The received wisdom is that conversion is cheaper. It usually is. This screen is about the specific, checkable ways that stops being true — and about which single unknown would most change your answer.
7 factsthree of which can end the comparison outrightCeiling height after the floor build-up, slab and footing condition, whether the garage is attached, whether a compliant escape opening exists, whether the parcel is in a mapped flood zone, whether replacement parking can be required, and whether a buildable pad survives the setbacks. Every weight is published on this page.
On an empty form the tool reports level, 0 to 0 — and names the ceiling height as the unknown to go and settle first, because it is worth up to five points, can remove the conversion route outright, and costs nothing but a tape measure.
Change one thing and the default assumption reverses fast: a cracked slab or one with no perimeter footing scores new build by 3, a mapped flood zone new build by 4, and a measured ceiling under 7 feet blocks the conversion entirely.
A good garage — 96 inches of height, sound slab, attached, an existing egress opening, no flood zone, no replacement parking required — scores conversion 12 to 1. Both results come from the same published table.
This is a screen to take to a designer and a planning counter, never a determination that you may build either one, and never a quote.
Seven facts about your garage
“I don’t know” is the default and a valid state — the tool tells you which unknown to go and settle first. Nothing is emailed.
—Which route is favored
—The one unknown that would most change this
—Weights, in full
—Pushing toward conversion
—Pushing toward a new build
—Your two figures
—Timing, measured (one city)
—What this is not
What this assumed—
A screen over seven facts, with every weight published on this page. Not a determination that you may build either one, not a flood determination, not a structural opinion and not a quote. Nothing here is an instruction to perform structural, electrical or gas work. HyreADU does not design, permit, build or finance ADUs.
Conversion is usually cheaper. The interesting question is when it stops being
The advice is not wrong. A garage has a slab, four walls and a roof, and those are real money you do not have to spend again.
The advice is incomplete, because a garage was not built to be lived in, and the list of things that separates a garage from a dwelling is longer than the list of things it already has.
Start from what a garage actually is. In most of the country it is a slab poured at or near grade, sloped to the door so water runs out, sitting on a thickened edge rather than a footing designed to carry a heated, occupied, code-compliant room.
The walls are usually uninsulated, often unfinished on the inside, and sometimes not sheathed the way an exterior wall of a dwelling has to be. There is rarely a vapour retarder under the slab.
There is frequently no compliant escape and rescue opening anywhere in the building. The ceiling may be at seven feet on the nose, or under it.
Every one of those is a line item, and any one of them can close the gap on its own. That is the whole point of this page. It is not that conversions are secretly expensive — most are not.
It is that the ones that go wrong go wrong for a small number of specific, checkable reasons, and the check costs an afternoon and a tape measure while discovering it late costs a set of drawings and a change order.
The five that matter, in the order they close the gap. A slab or footing that cannot carry a dwelling.
A ceiling that will not clear seven feet once a floor build-up and a ceiling finish are added.
A mapped flood zone, where converting a garage almost certainly trips the substantial-improvement threshold and pulls the lowest floor up to the base flood elevation. A jurisdiction that will make you replace the parking.
And an absent escape and rescue opening in a masonry or shear wall, which is structural work rather than glazing.
And one that is not on that list, though everyone expects it to be. This is the single most common objection people brace for, and in California it is not available to the agency.
It cannot make you replace the parking the garage was providing. In California the agency is forbidden by statute (Gov. Code § 66314(d)(11)) from requiring it: "When a garage, carport, covered parking structure, or uncovered parking space is demolished in conjunction with the construction of an accessory dwelling unit or converted to an accessory dwelling unit, the local agency shall not require that those offstreet parking spaces be replaced." That is California law.
It is not yours unless you are in California, and it is one of the two inputs on this form most worth confirming rather than assuming.
What you lose is not a cost line, and it is the part budgets never carry. You lose the garage.
Not just a parking space: the workbench, the bicycles, the bins, the ladder, the boxes that have lived there for a decade.
On a lot with no basement and no shed, converting the garage means that storage has to go somewhere, and "somewhere" is frequently a new shed, a rented unit, or a genuinely smaller life.
A new detached build keeps the garage. That is a real difference between the two routes and it never appears in a bid from either side.
The constraint that argues the other way is the footprint. A conversion is stuck with the garage: its dimensions, its orientation, its window positions, its relationship to the street and to the neighbor.
If the garage is 20 by 20 and faces a blank fence, that is the unit you are getting.
A new build chooses where the light comes from, where the door faces, and how large it is inside whatever the setbacks leave.
People who have lived in a converted garage they did not design tend to talk about the orientation more than the money.
HYRE analysis, assembled from the model-code and statutory material sourced on this page. It is not a cost ranking: this site publishes no ADU price and holds no dataset of bids.
Every weight this tool applies, published in full
A screen like this is an editorial judgment dressed as arithmetic, and hiding the arithmetic would be the dishonest version of it. Below is the complete table. It is the table that runs — the engine has no other rules — so you can disagree with one row rather than with a number.
