DEALINTEL
PRIVATE FIX & FLIP INTELLIGENCE
LOG INSIGN UP
Blog · Strategy · 11 min read

ADU Cost-Yield Math: When the Numbers Actually Work

ADUs are pitched as universal value-adds. The math is more selective. $140k all-in cost vs $1,950/month rent can pencil as a 12% yield or a 4% yield depending on how you account for cost, vacancy, and exit value. Here is the institutional underwrite.

Accessory Dwelling Units have a marketing problem: they are sold to operators as a universal value-add, with broad-stroke math ("rent will cover the loan") that holds only on a narrow band of properties.

The institutional question is not "should I build an ADU" but "does the ADU on this specific property, in this specific submarket, at this specific cost basis, return more than the alternative use of the same capital." Often yes. Sometimes no. The math is the answer.

The four ADU configurations

  • Detached new construction (separate building). 600–800 sqft typical. Most flexible, highest cost. $180–280/sqft all-in in most metros, $300+/sqft in California and Northeast.
  • Attached new construction (addition with kitchen + separate entrance). Shares wall with primary structure. $150–230/sqft. Often slightly cheaper than detached because shared structure but loses some privacy premium at rental.
  • Conversion of existing space (garage, basement). $80–160/sqft typically. Lowest cost path. Constrained by existing space's ceiling height, egress, plumbing routing.
  • Internal carve-out (in-law unit within existing structure). $60–120/sqft for finishes; main cost is reconfiguring the primary unit. Usually requires the existing kitchen + bath to be relocated or duplicated.

The honest cost stack

Most ADU pro formas underweight soft costs and miss contingency. The honest cost stack on a typical 700 sqft detached new-construction ADU:

  • Hard construction: $140,000 (700 × $200/sqft)
  • Design + plans: $6,500–14,000
  • Permit fees: $4,000–18,000 depending on jurisdiction (California permit fees alone can hit $25k)
  • Utility connections (separate water meter, electrical service): $4,000–18,000
  • Site work (foundation prep, drainage, utility trenching): $6,000–14,000
  • Landscaping / hardscape after construction: $3,000–8,000
  • Contingency (institutional standard 15%): $24,000–32,000
  • Carrying cost on the project loan during build (8–12 months): $8,000–18,000
  • Realistic total all-in: $195,000–245,000 on a 700 sqft detached ADU. Operators who budget $140k are budgeting hard cost only.

Three ways to compute yield (and they give different answers)

Yield method 1 — Gross rental yield

Annual rent ÷ total cost. The marketing number.

$1,950/month × 12 = $23,400. At $220,000 all-in cost: $23,400 ÷ $220,000 = 10.6% gross yield.

This is the number that sounds great in pitch decks. It does not account for vacancy, expenses, or the time value of money during the build.

Yield method 2 — Net operating yield (cap rate equivalent)

Net Operating Income ÷ total cost. Expenses included.

$23,400 gross rent − 30% operating expenses (vacancy, tax reassessment increment, insurance increment, management, maintenance, capex reserve) = $16,380 NOI.

$16,380 ÷ $220,000 = 7.4% NOI yield.

Closer to honest. This is the cap rate an institutional buyer would apply.

Yield method 3 — Levered cash-on-cash (with a HELOC or refi)

If the operator funds the ADU with a HELOC at 9% interest- only, the levered yield looks different:

  • Annual interest on $200k HELOC at 9%: $18,000
  • NOI: $16,380
  • Annual cash flow: $16,380 − $18,000 = −$1,620 (negative carry)

Levered with debt at typical rates, the ADU bleeds cash. The play is to assume property appreciation absorbs the negative carry and the cash-out refi at year 3–5 unlocks equity. That is a bet on appreciation, not on yield.

Yield method 4 — Exit-value-driven (the realistic underwrite)

Most ADU projects pencil on exit value, not on holding yield. The right question:

Does the ADU add more than its cost to the property's exit value?

Appraisers value ADUs at one of three approaches depending on local market:

  • Income approach. Add capitalized rent value. $16,380 NOI ÷ 6.5% cap rate = $252,000 added value.
  • Cost approach. Add depreciated reproduction cost. $220,000 minus a small depreciation factor.
  • Sales comparison approach. Add the average premium of comparable ADU-improved sales over non-ADU comps. In many submarkets this is $100,000–180,000.

On a $220,000 cost and a $252,000 income-approach value: 14% equity gain. On a $130,000 sales-comp premium: $90,000 equity loss (the cost exceeded the comp premium). The difference between markets is enormous.

The four conditions for ADU yield to actually work

  • 1. Strong rental demand. Submarket rent per sqft of comparable studio/1BR ADU rentals must be in the top third for the metro. A weak rental submarket leaves the ADU underutilized.
  • 2. Constrained ADU supply. If the submarket has hundreds of ADUs already, rent compression is happening. The first ADU on a block prices well; the 30th prices poorly.
  • 3. Cost path under $180/sqft. Above that, the math gets thin. Conversion ADUs (garage, basement) are often the right answer because cost path is lower.
  • 4. Buyer pool that values the ADU. Some submarkets price ADUs at full income-approach uplift. Others discount them as buyer-pool restrictions. Pull comp sales of ADU-improved properties before signing.

When ADUs do NOT pencil

  • Cost path over $200/sqft, modest rent submarket. California typical math: $220k spent, $1,650 rent, 4–6% yield. Capital better deployed elsewhere.
  • Lot with no ADU comp sales nearby. Exit appraiser will not credit the income; cost approach undervalues. Stuck capital.
  • Permit-restrictive jurisdiction. Timeline of 12–18 months kills the carry math.
  • Property already at high cost basis. If the primary property is already near comp-set ceiling, adding $220k of cost rarely returns at exit.

The kill list check

DealIntel's ADU strategy module computes all four yield methods and flags ADU projects where (a) all-in cost exceeds 95% of incremental exit value at the sales-comp approach, or (b) levered cash-on-cash is negative through year 5 with no appreciation assumed. Roughly one in three ADU pitches the platform sees triggers one of these flags.

Related reading

Keep reading

Written by
Matt Abadi
Founder, DealIntel

Matt Abadi is the founder of DealIntel. He leads the development of the platform's six-strategy underwriting engine, 25-point Kill List, and Monte-Carlo financial model — the institutional analysis stack DealIntel applies to every fix and flip deal. DealIntel was founded in 2025 with the central thesis that knowing when not to invest is the most valuable number on the page.

Last reviewed: 2026-05-10