Real Estate Investor · Austin, TX · Member since 2017 · 142 posts · 35 votes
I keep running into the same situation with older rentals: bought years ago, no original invoices, no blueprints, seller long gone. The owner wants a cost segregation study, but there's almost no paper trail — just the closing statement and county records.
Can a study still be done properly on a property like that, and what do the engineers actually use instead of the plans? Site inspection and measurements, estimating guides, photos? And does the missing documentation change anything — the price, which components they can reclassify, or how defensible the study is if it's ever questioned?
Asking because I see a lot of owners assume no paperwork means no study, and I'm not sure that's right. Curious what CPAs and engineers here have actually seen work (or not work) on older properties.
Yes, a study can still be done, but it will look different from one on a new build with full records. Engineers do this regularly on older properties. They start with a site inspection, photos, measurements, and the county records, then estimate component costs from construction cost data for the year the property was built or renovated. Your closing statement helps establish the starting basis.
What missing documents do affect is defensibility and precision. Without invoices, more of the allocation rests on the engineer's estimates and judgment, which is fine as long as the firm documents its methods. A reputable firm should tell you upfront how much of the study will be estimated versus documented, and that is worth asking before you pay.
Two things I would do: get a free feasibility estimate first, most cost seg firms will give you a rough number before you commit, and run it past your CPA, because the study only helps if your tax situation can actually use the accelerated depreciation this year. If you cannot use the losses now, they may just sit as suspended passive losses.
Yes, and it’s common on older rentals. For a purchased building, the depreciable basis is what you paid (less land), not what it cost to build, so original invoices aren’t needed. The engineer works from a site visit: measurements, a component inventory, and photos. Then they price everything with standard cost-estimating guides and allocate your purchase price across the components. The IRS Cost Segregation Audit Techniques Guide accepts estimation-based approaches as long as the methodology is documented and ties back to your basis.
Where I see these studies go wrong on older properties:
Land allocation. The county assessor’s ratio is a starting point, not gospel. A bad land split moves the numbers more than the study does.
Improvements since purchase. Receipts for roofs, kitchens or HVAC do matter. Replaced components may also qualify for a partial disposition write-off.
Catch-up. You don’t amend old returns. You file Form 3115 (automatic change in accounting method) and take the missed depreciation as a §481(a) adjustment in the current year.
Bonus. The bonus percentage follows the acquisition and placed-in-service dates: none for used property bought before late 2017, phased-down rates for 2023 through early 2025, and 100% for property acquired after January 19, 2025.
Missing paperwork affects how the report supports its estimates, not whether you can do one. Whether it’s worth the fee depends on when it was bought and for roughly how much.
Yes, and you're right to question that assumption. Missing paperwork changes which method the study uses, not whether one can be done. Rosey's right that the IRS audit guide is the benchmark, and what it adds is a ranking of methods. A detailed engineering estimate built from a site visit, which is where an older building with no records lands, sits second only to a study built from actual cost records. Residual, sampling and rule-of-thumb studies sit at the bottom. The cheap desktop studies usually fall into that group because nobody walked the building and counted what's in it, and under audit a component list nobody saw is hard to defend. A study is only as strong as how much of the building someone actually looked at. Do the quotes the owner is getting include a site visit?