🇺🇸 Made in America·100% U.S.-Based Team · 24+ Years in Cost Segregation
The IRS publishes the Cost Segregation Audit Techniques Guide, Publication 5653, to tell its own examiners what a quality study looks like and what to challenge when one falls short. It lists 13 elements: a qualified preparer, a described methodology, source documents, interviews, common nomenclature, a numbering system, legal analysis, take-offs and unit costs, organized asset lists, reconciliation to actual cost, treatment of indirect costs, identification of Section 1245 property, and related issues such as method changes and sampling. The guide is not law and cannot be cited as authority, and it does not pre-approve anyone's study. What it does is show you exactly how your study will be read if your return is examined. If the study can answer all 13, you are holding a defensible position. If it cannot, the deduction is a number waiting for a letter.
Diane and her husband bought a small medical office building on College Avenue in Fayetteville, Arkansas, in March 2025 for $2,400,000. Diane is a physical therapist who runs her practice from the first floor; the second floor is leased to two other practices. Their CPA suggested a cost segregation study, and two proposals arrived the same week. One was a low-priced report produced from a questionnaire and a purchase price, turned around in days, with nobody looking at the building. The other was a fully engineered and accounted study with a document review, a live video walkthrough of every room, and a written methodology. The first cost a fraction of the second.
Her husband's first instinct was the cheap one. The projected deduction looked the same on paper. Then their CPA asked the better question: "If the IRS opens this return in three years, which of these do you want to hand them?"
Diane and her building are illustrative; the guide and its 13 elements are quoted from the IRS.
Publication 5653 is an internal training and reference document from the IRS Large Business and International division that the IRS makes public. It explains why studies are done, how they are prepared, what to look for in one and when an issue deserves a closer look. It carries a disclaimer that it has no status as an official pronouncement and may not be cited as authority, so a court is unbound by it and so are you. It is also the most candid published description of how an examiner approaches a study, which makes it the best rubric you will find for judging any provider, including this one.
Residential rental buildings depreciate over 27.5 years and nonresidential buildings over 39. Those periods apply to the structure and its structural components. Property that is tangible personal property under Section 1245, or a land improvement, may be examined for 5-, 7- or 15-year treatment. We explain the line in our guide to Section 1245 versus Section 1250 property.
Two cases anchor the analysis and the guide discusses both. Hospital Corporation of America v. Commissioner, 109 T.C. 21 (1997), held that the personal property tests developed under the old investment tax credit apply to MACRS classification. Whiteco Industries v. Commissioner, 65 T.C. 664 (1975), gave the six factors for deciding whether an item is inherently permanent, which is the center of every classification argument. Our article on the Whiteco six-factor test shows how they work on real components. Revenue Procedure 87-56 sets the class lives. And since the One Big Beautiful Bill Act, property with a recovery period of 20 years or less that is acquired and placed in service after January 19, 2025 gets 100 percent bonus depreciation, which is what turns a reclassification into a first-year deduction.
The law permits the classification. It does not tell you how to document it. The guide fills that gap.
Illustrative. Land comes out first; the March mid-month rate on 39-year property is 2.033 percent.
| Line | No study | Engineered study |
|---|---|---|
| Purchase price | $2,400,000 | $2,400,000 |
| Land (appraisal and county ratio) | $400,000 | $400,000 |
| Depreciable basis | $2,000,000 | $2,000,000 |
| Reclassified to 5-, 7- and 15-year (24%) | $0 | $480,000 |
| Bonus depreciation, year 1 (100%) | $0 | $480,000 |
| Building on 39 years | $2,000,000 | $1,520,000 |
| Building depreciation, year 1 (2.033%) | $40,660 | $30,902 |
| Total first-year deduction | $40,660 | $510,902 |
| Additional deduction | $470,242 | |
| Federal tax at 37% | $173,990 |
Diane and her husband file jointly, so the Section 461(l) excess business loss cap of $626,000 for 2025, adjusted annually, sits above any loss the building produces. Whether they can use the deduction in year one depends on the passive activity rules, covered below. What the table does not show is the other side. A study is a supporting schedule for a tax position. If the position is challenged and the study cannot be defended, the $173,990 flips from an advantage to an exposure, with interest and a possible 20 percent accuracy-related penalty on a substantial understatement. How big the number is matters less than how well the document behind it holds up, and the 13 elements are the IRS's answer to that second question.
