Have you hired Cost Segregation Guys for a cost segregation study?

Question

Hi All, I'd love to hear from you!

I have some properties that I'd like to have cost seg studies completed on, and would LOVE to hear what your experience has been, specifically with the company "Cost Segregation Guys."

The reclassification estimates are fairly aggressive in comparison to what Google/AI deems reasonable, but I'd really appreciate hearing your experience.

Some specific questions (but please feel free to share anything else! I'm all ears :)

1. Did the study produce more of the 'optimal' benefit or the 'moderate' that is quoted in the estimate?

2. Did the IRS flag the study at all for either audit or other? (If it triggered an audit- what was the outcome- did you win the audit once they provided a detailed version of the engineering study?)

THANK YOU SUBSTANTIALLY POCKETED CITIZENS!

Ana Garcia · Answer

Hi Kristin,

As a CPA, I always advise my clients to go with a quality, engineering-based cost segregation firm rather than relying solely on aggressive estimate algorithms or software-only shortcuts.

When you do a study, you want an audit-tested, fully documented engineering report that can actually back up the accelerated depreciation if the IRS ever comes knocking.

To address your second question: using an engineering-based study is a very safe strategy. The IRS generally accepts studies performed by qualified professionals following engineering principles, making audit triggers or adjustments far less likely compared to "DIY" or software-only estimates.

Definitely ask any firm you're considering for sample reports and whether they offer full audit defense included in their fee!

Sean Dougherty · Answer

@Kristin Perl These are exactly the right questions to ask before committing. On the estimate: any honest firm quotes a range, because the real numbers come from the engineering work, not the proposal. The tell is whether they can explain what would push your property toward the high vs. low end of that range, if they can't, that's a red flag.

On the IRS question: engineering-based studies with a site visit and full documentation are what hold up under scrutiny. The ones that run into trouble are desktop-only or software-generated reports that can't show their work.

Happy to run a free estimate so you have an independent number to compare against theirs: https://www.unitedtaxadvisors.com/sean-dougherty

Nathan Resnick · Answer

Ana, this is exactly the right framing, and I'll disclose up front that I'm a partner at Cost Segregation Guys, so I'm hardly neutral. But I wanted to reply because your two asks, sample reports and audit defense included in the fee are the sharpest filters in this whole conversation, and I don't think most investors realize how much signal they carry.

On sample reports specifically: what I'd tell Kristin to look for is whether the sample is a report or a summary. A real one runs long, assetby asset listing with cost basis assigned to each classification, takeoff detail, photo documentation, a methodology narrative explaining which ATG approach was used and why, and citations to the authority supporting each reclassification decision. If the sample is eight pages and ends in a summary table, that's a modeled output with a cover letter on it. Ask for a redacted sample from a property comparable to yours, too as a firm's best looking sample is often a large commercial asset that tells you nothing about how they handle a $700K residential.

Your point about the engineering-based approach reducing adjustment risk is worth underlining for one reason people miss: it's not that the IRS treats engineering-based studies as privileged, it's that the ATG lays out what a quality study contains, and a report built to that outline gives an examiner a clear path through it. Most adjustments I've seen didn't come from a disagreement about §1245 classification, they came from a report that couldn't show its work, so the examiner had nothing to test and defaulted to disallowing. Documentation isn't just defense, it's what keeps the conversation technical instead of adversarial.

The one thing I'd add to your list, since you're the CPA in the thread and clients rarely ask this unprompted: ask what happens on the 3115 side and who's responsible for it. For prior year acquisitions the study is only half the deliverable as someone has to produce the §481(a) computation and the 3115 support, and I've seen deals where the investor assumed the cost seg firm handled it and the CPA assumed it came with the report. Worth pinning down in the engagement letter which side owns it. We typically handle both for our clients alongside their CPA.

And on audit defense, agreed it should be included, but "included" is doing a lot of work in most people's marketing. The questions that actually matter: is it lifetime or a fixed term, does it survive a sale, does it cover the study only or extend to representation, and is it in the engagement letter or just on the website? We include lifetime audit support and I'd want any investor to hold us to the contract language rather than the phrase.

Appreciate you weighing in here as Kristin's getting better guidance from a CPA in-thread than from any of us on the vendor side.