Input
Answer
Weight
Why it is set there
Ceiling height
Under 84 in (7 ft)
Blocks conversion
IRC R305.1, ceiling height sets 7 feet for habitable space and this is the figure that decides whether the space can be a dwelling at all. Raising a roof to fix it is not a conversion, it is a rebuild with a slab you kept.
84–89 in
+2 new build
It clears today and may not once a floor build-up, a ceiling finish and any duct crossing the space have taken their share. This is the band where projects get discovered late.
90–95 in
+1 conversion
Clears with a little to spare. Not much, and the build-up is still worth drawing before anyone commits.
96 in or more
+3 conversion
Room for a floor build-up, a finished ceiling and a service run, and still over seven feet. This is the case where the received wisdom holds.
Slab and footing
Cracked, settled, or no perimeter footing
+3 new build
Underpinning or replacing a slab is the single line that most often closes the gap between the two routes, because it removes the main thing the conversion was getting for free.
Level, uncracked, footing confirmed
+3 conversion
The largest thing a conversion gets for nothing. Weighted symmetrically with the bad case, because it is the same fact pointing the other way.
Flood zone
In a mapped special flood hazard area
+4 new build
Under 44 CFR § 59.1 an improvement costing 50 per cent or more of the structure’s pre-improvement market value is a substantial improvement, and a garage has a low market value. 44 CFR § 60.3(c)(2) then pulls the lowest floor to or above base flood elevation. Elevating a slab-on-grade garage is close to rebuilding it.
Not in a mapped SFHA
+1 conversion
A modest credit, not a large one: it removes a risk rather than creating an advantage.
Parking
The garage has to stay
Blocks conversion
A converted garage is not a garage. If the off-street space must remain, this is not a decision between these two routes — look at a unit above the garage or an attached ADU instead.
Replacement parking is required
+3 new build
A replacement stall has to physically fit on the lot after setbacks, and on the small lots where conversion is most attractive it frequently does not.
Replacement cannot be required here
+2 conversion
Where a state has removed this objection it removes the single thing people most often abandon a conversion over. Gov. Code § 66314(d)(11) is the example; most states have not written one.
Attached or detached
Attached to the house
+2 conversion
The shortest utility runs of any route. Water, drainage and the electrical panel are already at that wall, and trench length is most of what separates the two routes on utilities.
Detached
+1 new build
A detached garage carries much of the same trench a new build would, which removes part of the conversion’s advantage rather than all of it.
Escape and rescue opening
None that complies
+2 new build
IRC R310, emergency escape and rescue openings requires a net clear opening of 5.7 sq ft with minimum dimensions of 24 in high and 20 in wide, no more than 44 in above the floor. Cutting one into masonry or a braced wall is structural work.
One already exists
+1 conversion
A small credit. It removes one structural cut, not the whole scope.
Buildable pad
None survives setbacks and coverage
Blocks new build
A detached unit needs a rectangle that exists after rear and side setbacks and lot coverage. If it does not exist, the comparison is over and the conversion is the only route left on this lot — which is not the same as it being a good one.
A pad survives
+1 new build
Only a point. It establishes that a new build is genuinely on the table rather than a drawing, which is a precondition, not an advantage.
HYRE analysis. A lead of three points or more is reported as a decision; one or two points is reported as too close to call, because a one-point lead is not a real distinction.
Blockers are read before any weight and settle the answer on their own. Weights are a judgment about which facts most often close the cost gap — they are not derived from a dataset of outcomes, and we do not claim they are.
Which single unknown would most change your answer, and how the tool picks it
The output most people should read first is not the verdict. It is the line naming the one fact you do not yet have that would move the answer furthest — because that is the thing to go and get before you spend anything.
The rule, in full. Each input has a swing: the distance between the most pro-conversion and the most pro-new-build answer it can give.
Slab condition swings six points, ceiling height five, flood zone five, parking five, attachment and egress three, buildable pad one.
An input that can remove a route outright gets a further ten, because a blocker settles the question rather than nudging it. The unanswered input with the highest total is reported.
Ties go to whichever comes first in the declared order, which puts the free tape-measure questions ahead of the ones needing a phone call.
Which means, on an empty form, the answer is always the ceiling height. That is not a bug and it is not a coincidence.
Ceiling height ties with parking on score, and it wins the tie deliberately, because you can settle it yourself in ten minutes for nothing while the parking question needs a jurisdiction to answer it.
If a tool is going to send you to do one thing before anything else, it should be the free one.
Once everything is answered, the question changes shape. The tool then looks for an input whose swing is larger than the current margin — one where being wrong could reverse the verdict — and names it.
If the margin is wider than any single swing, it says so, and points at the three blockers instead, because those can still reverse an answer no weight can.
How to measure the one it will send you to. This is the figure that decides whether a basement or an above-garage space can become a dwelling at all, and it is the one most often discovered after money has been spent on drawings.
Measure it first. Measure from the slab to the underside of the structure, then again to the lowest thing crossing the space — a beam, a garage door track, a duct, a light.
