Cost Segregation in San Diego: The Look-Back Study Walkthrough

You bought the building eight years ago. A dental practice in Mission Valley, a warehouse in Kearny Mesa, a retail strip in La Mesa. Every year since, your tax return has shown the same quiet line: the building depreciating straight-line over 39 years, a modest deduction that never changes.

Here's the part that surprises most owners: that 39-year clock is a choice the tax code makes, not a law of physics. And for property you already own, you can change it. An engineering-based cost segregation study can reclassify parts of the building onto faster schedules, then catch up on the years you've already owned it. In San Diego, most commercial owners never hear about the second part.

What a cost segregation study actually does

A cost segregation study is an engineering-based analysis that breaks a building's cost into its component parts, so each part depreciates on the schedule that fits its actual life instead of the building's shell. The tax code does not require everything to depreciate over 39 years. The structure does. A lot of what's inside it doesn't.

• Structural components, the walls, the roof, the shell: 39 years for commercial property.
• Personal property, things like specialized electrical, lighting, plumbing, millwork, and interior finishes: 5 or 7 years.
• Land improvements, parking lots, fencing, landscaping: 15 years.

An engineer walks the property, or reconstructs costs from your closing statement and improvement records, and produces a report that puts every dollar in its proper class. It's not a percentage off a spreadsheet. It's an engineering document, which is the reason the IRS treats it seriously.

The look-back: catching up on property you already own

Here's the common assumption: cost segregation is only for the year you buy. Not true. If you've owned the building for years, you can still run the study and change how that property has been depreciated since day one. This is the look-back study, and it works through a change in accounting method.

You don't amend old returns. The catch-up for prior years lands as an adjustment in the current year, on top of the current year's depreciation. That's the point of the strategy: the deduction arrives now, when it can actually do something, instead of leaking out a little every year for four decades.

Most owners never hear about this because their preparer files the depreciation the software shows, year after year, and never asks whether the building's cost was classified correctly in the first place. Filing a return isn't the same as designing the outcome.

A walkthrough: a San Diego owner who bought in 2019

Let me walk through the mechanics in a representative situation, not any specific client's numbers. Say a business owner bought a commercial building for $1.4 million in 2019, with the land worth about $300,000. Excluding land, that leaves about $1.1 million depreciating straight-line over 39 years, roughly $28,000 a year.

A study doesn't change the total cost. It reclassifies it. A meaningful share of that $1.1 million sits in the shorter classes: interior finishes, specialized electrical, plumbing, and land improvements. How much depends on the building, which is why the engineering matters and why no one should quote you a fixed percentage sight unseen.

For tax years 2019 through 2025, the depreciation that should have been taken on those shorter classes gets caught up as an adjustment in the current year, on top of this year's depreciation. The result is a much larger deduction now, in a year you can use it, instead of the same modest amount every year.

Bonus depreciation is part of the picture, and it's worth being precise about it. In 2026, bonus depreciation is at 20%, down from 100% a few years ago. That changes the math for a new purchase, but it doesn't kill the value of a look-back study. The reclassification to 5 and 7-year classes still front-loads depreciation on the remaining cost, and the catch-up adjustment is the real prize. We laid out how the phase-down shifts the numbers in our post on bonus depreciation in 2026.

When a study pays for itself, and when it doesn't

Cost segregation isn't for every property, and a good strategist will tell you when it's not worth it. It pays for itself fastest on buildings with real components to reclassify: commercial buildings, retail, warehouses, medical and dental offices, rental properties, and short-term rentals. A small condo might not justify the cost of the study. A $1 million-plus commercial building usually does, often within the first year or two of the catch-up.

Two things decide whether the acceleration actually helps you: your income picture and the passive activity rules. Depreciation from a rental is generally passive, and passive losses are limited unless you qualify as a real estate professional. A study that produces losses you can't use is just moving numbers around. That's why the study should be paired with a look at your whole plan, not run in a vacuum.

For owners of short-term rentals, the calculus can be different. We walked through how a short-term rental's 39-year depreciation can be reclassified in this earlier post. And yes, accelerated depreciation is recaptured on sale. You don't make the tax disappear. You decide when it lands, which is the entire point of doing this on purpose instead of by default.

What the study looks like in practice

The engineer walks the property or works from plans and cost records, itemizes the components, and delivers a written report. You bring that report to a strategist, who applies it to your return and coordinates it with everything else you're doing: entity structure, retirement plans, and the timing of any sale.

You'll want your closing statement, any construction or improvement costs, and the plans if you have them. If the property has been renovated, those costs matter too.

Most firms file the depreciation their software shows. A strategist asks whether the classification is right in the first place. That difference is this whole post in one line: most firms handle the return, and proactive strategy handles the building.

If you own a commercial building, a rental property, or a short-term rental in San Diego and have never had an engineering-based study done, that's worth a 15-minute conversation. We'll tell you whether a study is likely to pay for itself before you spend a dollar on one. As tax strategists in San Diego, we work with property owners across California and nationwide. Book a free discovery call at (619) 280-2700 or info@RoadmapTax.com. The paid strategy session that follows is where you get the real deliverable: a plan built around your property, your income, and your goals.

FAQ

What is a look-back cost segregation study?

A look-back cost segregation study reclassifies parts of a building you already own into shorter depreciation schedules of 5, 7, or 15 years. The catch-up depreciation for prior tax years lands as an adjustment in the current year, without amending old returns.

Do I need to amend past tax returns for a look-back study?

No. The catch-up works through a change in accounting method, and the adjustment for prior years is taken in the current year. That is what makes the strategy useful: the deduction arrives now, when it can do something.

Which properties qualify for a cost segregation study?

Commercial buildings, retail and office space, warehouses, medical and dental offices, rental properties, and short-term rentals all qualify. Land does not depreciate, and a study only reclassifies the depreciable cost of the building and its improvements.

Does cost segregation still make sense in 2026?

Yes. Bonus depreciation is at 20% in 2026, but the 5 and 7-year reclassification still front-loads depreciation on the remaining cost, and a look-back study adds the catch-up adjustment. Whether it pays for itself depends on the building and your income situation.

Is accelerated depreciation recaptured when I sell?

Depreciation taken faster than straight-line is generally recaptured on sale. Cost segregation changes when the tax lands, not whether it lands, which is why it works best as part of a longer plan rather than a one-off.