DIY Cost Segregation: What a Defensible Study Needs

Cost segregation splits the price of a building into its parts so that some of them depreciate over 5 or 15 years instead of 27.5 or 39. If you want the full strategy, start with our cost segregation guide. This page covers one narrower question: can you do the study yourself?

The best source is the IRS Cost Segregation Audit Technique Guide , Publication 5653, revised February 2025. It is written for examiners, so it tells you how your study would be read. It also says it is not an official pronouncement of the law, so treat it as a road map, not a safe harbor.

Four statements in the guide frame the whole DIY decision:

• No required standard. The IRS has not set requirements or standards for preparing studies, and it does not prescribe a method.
• No required credential. There are no prescribed qualifications for preparers, though credentials may have a bearing on accuracy and quality.
• Engineers carry more weight. A construction engineer's study is generally more reliable than one by someone with no engineering or construction background.
• You still carry the proof. Taxpayers must substantiate depreciation deductions and property classifications, and actual costs are more accurate than estimates.

The guide adds a point most DIY articles skip. It says studies on used real property should be performed by qualified appraisers, and that the purchase price should be allocated among land, building, and personal property by value on the purchase date, citing AmeriSouth XXXII, Ltd. v. Commissioner, T.C. Memo. 2012-67. If you bought an existing rental, which is the usual DIY case, expect that sentence to be read back to you.

The guide also rates the common approaches. A detailed engineering study from actual cost records is, in general, the most methodical and accurate, but it mostly fits new construction. A residual approach, which prices only the short-lived assets and assigns the rest to the building, is simpler but can be less accurate. A rule of thumb, such as an assumed percentage from industry averages, is the one examiners are told to view with caution because it lacks documentation.

For an acquired property, the land comes first. The guide says land value is always determined first, based on highest and best use, and the balance is then allocated to the building and other assets. A quality study on an older building also accounts for physical deterioration and functional obsolescence. A new carpet and a 20-year-old carpet do not carry the same value.

Publication 527 gives a simpler allocation rule for ordinary rentals: divide the cost by fair market value, and if you are unsure of the values, you can use the assessed values for real estate tax purposes. See Publication 527 . That helps with land versus building. It does not break the building into components.

For the spreadsheet behind the report, we suggest these columns: unit number, plain-English description, location on the drawing or photo number, quantity, unit cost, cost source, direct cost, allocated indirect cost, total cost, section 1245 or 1250, asset class, recovery period, placed-in-service date, and the legal authority for the classification. The column list is our suggestion. The audit guide sets no format.

Add two checks at the bottom. The first is that total allocated cost equals the purchase price or project cost. The second is that any separately purchased furniture, fixtures, or equipment is not also counted in the allocation. The guide flags both.

The allocation adds $193,636 of first-year depreciation. At a hypothetical 35% federal rate, that defers about $67,773 of federal income tax in year one. Later annual depreciation drops by $7,273, because $200,000 of basis is already gone. This is an illustration of the math, not a projection of what any owner will get.

The rules behind the numbers:

• 100% bonus is permanent for qualified property acquired and placed in service after January 19, 2025. See Notice 2026-11 and the IRS release .
• Only property of 20 years or less qualifies, under section 168(k) . The 5-year and 15-year pieces do. The 27.5-year building does not.
• Contract date matters. Property is not treated as acquired after the date a written binding contract is entered into. A contract signed on or before January 19, 2025 stays under the old phase-down, which was 40% for 2025 and falls 20 points a year.
• The 3.182% rate is the February first-year rate for 27.5-year property in Publication 527, Table 2-2d . It reflects the mid-month convention. Other months differ.

What the example leaves out, and it matters:

• Passive loss limits. Most rental activities are passive, so the loss may only carry forward. Publication 527 lists a $25,000 allowance for many owners, phasing out from $100,000 to $150,000 of modified AGI. If you cannot use the deduction now, the study buys you little this year. See the short-term rental loophole and real estate professional status for how some owners change that.
• Recapture. This is largely timing. Gain on section 1245 property is ordinary income to the extent of depreciation (Publication 544 ), and unrecaptured section 1250 gain has a maximum 25% rate (Topic 409 ).
• State tax. Some states do not follow federal bonus depreciation. California, for one, does not conform to section 168(k) .
• Property type. The 27.5-year life applies to residential rental property. Publication 527 excludes a unit in a hotel, motel, inn, or other establishment where more than half of the units are used on a transient basis. Confirm this before you copy the math onto a short-term rental.

For your own rough numbers, try our cost segregation estimator. It is a self-service screen, not a study. For break-even math at smaller price points, see our cost segregation for rental property guide.

Software sits in the middle. The guide does not grade tools by name, so judge the output. Ask four questions. Does it list items or just percentages? Does it use your photos, records, and quantities? Does it reconcile to your price? Does it name a preparer who can explain it? Tools that apply set percentages by property type look a lot like the rule of thumb approach.

A weak study has a cost. If assets move back to longer lives, you owe the tax and interest, and a 20% accuracy-related penalty can apply to the underpayment from negligence or a substantial understatement. In fairness, the guide tells examiners not to reject a rule of thumb study categorically and to verify its documentation on the merits. That helps only if the documentation exists.

Filing is its own job. For property already depreciated on your returns, the guide states the IRS position that using a study to reclassify it requires consent as a change in accounting method, which is requested on Form 3115 . Read our Form 3115 catch-up guide if the property is already on your books, and our year-end timing guide if you closed this year.