Then subtract what you will add: a floor assembly over the sloping slab, insulation, and a ceiling finish. The governing number is the smallest one after the build-up comes off, and it is almost never the number people remember.
HYRE analysis. The swing figures above are the ones in the engine, not a simplification of them.
The one part of this comparison we can measure rather than assert
Conversions are usually described as faster. On the permit side, in the one city where we could measure it, that is true and the gap is large. Read the caveats under the figures before you carry them anywhere.
102 daysMedian application to issuance, ADU-flagged alteration and repair permitsLos Angeles, n = 9,683. LADBS Building and Safety — Building Permits Issued from 2020 to Present (N)
218 daysMedian application to issuance, ADU-flagged new-build permitsLos Angeles, n = 9,145. LADBS Building and Safety — Building Permits Issued from 2020 to Present (N)
35%of alteration and repair permits issued within 60 daysAgainst the 60-day statutory decision clock.
7%of new-build permits issued within 60 daysThe same clock, the same city, a different route.
Alteration / repair (conversion proxy)102 days
n = 9,683, interquartile range 38 to 203 days
Addition175 days
n = 8,360, shown for context — an addition is neither of the two routes on this page
New build218 days
n = 9,145, interquartile range 130 to 373 days
Read these carefully. HyreADU calculation on LADBS Building and Safety — Building Permits Issued from 2020 to Present (N), retrieved 2026-09-05, published with its method at permit times by city.
Permit type is a proxy for route: an alteration permit is usually a conversion and a new-build permit is usually a new building, but the dataset does not say so and we have not verified any individual record.
It is one city, and one city is not a country. These are calendar days from the published application-side date to issuance, on permits that issued.
The statutory 60 days runs from a COMPLETED application, and none of the three extracts marks the completeness determination.
The gap between the two therefore includes every day an applicant spent curing an incomplete set, paying fees, answering plan-check comments, or asking for delay.
A median above 60 days is not, by itself, a finding that a city missed the shot clock. And none of it says anything about how long the construction takes — only how long the paperwork took to issue.
The four conversion-specific drivers, in the detail the bid will not give you
Four conversion-specific drivers turn a cheap-looking garage conversion into an expensive one. Each is checkable before you commission anything, and each is a question to put to a designer or the building department rather than a thing to do yourself.
The slab, and what is under its edge
A garage slab is designed for a car, not a room. It slopes to the door, usually a quarter-inch per foot, so that water leaves. A habitable floor does not slope.
Levelling it means either a poured self-levelling topping, which adds height you may not have, or a framed floor over the slab, which adds more.
Then the edge. A dwelling’s exterior walls sit on a footing sized and embedded for the load and the frost depth. A garage frequently has a thickened slab edge instead.
Whether that is adequate is a structural question with a real answer, and it is the answer that most often closes the cost gap — because if it is not adequate, the remedy is underpinning or a new perimeter footing poured against an existing slab, which is slow, awkward work.
And underneath. A vapour retarder under the slab is standard for habitable space and usually absent under a garage. Retrofitting it is not possible; the alternatives are a floor assembly designed for the condition, which again costs height. This is why ceiling height and slab condition are not really two independent questions.
What to do with this. Ask a designer or a structural engineer what is under the perimeter, and ask for it in writing before drawings. Do not dig to find out and do not remove anything structural. This page is for understanding a proposal, not for performing structural work.
The ceiling, measured after the build-up
IRC R305.1, ceiling height: "Habitable space, hallways and portions of basements containing these spaces shall have a ceiling height of not less than 7 feet (2134 mm). Bathrooms, toilet rooms and laundry rooms shall have a ceiling height of not less than 6 feet 8 inches (2032 mm)."
The trap is the order of operations. People measure the garage, find seven foot four, and conclude it works.
Then the floor is levelled and insulated, which takes three or four inches, and a ceiling finish goes on, which takes another, and the finished room is under seven feet.
There is no design that recovers those inches, because the slab is where it is and the structure is where it is.
Sloped ceilings buy you something. "For rooms with sloped ceilings, the required floor area of the room shall have a ceiling height of not less than 5 feet (1524 mm) and not less than 50 percent of the required floor area shall have a ceiling height of not less than 7 feet (2134 mm)." Which is why a garage with an open truss or a gable can sometimes work where a flat-ceilinged one cannot — but only if the trusses can be opened up at all, and a standard fink truss cannot be cut without engineering.
Model-code caveat, stated every time. The IRC is a model document with no legal force until a jurisdiction adopts it, and adopting jurisdictions routinely amend it. Confirm every figure with the authority that will inspect the work.
Neither retrieved document states its IRC edition on its face. The numbering matches the 2021 structure and identical language appears in a state adoption of the 2018 edition, so we present these as model-code figures without an edition label rather than assert one we did not verify.
ICC’s own code library returned 403 to automated retrieval on 2026-09-05. The text below comes from jurisdiction-published copies, named in the sources.
