What Is Xactimate? An Operator's Guide

Ask ten restoration owners what Xactimate is and you’ll get ten versions of the same shrug: “It’s the software the insurance company makes us use.” That’s true the way “a car is the thing that gets you to work” is true. It’s correct, it’s useless, and it’s the reason a lot of shops leave six figures a year on the table without ever knowing it happened.

Xactimate isn’t just an estimating tool. It’s the pricing language of the property insurance industry, and the carrier owns the dictionary. If you write estimates for insurance work and you don’t understand how that language is built, who controls it, and where it bends, you’re negotiating in a language you only half speak against people who wrote it. This is the operator’s guide I wish someone had handed me years ago. No Verisk marketing gloss, no aggregator fluff, real numbers where I have them.

What Xactimate actually is

Xactimate is estimating software owned by Verisk. It started life in 1986 under a company called Xactware, got acquired by Verisk in 2017 for around $1.3 billion, and is now the default estimating platform for property insurance claims in North America. When an adjuster, a TPA, or a carrier talks about “the estimate,” they almost always mean an Xactimate estimate, priced against an Xactimate price list, reviewed in an Xactimate-adjacent system.

The thing most owners miss: Verisk sits on both sides of the table. They sell the software you write your scope in, and they sell the analytics and review tooling the carrier uses to check your scope. The same company that defines what a “remove and reset toilet” is worth in your ZIP code also sells the carrier the tooling to flag your estimate if it deviates. That’s not a conspiracy, it’s a business model, and understanding it is the difference between an operator and a button-pusher.

There’s one current platform that matters: X1. The old desktop version, Xactimate 28 (“Classic”), reached end of life in December 2021. If you’re still running 28, you’re on borrowed time and stale pricing. X1 runs on Windows, syncs to the cloud, and has online and iOS companions. Android mobile is being sunset. That’s the lay of the land in 2026.

The line-item system: how scope becomes money

Everything in Xactimate is a line item. A line item is a coded unit of work with a unit, a quantity, and a price. “Detach & reset toilet,” “remove & replace 1/2″ drywall,” “clean stud wall.” Each one carries a code, and the codes are where most people’s understanding stops.

There are roughly 20,000 line items in the database. That number scares people into thinking the system is impossibly complex. It isn’t. In water mitigation, about 50 codes cover 80% of the work you’ll ever write. Learn those cold and you’re faster than most estimators who’ve been doing it for a decade and never bothered to internalize the pattern.

The codes look intimidating but they’re actually a flat string with modifiers baked in. Water extraction is WTREXT. Category 2 water? WTREXTG — the G is the modifier. Category 3? WTREXTS. The pattern repeats across the database: a base code plus a letter that bumps the category, the equipment grade, or the surface type. Once you see the grammar, you stop memorizing 20,000 items and start reading them like words.

One trap worth naming early: the three-letter category codes you see in the “Recap by Category” section of a PDF (DRY for drywall, WTR for water, CLN for cleaning) are not the same as line-item codes. The category is the bucket. The line item is the specific unit of work in that bucket. People conflate them constantly and it leads to sloppy supplements.

How pricing actually works (the part nobody explains)

This is the section the aggregator sites get wrong and the Verisk pages won’t spell out, so read it twice.

You do not set the price in Xactimate. The price list does. Every line item resolves to a unit price that comes from a price list tied to a specific geography and month. Verisk publishes general price lists monthly per region (vendor-specific pricing updates as often as nightly). The price for “remove & replace drywall” in Houston in June is a different number than the same line in Dallas, or in Houston in March. The software just looks it up.

That unit price is itself built from three components: material, labor, and equipment. Verisk derives those from ongoing market surveys, supplier data, and wage data in each region. You don’t see the seams unless you open the Pricing Editor, but they’re there. This matters because when an adjuster says “the price list already accounts for that,” they’re sometimes right and sometimes wrong, and you can only tell which if you understand what’s baked into the unit price versus what’s a separate line.

A few hard truths about pricing that change how you operate:

• There is no public pricing API. You cannot programmatically pull “what does this cost” from Verisk without the right license and partner enablement. Anyone selling you a magic real-time pricing integration is either licensed or lying.
• Pricing above the published price list triggers review. Go over the list without an overhead-and-profit justification or a documented reason, and you’ve handed the carrier a reason to kick the estimate. Don’t fight the price list head-on. Win on scope — the line items you include — not on inflating unit prices.
• Stale pricing quietly costs you. If your estimator is working off a price list that’s two months old, every estimate is slightly wrong, and over a year that’s real money in both directions.

As for what the software itself costs you: Xactimate is a per-user subscription that runs into the hundreds of dollars per user per month depending on tier and term. That’s a fixed cost most owners never revisit even as their seat count creeps up. Worth an audit once a year.

How to read an estimate without getting played

An Xactimate estimate has an anatomy, and once you know it you can scan any estimate — yours or the carrier’s — in about ninety seconds and know where the fight is.

