The honest answer to “how much does cost segregation cost” has three parts: what the market charges, what you should expect to pay for your specific property, and the one question that matters more than price — will it survive examination?

The 2026 market, at a glance

Tier Typical 2026 pricing What it actually is
Promotional questionnaire $299–$699 (loss-leader) Table-driven estimate, minimal records analysis
Standard questionnaire study ~$1,000–$2,500 Structured homeowner inputs + published tables
Full engineered study ~$3,000–$15,000+ Asset-by-asset engineering from cost records; the standard the IRS ATG rewards
Rush / 24-hour engineered premium on engineered work Same rigor, compressed; rarely offered honestly

Ranges here are our observed-market framing from reviewing category players — every firm quotes by property; get the number in writing.

What actually drives your price

  1. Document quality. A builder’s accounting or itemized appraisal is worth hours of manual reconstruction skipped. The better your cost detail, the cheaper the study — and the bigger the reclassification, because the study measures instead of estimates.
  2. Property type & size. A single-family rental is not a 200-unit apartment complex with three HVAC plants and a pool. Commercial systems, multiple entities, and portfolio batching all add engineering hours.
  3. Inspection needs. Virtual walkthroughs handle most residential and small commercial work at standard price. Genuinely complex commercial assets need boots on the ground — pay for it where the engineering demands it.
  4. Speed. 24-hour turnaround requires a standing pipeline with slack in it. Ours (SegWise Quick™, from $1,499) is AI-accelerated classification with human sign-off, which is how we price the rush this low.

Our published fees (flat, quoted before work starts)

From $855 (SegWise Rapid, questionnaire, small rentals) · $1,499 (SegWise Quick™, 24-hour engineered) · $2,088 (Full Engineered residential; $2,457+ commercial) · $450 (pre-acquisition analysis, credited at closing) — published starting points that sit about 10% under the only national firm that publishes its rates. Form 3115 catch-up and audit support are included in every tier — no surprise line items.

The $299 study trap (and the 2026 version of it)

Low-cost studies are subsidized marketing: the fee is engineered to be a yes, the deliverable is a questionnaire on published tables. Two ways that costs real money in 2026:

  • The audit question. Pub 5653 (Feb 2025) tells examiners to reward “detailed engineering approach from actual cost records” — and to discount guesses. A table-based study answers “did your expert measure it?” with “he assumed it.”
  • Stale-law math. Some shops still quote year-one savings on pre-OBBBA phase-down tables, or misstate the January 19, 2025 test (it’s a fixed statutory date — the binding-contract date controls, not your closing date and not “the Senate passage date,” a phrase floating around the internet that is simply wrong). Ask any firm you’re quoting: “what bonus percentage do you assume for a contract signed February 2026?” If they hesitate, you have your answer.

When a study is not worth it (we say this out loud)

A study needs a runway: enough depreciable basis (very roughly $250K+) and enough holding years to enjoy the deferral. Skip-or-wait cases: a house about to sell with little basis left, a fast-flip BRRRR where placed-in-service timing fights the study, or a tiny rental where the deduction delta barely clears the fee. Any reputable firm should model this before taking your money — ours put it in the written proposal, in writing, and we’ve talked owners out of studies at tax time.

The price question is really a return question. At our engineered-tier fees, a typical $500K residential purchase producing $30–40K of year-one tax savings returns 10–20× the fee in year one alone. Run your property through the calculator, then let the written proposal make the number concrete.