The flood rule almost nobody checks first
44 CFR § 59.1: "Substantial improvement means any reconstruction, rehabilitation, addition, or other improvement of a structure, the cost of which equals or exceeds 50 percent of the market value of the structure before the “start of construction” of the improvement. This term includes structures which have incurred “substantial damage”, regardless of the actual repair work performed."
Read the denominator. Fifty per cent of the market value of the structure, before construction. A detached garage is a low-value structure.
A conversion that turns it into a dwelling will, in a great many cases, cost more than half of what the garage is worth — which means the project is a substantial improvement, and it is treated as if it were new construction.
And then 44 CFR § 60.3(c)(2): "Require that all new construction and substantial improvements of residential structures within Zones A1–30, AE and AH zones on the community’s FIRM have the lowest floor (including basement) elevated to or above the base flood level"
Which is the whole problem. Elevating the lowest floor of a slab-on-grade garage to base flood elevation is not a modification of a conversion; it is a different project that happens to be on the same footprint.
A new build, by contrast, is designed to that elevation from the start, which is ordinary rather than remedial. In a mapped flood zone the received wisdom can invert completely.
What this is not. These are the federal minimum criteria a community adopts to participate in the National Flood Insurance Program.
Communities may and often do adopt stricter versions, including cumulative tests that add improvements together over a period of years. The market-value determination is made locally. This site does not determine flood zones and does not read your community’s ordinance.
Look your parcel up on the FEMA Flood Map Service Center and then ask your local floodplain administrator.
The two questions are "what zone is this parcel in" and "how does this community administer substantial improvement", and the second is the one that varies.
The parking space, and the storage nobody budgets
The legal half. This is the single most common objection people brace for, and in California it is not available to the agency. It cannot make you replace the parking the garage was providing.
The California text (Gov. Code § 66314(d)(11)) is: "When a garage, carport, covered parking structure, or uncovered parking space is demolished in conjunction with the construction of an accessory dwelling unit or converted to an accessory dwelling unit, the local agency shall not require that those offstreet parking spaces be replaced." Several other states have written their own version of the same idea; several have not.
This is a specific legal question with a specific answer, and it is worth getting before anything else on the list that costs money.
There is a separate cap worth knowing about too (Gov. Code § 66314(d)(10)(A)): "Parking requirements for accessory dwelling units shall not exceed one parking space per accessory dwelling unit or per bedroom, whichever is less. These spaces may be provided as tandem parking on a driveway." A parking requirement and a replacement-parking requirement are different instruments, and a jurisdiction can have removed one and not the other.
The practical half is not a legal question at all. Even where no replacement space can be required, you still lose the place the car was, and in a street-parking-constrained neighborhood that is a daily cost paid by the household rather than by the budget.
Some people find it trivial. Some find it is the thing they most regret. Neither reaction shows up in a bid.
And the storage. The garage was doing work. If there is no basement, no attic and no shed, that work has to move somewhere. Price the shed, or price the rented unit for as long as you expect to need it, and put the number in the conversion column where it belongs.
One input changed at a time, from an empty form
Every figure below is the engine’s actual output, produced by running it with the stated single change and nothing else. The baseline is the form as it loads: nothing measured, nothing known.
What changed
Score
Verdict
The unknown it then names
Nothing — the empty form
Conversion 0, new build 0
Level. Nothing entered separates the two routes yet.
Slab-to-structure height. Worth up to 5 points, and it can remove a route outright. Six other inputs still unanswered.
Ceiling measured at 96 in
Conversion 3, new build 0
Conversion favored by 3 points.
The parking question — same score, but it now needs a jurisdiction rather than a tape.
Ceiling measured at 86 in
Conversion 0, new build 2
Too close to call: a 2-point lead is not a real distinction.
The parking question. The ceiling is answered but sits in the band where a build-up could still end it.
Ceiling measured at 82 in
Blocked
A detached new build, by elimination rather than by preference.
Still the parking question — but the conversion is gone regardless of how it is answered.
Slab cracked or without a footing
Conversion 0, new build 3
New build favored by 3 points.
Ceiling height. One driver has fired; the free measurement is still outstanding.
Parcel in a mapped flood zone
Conversion 0, new build 4
New build favored by 4 points.
Ceiling height.
Replacement parking required
Conversion 0, new build 3
New build favored by 3 points.
Ceiling height.
The garage has to stay
Blocked
A detached new build, by elimination. A converted garage is not a garage.
Ceiling height — though the honest next move is the type selector, because a unit above the garage may be the real answer.
No buildable pad after setbacks
Blocked
A garage conversion, by elimination. The new build was removed by your own answer.
Ceiling height — which now matters more, because it is the only route left and it may not be available either.
Produced by running the published engine. Note the shape of the results: a single bad fact is enough to reverse the default assumption, and three of the nine rows end the comparison outright rather than scoring it. That is the finding this page exists for.