1. The line items. The body of the estimate. Every unit of work, its quantity, its unit, its unit price, its total. This is where scope lives or dies.
2. RCV, depreciation, and ACV. Replacement Cost Value is what it costs new. Depreciation is what the carrier subtracts for age and wear. Actual Cash Value is RCV minus depreciation — the first check. The difference (recoverable depreciation) is usually released after the work is done and documented. A huge amount of underpayment hides in aggressive or non-recoverable depreciation that nobody contested.
3. Overhead & Profit (O&P). The convention is “10 and 10” — 10% overhead, 10% profit — applied when a job is complex enough to require general-contractor coordination. The rule of thumb adjusters use is the “three trades” test. Whether O&P applies is one of the most common, highest-dollar disputes in the business, and most shops just accept the carrier’s call on it.
4. Recap by Category. The summary at the back that buckets everything by trade. Great for a fast sanity check: if the cleaning category is suspiciously thin on a Cat 3 loss, you found your supplement.

You don’t win insurance estimates by pricing higher. You win them by scoping completely and defending every line with documentation. The price list is fixed. The scope is yours.

ROM vs. detailed estimates

Not every estimate needs to be a 300-line masterpiece. A Rough Order of Magnitude (ROM) is a fast, ballpark estimate — useful on the first call, for triage, for a reserve number. A detailed estimate is the line-by-line document you actually get paid against.

The mistake operators make is treating the ROM as if it commits them, or treating the detailed estimate as if speed doesn’t matter. The discipline is knowing which one the moment calls for. A ROM that accidentally becomes the number the adjuster anchors to has cost you before you wrote a single real line item. Give a range, label it a range, and make the detailed scope the document that does the talking.

Macros: speed that can get you flagged

Macros are saved bundles of line items you can drop into an estimate in one move — a “water extraction bedroom” macro, a “reset bathroom” macro. Used well, they’re the single biggest speed lever an estimator has. A shop with a tight, current macro library writes scopes in a fraction of the time and with far more consistency.

Used lazily, they’re a liability. The exact same macro applied to every room, with identical quantities, reads to a review system like templating — because it is. If every bedroom in the house has precisely the same square footage and the same line items, the carrier’s review queue notices. The rule: macros should accelerate the structure of your scope, but the per-room quantities have to reflect the actual loss. Speed is fine. Sloppiness is what gets kicked.

The ESX file: the part that’s actually a moat

Here’s the piece almost no operator-level guide mentions, and it’s the most strategically important thing in this whole article.

An ESX file (Estimate Exchange) is the portable form of an Xactimate estimate. Under the hood it’s just a ZIP archive containing an XML document plus your photos, sketch, and assets. It’s how estimates move between parties, and critically, it’s an import format. If you can produce a valid ESX, Xactimate will ingest it — line items, sketch, measurements, photos, contacts — without you ever clicking through the UI to build it by hand.

That’s the seam where the future of estimating gets pried open. The slow, expensive part of estimating isn’t the pricing — the software does that on import. It’s the human translating a flooded house into 80 correct line items with the right modifiers and quantities. Anything that can generate a clean ESX from the field data side-steps the majority of that labor. This is exactly the lane Restoration 360 builds in, and I’ll be honest about it rather than coy: the ESX-first path is why an independent automation play is viable at all.

Where AI is actually changing estimating (and where it isn’t)

2026 is the year the hype caught up to the trade, so let’s separate signal from noise.

Verisk launched XactAI in September 2025 — carrier-side first. In May 2026, Verisk announced MCP connectors that wire its analytics directly into Anthropic’s Claude, which I broke down in detail in the Verisk × Claude piece. And in December 2025, the FTC blocked Verisk’s $2.35B acquisition of AccuLynx, which means Verisk can no longer simply buy the contractor-side software that threatens its position. For the first time in years, the antitrust weather actually favors independent tools.

What AI does well today: detecting supplement candidates from photos, drafting first-pass scope from field capture, reviewing a scope against IICRC S500, and triaging carrier communications. What it does not do, and won’t for a while, is replace the operator’s judgment about which fight is worth having on which loss. I wrote about that balance more broadly in AI in Restoration. The honest framing: AI is absorbing the volume of low-judgment estimating work, not the judgment itself. The estimator who learns to drive it writes three times the scope at the same quality. The one who ignores it competes on hours against people who don’t have to.

What most guides won’t tell you

Xactimate is not neutral infrastructure. It’s a privately owned pricing standard, controlled by a company that also serves the carriers reviewing your work. That’s not a reason for paranoia, it’s a reason for fluency. The operators who treat Xactimate as “the annoying software” stay button-pushers and accept whatever the price list and the adjuster hand them. The ones who treat it as a language they intend to be fluent in — the codes, the pricing mechanics, the ESX seam, the O&P fight — consistently collect more on the same losses, with less rework, and sleep better doing it.

What to do Monday

Pull your last ten closed estimates. For each one, check three things: Was the price list current for the month the work happened? Was O&P applied or contested where the three-trades test was met? Did the depreciation get reviewed, or did you accept the carrier’s number without a second look? If you can’t answer all three on most of those jobs, you don’t have an Xactimate problem — you have a fluency problem wearing an Xactimate costume. Start there. The price list is fixed, but almost everything else on this page is yours to win.