Aaron Zimmerman · Answer

I'd recommend an engineered based study too. Happy to refer you towards a couple of my preferred vendors. They're timely and I feel like they're not overly aggressive. I have never worked with the cost segregation guys

Divin Kanyama · Answer

CPA here, @Kristin Perl. We’ve had clients use Cost Segregation Guys, and overall the experience has been positive, but we always review the assumptions carefully before relying on the estimate. In my experience, the final benefit often lands closer to the moderate side than the most aggressive estimate, depending on the property type, improvements, purchase allocation, and documentation available. I have not personally seen one of their studies automatically trigger an IRS audit, but I also don’t treat any cost seg as “audit-proof.” The key is making sure the study is detailed, engineering-based, well documented, and consistent with the return position. If the numbers look unusually aggressive, I’d ask them to walk your CPA through the methodology and asset classifications before filing. Done correctly, cost seg can be a great tool, but I’d focus less on the biggest projected deduction and more on whether the study is defensible.

Ashish Acharya · Answer

Kristin, I’d be less focused on whether the estimate looks “aggressive” compared with Google or AI and more focused on two things: how defensible the study is and whether you can actually use the losses it creates.

The IRS has a dedicated Cost Segregation Audit Techniques Guide for evaluating these studies, so methodology, asset classification, engineering support, documentation, and how well the report ties back to the property’s basis all matter.

Before hiring any firm, I’d ask for a sample full study, who performs the engineering analysis, how they support the 5-, 7-, and 15-year classifications, whether audit support is included, and how the report reconciles to your closing statement and improvements.

But I’d also do a detailed tax analysis before committing to the study. A large accelerated depreciation number can look great on paper, but if the rental losses are passive and you do not currently have enough passive income or another exception that makes them usable, much of that loss may simply carry forward. The IRS passive-activity and at-risk rules can limit the amount currently deductible.

So I would not judge the study by “optimal” versus “moderate” alone. The better question is: How much of this deduction is technically supportable, and how much of it can I actually use this year?

Happy to connect!

Mak J. · Answer

What is benefit of cost segregation? My understanding is accelerated depreciation. But it only is helpful if you have enough of the profits to write off each year? Is my understanding correct?

Nathan Resnick · Answer

Kristin, full disclosure before anything else: I'm a partner at Cost Segregation Guys, so I have an obvious interest here and I'm not a neutral voice in this thread. I'd rather answer your technical questions straight than let them sit unanswered, but weight actual client responses far more heavily than mine. I'll try to make this useful whether or not you ever work with us, because most of it applies to evaluating any provider.

Why the estimate probably looks aggressive next to what Google and ChatGPT tell you

The ranges you'll find online is usually "20–30% for residential, 20–40% for commercial" and they aren't wrong. They're just blended averages across every property type, vintage, price point, and market in the country, and they're often quoted against purchase price in one source and depreciable basis in another without anyone saying which. Comparing a property-specific estimate to that number is like comparing your actual appraisal to a Zillow national median.

Here's what actually moves the percentage on a specific deal:

Land allocation. This is the single biggest swing factor and the one most people never ask about. Land isn't depreciable. If you buy a $1M property in San Luis Obispo, land might be 35–45% of that. The same building in rural Texas might be 22%. Every downstream percentage is computed off what's left. A provider quoting 30% reclassification while assuming a 15% land allocation is quoting a very different dollar number than one assuming 40% and the difference is invisible unless you ask.

Property type and use. A furnished short-term rental with appliances, furniture, decorative lighting, and window treatments included in the purchase will blow past the generic ranges. I've seen furnished STRs land well above what any online average would predict. A restaurant, medical office, gas station, or hotel carries dense specialty electrical, plumbing, and process equipment that also lands high. Meanwhile a plain-vanilla suburban office building or a small unfurnished SFR often lands at or below the "average" range. The average is an average of these extremes; almost no individual property is actually average.

Site improvements. Parking lots, curbing, sidewalks, landscaping, site lighting, fencing, retaining walls, signage, and site utilities all go to 15-year land improvements. A property on 2 acres with a big lot has a fundamentally different profile than the same square footage on a small urban footprint. This category alone routinely accounts for 8–15% of basis and is a common reason a study "beats" the generic range legitimately.