The guide says there are no prescribed qualifications for cost segregation preparers; anyone can write one. It then says a study prepared by a construction engineer is generally more reliable, that cost estimating experience and tax knowledge both matter, and that a quality study names the preparer and states the credentials. Ask who did the work; the name on the cover is a separate question. Our list of questions every CPA should ask a provider covers how to test this. Diane's cheap report listed no preparer at all.
The report has to say which approach it used, why, and how it was applied. The guide describes six common approaches and prescribes none; it calls Approach 1, detailed engineering from actual cost records, the most methodical and accurate, and Approach 2, engineering with documented estimates reconciled to actual cost, methodical and reasonably accurate. "Industry standard methods were used" is no description at all. A sequence from document review through walkthrough, take-offs and reconciliation is.
Contemporaneous records: plans, specifications, change orders, pay applications and the cost ledger on new construction; the purchase agreement, closing statement, appraisal, leases and tenant improvement records on an acquisition. Better documents mean less estimating. Owners who built the property have a natural edge, and gathering the documents before the study starts saves time either way.
The owner, the property manager and, where possible, the contractor and subcontractors. Interviews catch what drawings miss: which finishes went in, which rooms serve a specific business use, what a prior owner replaced. The guide expects the interviews to be documented.
Asset descriptions in standard construction terms that match the project documents. "Misc. equipment" tells an examiner nothing. "Dedicated 30-amp circuit, second floor suite B, serving sterilizer" tells them everything.
Assets numbered consistently, ideally tied to bid documents and pay requests, so any line in the study traces back to a source. That trace is what makes a study auditable.
The statutes, regulations and cases behind each Section 1245 classification, with conflicting authority addressed. A spreadsheet of percentages is not a legal analysis. A decorative fixture in a lobby and the building's general electrical service are classified by function and relationship to the building, and the study has to say why.
The heart of the engineering approach. A take-off measures quantities: square feet of flooring, linear feet of millwork, counts of fixtures, lengths of conduit. Unit costs from contractor data or a cited published source turn quantities into dollars. The guide expects the quantities to be identifiable on the drawings or other support.
Results organized by category and recovery period, tied to the owner's fixed asset records, so the CPA can apply them and the examiner can see how they flowed through.
The components must add up to what you paid or built, no more and no less, with the same estimating technique applied across every item. This is where weak studies most often fail. When estimates from a cost guide total more than the purchase price, the lazy fix is to scale everything by one factor. The guide specifically warns against that pro-rata step-up, because it spreads the error evenly and hides it. The land allocation sits here too; our article on land value allocation shows how quietly that number moves a result.
Architectural and engineering fees, permits, general conditions, builder's risk insurance, construction-period interest and developer fees are allocated to the assets they relate to on a stated, consistent basis. An unexplained soft-cost allocation is a red flag.
A list of the Section 1245 property with amounts, including anything that would ordinarily be Section 1250 but is treated as 1245 because of its function, and an accounting for separately purchased personal property so nothing is counted twice.
Section 263A capitalization on property you construct, changes in accounting method, and sampling, including how a sample was defined and how sampling error was handled. On a building you have owned for years, the method change is made on Form 3115 with a Section 481(a) adjustment; our article on look-back studies explains the mechanics.
The residual method (Approach 4) values only the short-lived assets and assigns whatever is left to the building. The guide says it generally does not reconcile project costs and can skew the result toward Section 1245 property, because nothing checks the allocation.
Pro-rata step-up factors, covered under element 10. If a report says its estimates were adjusted to match your purchase price without explaining how, ask for the detail.
Sampling and modeling (Approach 5) can fit large, repetitive portfolios, but the guide warns it may not be statistically valid. For a single building, a full take-off is more defensible.
Rule of thumb (Approach 6) is a percentage applied to a purchase price with little or no documentation. The guide says to view it with caution. That is the estimated study, whatever its invoice says.
The guide expects a quality study to include an inspection of the building, to confirm what was installed, understand the design and photograph the property. On an acquired building that matters most, because the original plans rarely match what is there now. Cost Seg America does it one of three ways: on larger properties an in-house analyst does a proper inspection; on others the team runs a live video walkthrough with the owner or manager; or the owner follows a guided photo and video checklist room by room and the team reviews every frame. Nobody looked at Diane's building for the cheap report. The walkthrough for the engineered study found that the prior owner had renovated the second floor with dedicated circuits and finishes that no drawing showed.
