This is the first question most owners ask, usually in a lowered voice. The direct answer: a properly performed engineering-based study is an established, IRS-acknowledged practice. The IRS publishes a detailed Audit Techniques Guide telling examiners how to review these studies, and agencies do not write examiner manuals for strategies they consider abusive.
Cost segregation audit risk is real, but it does not live where owners think. It lives in identifiable weaknesses, all of them fixable before you file.
The short answer
- The strategy is not the risk. Component-based depreciation has been validated in Tax Court and has its own official IRS examiner guide.
- The study can be the risk. A rule-of-thumb or software-generated allocation with no engineering basis is the most common weakness.
- Land allocation is the first thing an examiner tests, because it is the largest single number and easiest to challenge.
- The biggest exposure is usually not the study. It is an unsupported material participation claim behind a large loss taken against W-2 income.
- A defensible study is a documentation package, not a number. A report that cannot be reconstructed cannot be defended.
What Actually Draws Scrutiny
Six weaknesses account for most exposure. None are inherent to the strategy.
An unsupported land allocation. Land is never depreciable, so understating it inflates every downstream number. A 6% land allocation with no appraisal and no stated source is an invitation. See how land allocation should be supported.
A rule-of-thumb or software-generated allocation. The IRS guide names rule-of-thumb approaches as the least reliable of the six it describes. A report that cannot name its methodology has conceded the point.
A preparer with no construction expertise. The first element of a quality study is expertise in both construction and tax. "Proprietary algorithm" is not a credential.
No reconciliation to actual total cost. If the asset schedule does not add back exactly to the building's cost, the examiner has found a plug.
A large loss against W-2 income with no material participation documentation. This one generates real assessments, and it is a taxpayer issue, not an engineering one.
A §481(a) catch-up with no workpapers. A look-back reports one large number on Form 3115. If nobody can rebuild it from the original and corrected schedules, it will not hold up.
Worked Example: What an Aggressive Land Allocation Costs You
Example: a $2.2M residential rental placed in service January 2026 Two studies, same building. One allocates land at 22%, matching the county assessor's ratio and an appraisal. The other allocates 8%, with no stated source. Both reclassify 25% of depreciable basis into 5- and 15-year property at 100% bonus.
| Supported (22% land) | Aggressive (8% land) | |
|---|---|---|
| Land (non-depreciable) | $484,000 | $176,000 |
| Depreciable basis | $1,716,000 | $2,024,000 |
| Reclassified to 5- and 15-year | $429,000 | $506,000 |
| First-year structure depreciation | $44,850 | $52,900 |
| First-year deduction | $473,850 | $558,900 |
The aggressive version buys $85,050 of extra first-year deduction, roughly $30,000 in cash at 35%. It also puts the entire $308,000 land difference, and the credibility of the whole report, on the table in an examination. Poor trade. Your result depends on your tax rate and whether you can use passive losses.
Score Your Own Risk
| Factor | Low risk | Medium risk | High risk |
|---|---|---|---|
| Land allocation | Appraisal or assessor ratio | Contract allocation only | No stated source |
| Methodology | Detailed engineering | Documented sampling plan | Rule of thumb, or unnamed |
| Preparer | Named engineer | Tax preparer, engineering review | Software output only |
| Reconciliation | Ties to actual cost | Ties within rounding | Not presented |
| Loss usage | Passive loss vs. passive income | REPS with contemporaneous log | W-2 offset, no log |
| Audit support | Written commitment in the engagement letter | Informal assurance | Provider unreachable |
Two or more entries in the right-hand column is worth fixing before you file.
The Defense File
A defensible study leaves behind a package a stranger could rebuild the conclusion from: the engineering report naming the preparer, methodology, and inspection; a line-item asset schedule with class life, convention, and in-service date for every reclassified item; photo documentation and, where available, drawings and unit cost sources; the land allocation with its source attached; a reconciliation tying every allocated dollar to actual total cost; and, for a look-back, the §481(a) workpaper and Form 3115.
Then the part owners forget to negotiate: what your provider does if a notice arrives. Ask before you sign, in writing. The standard you want: the firm supports and explains its own report to the examining agent at no additional cost. A provider who will not commit to that in writing has told you something. See also choosing a provider and the IRS guide.
A Study Cannot Fix a Weak Participation Claim
Here is the part sales pitches skip. The most common bad outcome here is not a reclassification dispute. It is a taxpayer who deducted a $300,000 loss against a $500,000 W-2 salary and cannot substantiate the hours claimed.
Under §469, rental losses are passive by default. Using them against ordinary income generally requires real estate professional status, which means more than 750 hours in real property trades or businesses, more than half of all personal services performed there, and material participation in the rental activity itself, or a §469(c)(7)(A) election grouping all rentals into one activity. The alternative is the short-term rental exception, where average customer use is seven days or less and you materially participate.
A study is silent on all of that. The engineer measures the building; nobody there can testify to how you spent your Saturdays. The only real defense is a contemporaneous time log kept during the year, not reconstructed the following spring. If you are relying on the STR route, read how to keep a defensible time log before the year starts.
The engineering risk is manageable and mostly a function of who you hire. The participation risk is yours alone, and for most owners with W-2 income it is the larger of the two.
Frequently Asked Questions
Has the IRS ever disallowed a cost segregation study outright?
Disputes typically turn on specific asset classifications, land allocations, or methodology rather than on the concept. The practice was validated in Tax Court in Hospital Corporation of America v. Commissioner, 109 T.C. 21 (1997), and the IRS's own guide treats it as legitimate work performed well or poorly.
Does filing Form 3115 for a look-back increase audit risk?
Form 3115 is a routine compliance filing, and a properly filed automatic change carries audit protection for the method itself. What it does not protect is a weak allocation behind the number, which is why workpapers matter more than the form.
Should I use a cheap software study to save money?
Understand what you are buying. A study with no engineering basis, no named preparer, and no cost reconciliation is weakest on exactly the points the IRS guide emphasizes, and the examiner reads the report first. See engineering-based versus software studies.
See What Your Property Would Yield
Every building is different, and the only way to know your number is to look at your building. Precision Cost Segregation provides a no-cost feasibility analysis: send us the property address, purchase price, closing date, and any improvements, and we'll model your likely reclassification and first-year benefit before you commit to anything.
Request a free feasibility analysis →
This article is provided by Precision Cost Segregation for general educational purposes and does not constitute tax, legal, or accounting advice. Tax outcomes depend on your specific facts, your tax rate, your passive activity position, and your entity structure. Figures shown are illustrative. Consult your CPA or tax advisor before acting, and engage a qualified professional to perform any cost segregation study. Information is current as of publication.
This article is general information, not tax or legal advice. Depreciation outcomes depend on your facts, elections, and current law — consult your CPA before acting. © 2026 Precision Cost Segregation.