What you actually bought. Basis isn't just the contract price as capitalized closing costs and certain acquisition costs get allocated across the same asset classes.

Bonus depreciation is a separate variable, and it's where a lot of "that seems too high" reactions come from. Reclassification percentage and Year 1 deduction are two different numbers. Under OBBBA, 100% bonus depreciation was restored for qualified property acquired after January 19, 2025 which means the entire 5, 7, and 15 year bucket can hit in Year 1 rather than being spread out. If your acquisition date falls under the older phase-down schedule (40% for 2025 placed-in-service under prior law, 60% for 2024, 80% for 2023), the Year 1 number is dramatically smaller for identical engineering. So confirm which bonus rate the estimate is applying and confirm it matches your actual acquisition dates. This is worth checking with any provider as it's the most common source of a Year 1 figure that looks inflated.

How I'd stress-test any estimate, ours included.

If you do nothing else, do these. A real study will survive all of them; a "modeled" study often won't.

• Ask for the reclassification broken into 5-year, 7-year, 15-year, and 27.5/39-year buckets with the asset categories driving each. If they can tell you "$X is site work, $Y is specialty electrical and dedicated circuits, $Z is finishes, cabinetry, and FF&E," there's engineering behind it. If it's a single blended percentage with no composition, be skeptical.
• Get the land allocation assumption in writing, and ask what it's sourced from. County assessor ratio? Appraisal? Comparable land sales? An appraisal-supported allocation is the strongest; assessor ratios are common and defensible but sometimes stale.
• Ask what methodology the report uses. The IRS Cost Segregation Audit Techniques Guide names several. Detailed engineering cost estimate (from actual construction records) and detailed engineering cost estimate approach (takeoffs and pricing where records don't exist) are the ones the ATG treats most favorably. This is what our firm does. "Rule of thumb" and pure modeling approaches are explicitly identified in the ATG as the weakest. Ask which bucket yours falls in.
• Ask whether the report is organized around the ATG's principal elements of a quality study. The ATG spells them out. A report built to that outline is materially easier to defend than a spreadsheet with a summary letter.
• Ask exactly what the audit support includes. Is it lifetime or a fixed term? Does it cover the study only, or does it extend to representation? Is it included or billed hourly? Does it survive if you sell? This varies enormously between providers and it's the thing nobody reads until they need it.
• If the property was placed in service in a prior year, ask about Form 3115. You don't amend prior returns, you file an automatic accounting method change and take the entire catch-up depreciation as a §481(a) adjustment in the current year. Confirm the provider produces the 3115 support and that your CPA is comfortable filing it. Some CPAs charge separately for this; know that in advance.

Your question 1: does the study land at "optimal" or "moderate"?

Honest answer on how those ranges are built: they're pre-study estimates derived from property type, purchase price, location, and comparable engineering data. Not a site-level takeoff. They're a forecast, not a commitment.

In our experience, final numbers land inside the quoted range in the large majority of studies. When one comes in at the low end or below, the cause is almost always one of three things:

• Land allocation came in higher than modeled once actual assessor or appraisal data was applied. This is by far the most common.
• The property has less site improvement than the model assumed like smaller lot, shared parking, no landscaping of consequence.
• Something we assumed conveyed didn't convey. Furnishings, equipment, or specialty systems that turned out to be tenant owned or excluded from the purchase.

Notice that all three are knowable before you engage. So: ask for the land allocation assumption in writing, tell the provider exactly what conveyed with the property, and describe the site accurately. That converts most of the variance risk into something you can check up front rather than discover at delivery.

I'd also push back gently on treating "optimal" as the goal. The right study is the accurate one. A study that reclassifies aggressively into 5-year property that doesn't belong there creates an exposure that outlives the deduction and it's your return, not the provider's, that carries it.

Your question 2: audit

A cost seg study is not itself an audit trigger. It's a permitted method for computing depreciation and the IRS publishes its own guide on how to examine one. Nothing about filing a return that includes a study puts you in a special queue.

