the same amount. The contingency is a reserve against work not yet found. Carrying it into the rule lowers the ceiling by the reserve before anything is found, which is the only way the convention can hold room for it: the flip model applies no contingency of its own and adds nothing to the figure it is given.

Rehab is also the one input that moves both outputs by the same amount, because it is subtracted whole from each. Enter $54,000 in place of $42,000 on the base deal, with the $190,000 purchase price and every other input as before, and the ceiling falls to $191,000 while the projected profit falls to $76,500 — each by exactly $12,000, which the build checks — with the return on cost at 28%. The engine has no cost data of its own; the figure is whatever is entered, and the budget tool exists so that the figure can be built rather than guessed.

Profit against margin, one input at a time

The remaining inputs reach the profit line and, with one exception, leave the ceiling alone. The exception is the after-repair value, which appears in both formulas and sets the selling-cost figure besides. Enter $330,000 instead of $350,000 and keep the $42,000 rehab, $190,000 purchase price, $12,000 holding costs, and 5% selling costs: the ceiling falls to $189,000, now $1,000 below the entered purchase price, so the price has moved from inside the convention to outside it without the price itself changing. Selling costs shrink to $16,500, and the projected profit falls to $69,500, a 26.7% return on $260,500 invested. The profit fell by $19,000, less than the $20,000 change in ARV, because selling costs are a share of the ARV and fell with it. It is still a positive figure: a ceiling below the entered price is not a loss, only a price the convention would not have reached.

The purchase price has no effect on the ceiling and a dollar-for-dollar effect on profit. Enter $198,000, still under the ceiling but with the headroom cut to $5,000, on the same $350,000 ARV, $42,000 rehab, $12,000 holding costs, and 5% selling costs: the ceiling reads $203,000 as before, selling costs read $17,500 as before — the build checks that they did not move, because they are a share of the ARV and not of the price — and the projected profit is $80,500 on $269,500 invested, 29.9% on cost, lower than the base run by exactly the $8,000 added to the price. Raise the price further, to $290,000 on the same other inputs, and the entered costs overtake the entered ARV: total invested is $361,500, the projected profit is -$11,500, and the return on cost is -3.2%. The ceiling still reads $203,000. The sign change is a statement about the inputs — their sum exceeds the ARV as entered — and nothing more; the engine does not flag it, and it does not compare the price to the ceiling on the way to the result.

Holding costs are one dollar amount for the whole project, as entered, and they appear only in the profit line. Enter $18,000 in place of $12,000 on the base deal, with the $190,000 purchase price unchanged, and the ceiling is still $203,000, while the projected profit drops to $82,500 at 30.8% on cost. Leave the field blank and the engine treats it as zero: profit rises to $100,500 on $249,500 invested, 40.3%. There is no month count and no monthly rate behind this input — the model has no timeline — so a project that runs long, or a loan that carries interest while it does, is reflected only if the figure entered already includes it.

Selling costs are entered as a percentage and applied to the ARV. At 9% rather than 5% on the base deal, with the $190,000 purchase price unchanged, they rise to $31,500 and the projected profit falls to $74,500, 27% on cost; at 0% they vanish and the profit is $106,000 on $244,000 invested, 43.4%, against 33.8% in the base run. The return on cost moves in both its numerator and its denominator when this input changes, because the engine counts selling costs in total invested even though they are paid out of the sale rather than up front. That is a convention of this engine, chosen so that the profit line and the total-invested line always reconcile: the percentage is the profit over every dollar the model counts as spent, including the dollars that leave through the closing statement.

When the rule says nothing

The ceiling has three terms and is silent about everything else. It does not know the purchase price, so it cannot say whether the entered price is inside the convention; it does not know the holding costs or the selling costs, so the margin it reserves is the only room those costs have; and it does not know how long the project takes, what it costs to finance, or what tax falls on the result. The profit line nets the holding and selling costs as entered, and that is the end of what the model computes. Across its 6 inputs there is no loan — no interest rate, term, points, or lender fees — so any financing cost counts only if it is inside the holding figure; there is no timeline, so a longer project changes nothing unless the holding figure is raised; there are no income taxes on the profit; the ARV is treated as fixed between purchase and sale; and neither the ARV nor the rehab budget is estimated — both are entered, and every output inherits whatever error they carry. The methodology page states these omissions for every calculator on the site, and the guide What these models leave out walks through each one. The rest of the guides cover the site's other calculators.

Inside those limits the engine computes whatever it is given, and outside them it answers with a message instead of a number. A selling-cost percentage beyond its bound — 150%, say — returns "Enter a selling cost percentage of 100% or less." and a negative rule percentage returns the message quoted in the ceiling series above. Every money field has a floor at zero and an upper limit the engine states in words when it is crossed, and the ARV and the purchase price must also be greater than zero. Within the bounds nothing is refused: a purchase price above the ceiling, a rule percentage of 100%, holding costs larger than the margin — each produces a result, and reading the result is the reader's work.

Readings the ceiling does not support

• Carrying the budget subtotal into the rule. The subtotal leaves the contingency out. On the guide's budget it raises the ceiling from $203,000 to $207,500, exactly the $4,500 reserve, and the flip model will not put it back.
• Reading the 100% case as the rule switched off. It is the rule with nothing held back: a ceiling of $308,000, the ARV less the rehab, and a loss of $29,500 when bought at — holding plus selling costs, by construction.
• Treating the percentage as a fact of the engine. It is a field with a default. 72% is as valid an entry as 70% and gives a ceiling of $210,000 on the base deal; the engine enforces bounds, not a list.
• Reading selling costs off the purchase price. The engine applies the percentage to the ARV. On the base deal that is $17,500, and the figure stays $17,500 when the purchase price is raised to $198,000.
• Reading a ceiling below the entered price as a loss. In the lower-ARV run the ceiling is $189,000 against a $190,000 price, and the projected profit is still $69,500. The ceiling and the profit line are separate chains; only the ARV links them.

Frequently asked questions

Not financial advice: a general educational guide to a screening convention, not a recommendation to offer, buy, or flip any property. The model excludes financing structure, income taxes on the profit, and market movement during the project — verify every deal with local comparables and qualified professionals. Values are processed locally in your browser and never transmitted. See the methodology page.