The trade that is not about money at all
What the conversion buys you
A shell that already exists, which is genuinely most of the envelope: foundation, walls, roof. Where the slab is sound and the ceiling is generous, that is a real and large saving and the received wisdom is simply correct.
Short utility runs, if the garage is attached or close. Trench length is one of the biggest single differences between the two routes, and it does not scale with floor area, which is why it hides inside a per-square-foot comparison.
A yard that stays a yard. A detached unit consumes open ground, and on a small lot the garden is the thing being spent. The conversion spends a building you already treat as not-quite-house.
Less disruption and, on the evidence above, a faster permit route. No excavation next to the house, no pad, and a review path that in Los Angeles issued in about half the time.
What the new build buys you
The freedom to choose the plan. Orientation, light, entrance, ceiling height, room sizes, where the bathroom goes. A conversion inherits all of that from a building designed to hold a car, and no amount of joinery undoes an east-facing wall that faces a fence.
Compliance designed in rather than retrofitted. Insulation in the wall as it is built, an egress window in the opening as it is framed, a floor that was always level, a footing sized for the load. Each of those is cheap when it is designed and expensive when it is remedial.
Your garage. Parking, storage, the workbench. This is the line that never appears in either bid and is often the one people feel most.
A cleaner path where a rule bites. In a mapped flood zone, elevating a new structure is ordinary design and elevating an existing slab is close to a rebuild. Where a code trigger fires, new construction is frequently the less contorted answer.
Neither column is a recommendation. Which of these you weigh more is a question about how you live, and the tool cannot see that — which is why the verdict is reported as a lead in points rather than as an instruction.
What to establish before either route gets any money
The first three you can do yourself today for nothing. The rest are short questions to specific people, and each has a specific answer.
Measure the garage ceiling to the lowest obstruction, then subtract the build-up
Slab to underside of structure, then slab to the lowest beam, duct or door track. Then subtract what a levelled, insulated floor and a finished ceiling will take. Compare the remainder with 7 feet.
Measure every existing opening and compute its net clear area
Height times width of the opening you can actually pass through, in inches, divided by 144. Compare with 5.7 square feet, and check the 44-inch maximum to the bottom of the clear opening.
That 44 inches is measured to the bottom of the actual clear opening, not to the sill height of the rough opening. The ICC Commentary is explicit about it, and it is a common and expensive misreading.
Look the parcel up on the FEMA Flood Map Service Center
It is free and it takes two minutes. If the parcel is in a mapped special flood hazard area, stop and speak to the local floodplain administrator about how this community administers substantial improvement before you compare anything else.
Ask planning, in these words, whether replacement parking can be required
Not "do I need parking" — ask whether the jurisdiction may require replacement of the off-street spaces lost when a garage is converted to an ADU, and ask for the section number. Several states have removed that power. Several have not.
Ask a structural engineer or designer what is under the garage perimeter
Whether there is a footing, at what depth, and whether it can carry a habitable structure under the adopted code. Get it in writing before drawings, not after. This is a question to ask, never work to do yourself.
Get both routes bid on the same line items
Design, permits and fees, site and foundation, utilities, building, wet rooms, contingency. Two bids in different shapes cannot be compared, and the conversion bid should carry the slab, footing, insulation, egress and flood lines explicitly rather than as an allowance. The ADU cost calculator is the structure to hold them in.
Price the storage you are about to lose
A shed, a rented unit, or the honest admission that the bikes are going. Put a number on it and add it to the conversion column. It is a real cost of that route and neither bidder will mention it.
The five ways this decision is actually got wrong
Comparing a lump sum with an itemized budget. The conversion arrives as one number and the new build arrives as a schedule, and the difference gets read as a saving. It is usually a difference in disclosure. Insist on the same lines from both.
Measuring the ceiling before the floor exists. Covered above, and it is the single most common way a conversion is discovered to be impossible after the drawings are paid for.
Assuming another state’s parking rule. California’s prohibition on replacement parking is quoted endlessly and is quoted accurately; it is also California law. Guides that present it as a national position are wrong.
Seven states, seven different drafting styles, and the same three objections removed in most of them: replacement parking, owner-occupancy, and a requirement that the unit match the house.
Several also attach a self-executing penalty — California deems the application approved, Arizona removes the city’s limits entirely, Montana voids the regulations — which tells you legislatures expected local resistance and drafted against it.
Treating a per-square-foot rate as comparable across routes. A rate that includes a foundation and a rate that assumes one exists are not the same rate, and neither includes trench length, which is one of the largest differences between the two. This site publishes no ADU price for exactly this reason.
Forgetting that a conversion has to be permitted as a dwelling too. The moment the garage becomes habitable space, the dwelling rules apply to it: egress, ceiling height, insulation and energy compliance, fire separation from the house, and the local ADU ordinance. "It is only a conversion" is a sentence about effort, not about jurisdiction.
HYRE analysis, drawn from the code and statutory material on this page. It is not a claim about frequency: we hold no dataset of failed projects and do not assert one.