What actually draws exam attention on these deals is usually not the engineering. It's whether you're entitled to use the loss and that question exists whether or not you ever do a study. A study just makes the number big enough to be worth arguing about. The three places it goes wrong:

• REPS (real estate professional status) under §469(c)(7). You need more than 750 hours in real property trades or businesses and more than half of your total personal services in those businesses and then material participation in the rental activity itself. The hours test is where cases are lost, and they're lost on documentation. Contemporaneous logs, not reconstructed calendars. The Tax Court has been consistently unimpressed by after the fact estimates. If you have a full-time W2 job, the "more than half" prong is generally fatal regardless of how many hours you log.
• The STR route. If average period of customer use is 7 days or less, the activity isn't a "rental activity" for §469 purposes under Reg. §1.469-1T(e)(3)(ii)(A) which means you don't need REPS. But you still need to materially participate under one of the seven tests, most commonly the 100 hour test where no one else participates more. Cleaners and co hosts participating more hours than you is the classic failure. Also: it's average stay across the year, and a few long bookings can push you over 7 days.
• §461(l) excess business loss limitation. Even when everything else works, there's a cap on how much business loss can offset non-business income in a year, with the excess carrying forward as an NOL. Worth modeling before you assume the full deduction lands in one year.

On the study side specifically: what protects you is a detailed engineering report, asset byasset listing, cost basis for each classification, photos, takeoffs, methodology narrative, and citations to the authority supporting each reclassification. Something an examiner can walk through line by line. We complete studies to that standard and include lifetime audit support, meaning if a study we produced is examined we defend the study at no additional cost for as long as you hold the property. I'd ask every provider on your list the same question and compare the actual scope, not the marketing phrase.

What I'd do in your position

Get your CPA in the loop before you engage anyone, not after the report arrives. Specifically, have them confirm: (a) you'll qualify to use the losses this year under REPS or STR material participation, (b) how §461(l) affects your particular income picture, and (c) whether they're comfortable filing the 3115 if any of these properties are prior year acquisitions. If the answer to a is "probably not this year," the study still isn't wasted as the losses suspend and release later or on disposition but you should make that decision deliberately rather than discover it in April. We are happy to chat with your CPA prior to doing a study.

Happy to answer technical questions in thread including about competitors' proposals, which I'll try to be fair about. And I'd genuinely encourage anyone in this thread who's used us to reply honestly. That's more useful to Kristin than anything I write.

Aaron Weikle · Answer

Used them on a duplex conversion and a small mixed-use building a few years back. My experience was solid, but let me give you the specifics you're actually asking about. The reclassification percentages landed closer to their "optimal" scenario on the residential property and right in the middle on the mixed-use. The difference came down to how much personal property and land improvements could be isolated. Things pertaining to specialized electrical, decorative lighting, and parking surfaces usually. If your properties have a lot of those components, aggressive estimates are defensible. If it's mostly vanilla stick-frame residential, tempering expectations makes sense. For audit risk, cost segregation itself doesn't trigger audits the way people fear. The IRS looks at the quality of the engineering study, not just that you did one. A study that documents every reclassified asset with actual measurements, photos, and cost allocation methodology is what protects you. Ask them directly whether the report follows AICPA and IRS Audit Technique Guide standards. A reputable firm won't hesitate on that. I haven't personally been audited on a cost seg return, but a CPA friend had a client go through an exam on a similar study from a different firm. The IRS pushed back on a few 5 year personal property items but the engineering backup held up so nothing was reversed. One practical thing: whatever vendor you use, make sure the study integrates cleanly into Form 4562 and your depreciation schedule before you file. Errors in how the reclassified assets get entered are more common than audit flags.

Kristin Perl · Answer

@Nathan Resnick Nathan, thank you so much for taking the time to provide this extremely thorough response! I absolutely understand better the range of estimates and I actually just signed the contract for the first property cost segregation study two days ago:) (Will do several more after the first is complete.) I totally understand the ranges now- when you're using averages across the entirety of the US, land values can look WAY different. We're definitely land-value-heavy here in SLO. It sounds like Cost Segregation Guys will be at BP Con this year! That's awesome! Thank you so much again for your reply! VERY useful and makes everything very straightforward. Truly, very much appreciated!!

Have you hired Cost Segregation Guys for a cost segregation study?

Have you hired Cost Segregation Guys for a cost segregation study?

Most Popular Reply

16 Replies