ADU Rules and Cost in Arizona

A state statute tells you what a jurisdiction must allow. It does not tell you how long the review will take, whether the building department has reviewed a factory-built submittal before, or whether local contractors know how to prepare a pad to a module manufacturer's tolerance. Those three things determine the experience, and they vary enormously between municipalities inside the same state.

Phoenix is generally the path of least resistance in Arizona for a simple reason: volume creates competence. A building department that has processed dozens of accessory dwelling permits has a checklist, a plans examiner who knows the statute, and an inspector who has seen a module set before. A rural jurisdiction processing its first one will get there, but the first submittal frequently comes back with comments that reflect unfamiliarity rather than actual deficiencies.

The second thing that varies is utility capacity. Statutes generally do not compel a water or sewer district to provide service that does not exist, so in areas at the edge of municipal infrastructure the binding constraint is a capacity letter rather than a zoning approval. Ask that question before you spend money on drawings.

The third is the assessor. Adding a permitted dwelling unit changes the assessed value of the parcel, and in states with rate caps or reassessment triggers the tax consequence can be material. It is almost always still worth doing where rents support the build, but it should be in the model rather than a surprise in the following tax year.

Factory-built detached unit. The whole building arrives finished and is set in hours. Structure runs about $332 per square foot at the Signature finish tier in this state. Fabrication overlaps permitting, so the calendar compresses. The trade-offs are real: you need truck or crane access to the pad, and the design is constrained to what can travel legally on a road.

Site-built detached unit. Complete design freedom and no transport constraint, at the cost of a schedule that runs six to fourteen months in most markets and a budget exposed to weather, subcontractor availability and material price movement across that whole period. For an unusual lot or a highly specific architectural requirement it is sometimes the only option.

Garage or basement conversion. Appears cheapest and frequently is not. The existing slab has no vapor barrier and no perimeter insulation, the ceiling height may be below the habitable minimum, the header over the vehicle opening was never sized to carry a wall, and there is no independent utility service. Once those are corrected the cost commonly lands within ten percent of a new factory unit — except that you have spent it on a building with no warranty, no new-construction envelope, and one less parking space.

The honest summary: conversions make sense when the existing structure is already sound, tall enough and dry. Otherwise the factory unit wins on cost per usable square foot and wins decisively on schedule certainty.

Cooling load rather than heating load drives the mechanical specification, and a unit with substantial west-facing glazing needs meaningfully more capacity than a square-footage rule of thumb suggests. Roof structures are commonly specified solar-ready from the outset because the payback in this market is short enough that most owners add panels within a few years.

The order of operations matters because two clocks run at once. A factory-built project sequences roughly like this: site survey and soils information first, then the foundation and site plan submitted for permit, then fabrication released to the production queue while that permit is in review, then foundation construction, then delivery and set, then utility connection and final inspection.

The reason this compresses the schedule relative to site-built construction is that fabrication and permitting overlap. On a site-built project nothing is framed until the permit issues. On a factory project the module is being built while the plans examiner works, so the permit clock and the build clock consume the same calendar days rather than consecutive ones. Statutory allowance in larger municipalities.

Two documents cause most of the delay when they are missing: a current survey showing setbacks and easements, and evidence of utility capacity at the service point. Both can be obtained before the permit application, and both routinely are not.

A factory runs a queue. When you reserve a slot you are buying a position in that queue, and the honest answer to "how fast" depends on three things: how many slots are already committed, the lead time on the specific materials your configuration calls for, and whether your permit clears before your slot opens.

Current windows run 30 to 90 days from released engineering to delivery for standard configurations. Smaller units with Essential finishes sit at the fast end. Anything with a ballistic package, a hurricane-rated envelope, custom glazing or a second story sits at the slow end because those components have their own supply lead times.

The reason this beats a site-built schedule is not that the factory works faster on any given day. It is that permitting and fabrication run in parallel instead of in sequence, and the on-site set is measured in hours rather than months. Weather stops a framing crew; it does not stop a production line.

Windows move with demand and material availability. A slot quoted today holds while capacity lasts.