What this tool cannot do, stated plainly
The comparison tool cannot tell you that you may build either one. Eligibility is a determination about a specific parcel that only the local authority can make. The feasibility checker exists for that question and it does not say yes either.
It cannot price anything. There is no cost per square foot in this tool, no national ADU average, no local fee schedule and no dataset of bids. The only dollars it handles are the two figures you type, and all it does with them is subtract.
It cannot see your building. Not the slab edge, not the truss type, not the moisture in the wall, not the panel capacity, not the distance to the sewer, not the tree between the garage and the main. Every one of those can matter more than any weight on this page.
It is not a flood determination and not a structural opinion. It reads back a federal rule and asks you a question about a map. Your floodplain administrator determines the first and an engineer the second.
And it is never an instruction to do the work. Structural alterations, electrical service, gas and fire separation are permitted, inspected trades. This page exists to help you understand a proposal and ask better questions of the people who will build it.
HyreADU does not design, permit or build accessory dwelling units. It does not appraise, lend, invest, let property or prepare taxes, it does not rank or refer contractors, and it takes no referral fee from anyone who does any of those things. That is what allows these tools to conclude that the numbers do not work, that the answer is to keep the garage, or that the honest next step is a question at the planning counter rather than a deposit.
How this calculator works
Blockers are read first, and settle the answer on their own. Then the weights are added. There is nothing else in the engine.
BLOCKERS ceiling < 84 in -> conversion removed
garage must stay -> conversion removed
no buildable pad -> new build removed
WEIGHTS ceiling >= 96 in +3 conv | 90-95 in +1 conv | 84-89 in +2 new
slab sound +3 conv | slab cracked / no footing +3 new
attached garage +2 conv | detached garage +1 new
egress opening +1 conv | no compliant opening +2 new
not in a flood zone +1 conv | mapped flood zone +4 new
no replacement req. +2 conv | replacement required +3 new
| buildable pad exists +1 new
VERDICT margin = conversion - new build
|margin| >= 3 -> decided
|margin| 1-2 -> too close to call
margin = 0 -> level
UNKNOWN rank each unanswered input by (swing + 10 if it can block),
ties broken by declared order, which puts the free
tape-measure question first
The full table with the reasoning behind every row is published above.
Weights are an editorial judgment about which facts most often close the cost gap; they are not derived from a dataset of project outcomes, and we do not claim they are.
The two dollar fields are not weighted at all — they are subtracted, and the difference is reported as the headroom the conversion work has to fit inside.
What each input means
Inputs on this tool, in the order they appear on the form.
Input
What it is actually asking
Ceiling height
Slab to the underside of the structure, or to the lowest beam, duct or garage-door track — whichever is smaller. Where you find it: a tape measure, ten minutes, today. Then subtract the floor build-up and ceiling finish you will add; the governing number is what is left.
Slab and perimeter footing
Whether the slab is level and uncracked and whether a footing sized for a dwelling exists under the perimeter. Where you find it: a structural engineer or a designer, in writing, before drawings. It is not visible and it is not a thing to dig for yourself.
Where the garage sits
Attached to the house or detached in the lot. Stands in for trench length to sewer and panel, which is one of the largest differences between the two routes and does not scale with floor area. Where you find it: you can see it.
Escape and rescue opening
Whether an existing opening gives 5.7 sq ft net clear, at least 24 in high and 20 in wide, with the bottom of the clear opening no more than 44 in above the floor. Where you find it: measure the opening you could actually climb through, multiply height by width in inches, divide by 144.
Mapped flood zone
Whether the parcel sits in a special flood hazard area. Where you find it: the FEMA Flood Map Service Center, free, two minutes — then your local floodplain administrator for how this community administers substantial improvement.
Parking
Whether replacement of the lost off-street space can be required, and whether you need the garage regardless. Where you find it: your planning counter. Ask for the section number, do not assume California’s rule applies to you.
Buildable pad
Whether a rectangle survives the rear and side setbacks and the lot coverage limit. Where you find it: your zoning ordinance, free to look up, or our size envelope calculator.
Your two figures
A conversion quote and a new-build quote, itemized on the same lines. Where you find it: from designers or builders who have permitted both routes in your city. Leave both at 0 until you have both — a one-sided number cannot settle a two-sided decision.
Worked examples
Including one where the naive answer misleads, which is the example most calculators leave out.
The empty form, which is the honest starting state
Nothing measured, nothing known. The tool reports level, conversion 0 to new build 0, and names the ceiling height as the unknown worth going to settle: up to five points, capable of removing the conversion outright, and free to answer.
Six other inputs remain open. This is the result most first-time readers should get, and the point of it is the instruction, not the score.
A good garage, everything confirmed
98 inches to the lowest obstruction, slab level with a confirmed footing, attached to the house, an existing opening over 5.7 sq ft, not in a flood zone, replacement parking cannot be required, and a pad does exist for a detached unit.
Result: conversion 12, new build 1 — favored by 11 points.
No single weight on the form can reverse that; only a blocker could, which is why the tool then names the three blocking questions instead of a weight. This is the case where the received wisdom is simply right.