Put the 13 elements together and you have the description of a fully engineered and accounted cost segregation study: every component priced under Approach 1 or Approach 2, a component schedule your CPA can import and the engineering support behind it, which an examiner can trace line by line, reconciled to the dollar, signed by credentialed professionals and engineered in the United States. An estimate produces a percentage and a cover letter. Cost Seg America defends every study it signs for as long as you own the building, with no time limit, hour cap or additional fee, through written responses and participation in examiner calls alongside your CPA, and its studies have been through 125+ IRS examinations with zero losses and $0 returned.
It does not say cost segregation is risky. Its premise is that a study done well produces a supportable result. It does not say that following it guarantees a clean examination; there is no safe harbor, the IRS does not pre-approve studies, and an examiner can disagree with a classification in a well-prepared report. The right standard is whether you can support every line when asked. Our article on common audit triggers covers that. And it says nothing about whether you can use the deduction on your own return.
The study answers what the building contains. Section 469 answers who can use the deduction. A real estate professional with more than 750 hours and more than half of working time in real property businesses they materially participate in treats rental losses as nonpassive. A short-term rental owner whose average stay is seven days or less and who materially participates is on a different, easier path. Diane's situation is a third one: she rents the first floor to her own practice, and the self-rental rule makes net income from that space nonpassive while a net loss stays passive, unless the practice and the building are treated as one activity or grouped under Reg. 1.469-4(d)(1)(ii), which requires the same owners in the same proportions on both sides. Her CPA ran that analysis before the study was ordered. If no path applies, the loss is suspended, and it comes loose on a taxable sale of the entire interest. Section 461(l) can cap a large nonpassive loss in a single year.
Depreciation on reclassified Section 1245 property is recaptured as ordinary income at sale; the 25 percent cap on unrecaptured Section 1250 gain covers only the building's straight-line depreciation. The study changes when you take the depreciation, never how much of it there is over the life of the building. Our article on depreciation recapture after a cost segregation study works the numbers.
If the building has less than $180,000 of depreciable basis after land, no study, however well built, pays for itself. If you plan to sell within a couple of years without exchanging, recapture outruns the time value. If the loss would be passive and nothing in your return can absorb it, the deduction waits regardless of how good the document is. And if a provider cannot answer the ten questions below in writing, the study you are being offered is not the one the guide describes, whatever the price.
A provider that welcomes these is behaving the way the guide describes a quality preparer. One that deflects them is telling you something. The cheap end of the market is the expensive end.
Diane chose the engineered study. It included a stated methodology, the walkthrough that found the second-floor work, take-offs for the therapy build-out, a land allocation tied to the county assessment and the appraisal, a reconciliation that matched the $2,000,000 basis to the dollar, and a schedule her CPA imported in an afternoon. If the return is ever selected for examination, the examiner will ask for the study, and she has a document that answers the 13 questions before they are asked. That is what the document is for. No provider can promise how any examination ends; a provider can only make sure the examiner has something to read.
The Cost Segregation Audit Techniques Guide, a public IRS document that trains examiners to review studies. It describes the legal background, the common approaches, the 13 elements of a quality study and review techniques. It carries no authority and cannot be cited as such.
No. The guide says there are no prescribed qualifications. It also says a study by a construction engineer is generally more reliable and that a quality study names the preparer and credentials.
There is no public IRS statement that a study by itself selects a return. What the guide shows is how closely the study will be read if the return is examined, which is the argument for a study built to its 13 elements.
Yes, while you still own it. A look-back study with Form 3115 and a Section 481(a) adjustment claims the missed depreciation in the current year without amending prior returns, at the bonus rate of the original placed-in-service year. Cost Seg America prepares the computation and, for a separate fee, the form; your CPA files it.
For property acquired and placed in service after January 19, 2025, the 5-, 7- and 15-year property a study identifies is bonus-eligible at 100 percent. Binding contract dates matter, so confirm the acquisition date with your CPA.
Real estate professionals and short-term rental owners who materially participate can apply it against other income. Everyone else carries a suspended passive loss forward until passive income or a taxable sale releases it.
Send the property address, the purchase or construction cost and the placed-in-service date. Jim Dougherty and his team will send a free proposal within 24 hours showing the projected first-year deduction and will say plainly if the study does not make sense on your building. Call 1-888-365-5023 or email info@costsegamerica.com.
This article is educational and is not tax, legal or accounting advice; Publication 5653 is an examiner reference and not authority, and you should review your situation with your CPA before acting.
Use the calculator, see your number, and request your free, no-cost proposal - delivered in 24 hours, with your flat fee quoted upfront and no obligation.