The one where the naive answer misleads
The same homeowner, same garage, but the parcel turns out to be in a mapped special flood hazard area. Nothing about the building has changed.
The score moves to conversion 11, new build 5 — the conversion still leads, but a four-point trigger has fired that nobody priced, and the consequence behind it is not a line item: it is a requirement to elevate the lowest floor to or above the base flood level.
On a slab-on-grade garage that is not a modification of the conversion, it is a different project. The tool cannot tell you what it costs. It can tell you to go and ask before you sign anything, which is the whole job.
Two quotes, and what the gap actually has to cover
A detached garage with a cracked slab, no compliant opening, replacement parking required, 92 inches of height, a pad available, and quotes of $145,000 to convert against $265,000 to build new.
Score: new build 10, conversion 2. The money line reports that the conversion sits $120,000 under the new build — and that this is the headroom the slab, footing, insulation, egress and parking work has to fit inside.
Five triggers have fired. Whether $120,000 covers them is a question for the person writing the bid, and the right way to ask it is line by line.
What changes the result
Slab condition has the widest swing on the form, six points from best to worst, because it is the fact that most directly removes the thing a conversion was getting for free.
Ceiling height, flood zone and parking each swing five — but ceiling and parking can also block, which is why they outrank slab in the unknown ranking despite scoring lower.
A single bad fact is enough to reverse the default assumption. From an empty form, a cracked slab alone produces new build by three; a flood zone alone produces new build by four. That is the shape of the finding: the received wisdom does not degrade gradually, it flips on one discovery.
Three of the seven inputs can end the comparison rather than score it. A ceiling under seven feet, a garage that has to stay, and the absence of a buildable pad each remove a route outright.
When a blocker fires, the tool says the surviving route won by elimination rather than by preference, which is a different and weaker statement than winning on the weights — and it is deliberately worded that way, because the surviving route may still be a bad idea.
The two dollar fields never touch the score. They cannot: a figure you typed is not evidence about which route is structurally sensible, and letting a cheap quote outvote a code trigger is precisely the error this page exists to catch.
Local considerations
Two of the seven inputs are almost entirely local questions, and the tool asks them rather than answering them.
Parking. California prohibits an agency from requiring replacement of the off-street spaces lost when a garage is converted to an ADU (Gov. Code § 66314(d)(11)), and caps ADU parking at one space per unit or per bedroom, whichever is less (Gov. Code § 66314(d)(10)(A)).
Several other states have written their own version of the idea; several have written nothing, and in those places the objection is live. Two neighboring cities in the same metro can differ. Ask, and ask for the section number.
Flood. 44 CFR § 59.1 and 44 CFR § 60.3(c)(2) are the federal minimum criteria a community adopts to participate in the National Flood Insurance Program. These are the federal minimum criteria a community adopts to participate in the National Flood Insurance Program.
Communities may and often do adopt stricter versions, including cumulative tests that add improvements together over a period of years. The market-value determination is made locally. This site does not determine flood zones and does not read your community’s ordinance.
The code figures are model-code figures. The IRC is a model document with no legal force until a jurisdiction adopts it, and adopting jurisdictions routinely amend it.
Confirm every figure with the authority that will inspect the work. Neither retrieved document states its IRC edition on its face.
The numbering matches the 2021 structure and identical language appears in a state adoption of the 2018 edition, so we present these as model-code figures without an edition label rather than assert one we did not verify.
When not to use this
Do not read the verdict as permission. Eligibility is a determination about a specific parcel that only the local authority can make, and the feasibility checker will not say yes either.
Do not use it as a price. There is no cost per square foot here, no national ADU average and no local fee schedule — deliberately, because we hold none we could re-verify. The line-by-line budget for either route is the ADU cost calculator.
Do not use it as a flood determination or a structural opinion. It reads back a federal rule and asks you a question about a map.
Do not use it if the form is still genuinely open across all six ADU types — the type selector comes first, and if you must keep the garage it will point you at a unit above it instead of at either route here.
And nothing on this page is an instruction to perform work. A converted garage is a dwelling the moment it becomes habitable space: structural alterations, electrical service, gas and fire separation between units are permitted, inspected trades.
HyreADU does not design, permit, build, finance or appraise ADUs.
Is a garage conversion always cheaper than building a new ADU?
No, and the exceptions are specific rather than random. A garage was not built as habitable space: its slab slopes and may sit on a thickened edge rather than a footing, its walls are usually uninsulated, it rarely has a compliant escape and rescue opening, and its ceiling may not clear seven feet once a floor build-up and a ceiling finish are added. Any one of those can close the gap. In a mapped flood zone the comparison can invert outright.
What is the single most important thing to check first?
The ceiling height, measured to the lowest obstruction and then reduced by the floor build-up and ceiling finish you will add. It is free, it takes ten minutes, and it is the measurement most often discovered after drawings have been paid for. The model code sets seven feet for habitable space and no design recovers inches the slab and the structure have already taken.
Will I have to replace the parking the garage was providing?
That is a specific legal question with a specific answer, and it varies by state and city. California prohibits an agency from requiring it: Gov. Code § 66314(d)(11) says that where a garage is demolished or converted in conjunction with an ADU, "the local agency shall not require that those offstreet parking spaces be replaced." Most states have written no such sentence. Ask your planning counter for the section number rather than assuming either way.
Why does a flood zone matter so much to a conversion?
Under 44 CFR § 59.1 an improvement costing 50 per cent or more of the structure’s pre-improvement market value is a "substantial improvement". A detached garage is a low-value structure, so a conversion frequently crosses that line — and 44 CFR § 60.3(c)(2) then requires substantially improved residential structures in A zones to have the lowest floor elevated to or above the base flood level. Elevating a slab-on-grade garage is close to rebuilding it. A new build is designed to that elevation from the start.
Is a conversion faster?
On the permit side, in the one city where we could measure it, yes and by a wide margin. On ADU-flagged Los Angeles records, alteration and repair permits ran a median of 102 days from application to issuance against 218 days for new-build permits. Permit type is a proxy for route rather than a verified split, it is one city, and it says nothing about how long the construction takes.
Why does this tool not tell me what either route costs?
Because we hold no dataset of ADU bids and publish no cost per square foot, and a rate that assumes a foundation exists is not comparable with one that includes building it. The only dollars here are the two quotes you type, and all the tool does with them is subtract, so you can see how much headroom the conversion has before the triggers eat it. The line-by-line budget for either route lives on the ADU cost calculator.
How is this different from the ADU type selector?
The type selector chooses among six forms — detached, attached, garage conversion, basement, above-garage, prefab — from the lot, the parking rule, any existing volume, the intended use and privacy. It never looks at the garage itself. This page assumes you have already narrowed it to two routes and asks the questions about your specific garage that decide between them. Use the selector first if the form is still open.
What does "which unknown would most change the answer" actually mean?
Each input has a swing: the distance between its most pro-conversion and most pro-new-build answer. An input that can remove a route outright gets a further ten points, because a blocker settles the question rather than nudging it. The unanswered input with the highest total is named, and ties are broken in favor of the question you can answer yourself for free. The whole rule is published in the weights section.
Does the tool account for losing the garage?
It accounts for the legal half — whether a replacement space can be required — because that is checkable. It cannot account for the storage, the workbench and the daily parking, which are real and are frequently the part people most regret. Price a shed or a rented unit and put the figure in the conversion column yourself before you compare.
Can I do any of this work myself?
No part of this page is an instruction to perform work. Structural alterations to a slab, footing or shear wall, electrical service, gas, and fire separation between dwelling units are permitted, inspected trades, and a converted garage is a dwelling the moment it becomes habitable space. This is a tool for understanding a proposal and asking better questions of the people who will build it.
Does HyreADU build or design ADUs?
No. HyreADU does not design, permit, build, finance or appraise ADUs, and takes no referral fee from anyone who does. That is exactly what lets this tool tell you to keep the garage.
Sources and methodology
Figures dated 5 September 2026. Last reviewed .
Definitions — "Substantial improvement" and "Substantial damage" (U.S. Government Publishing Office, 44 CFR § 59.1 (2023 annual edition), retrieved 2026-09-05. Quoted verbatim. Source of the 50 per cent threshold that makes a garage conversion a substantial improvement in a mapped flood zone. eCFR refused automated retrieval on the date, so the govinfo annual edition was used and is cited as such.)
Flood plain management criteria for flood-prone areas (U.S. Government Publishing Office, 44 CFR § 60.3(c)(2) (2023 annual edition), retrieved 2026-09-05. Quoted verbatim. The elevation consequence that follows a substantial-improvement determination for a residential structure in an A zone.)
State ADU law — replacement parking prohibition and parking cap (California Legislature, Gov. Code § 66314, retrieved 2026-09-05. Quoted verbatim at § 66314(d)(11) and (d)(10)(A). California law, cited as the example of a state removing the objection — not as a national position.)
IRC R305.1 ceiling height (code handout #434) (City of Boise, Idaho, retrieved 2026-09-05. The seven-foot habitable ceiling height. A jurisdiction-published copy: ICC’s own library returned 403, and neither retrieved document states its IRC edition on its face.)
ADU permit application-to-issuance durations, by permit type (HyreADU Research Desk, on City of Los Angeles open data, retrieved 2026-09-05. Our own calculation, with its method, exclusions and caveats on the research page. Permit type is a proxy for route, not a verified conversion / new-build split.)
FEMA Flood Map Service Center (Federal Emergency Management Agency, retrieved 2026-09-05. Where a reader looks up whether their own parcel is in a mapped special flood hazard area. This tool makes no flood determination.)
Related
ADU type selectorIf the form is still open across all six, start there.
ADU cost calculatorThe dollars for either route, as lines you can audit. This page ships no rate.
Feasibility checkerWhether either route is permitted here at all. It will not say yes.