Two connected posts break down State Farm's internal Operation Guides for the 2025 LA wildfire smoke/soot/ash claims and argue policyholders should stop litigating single files and instead demand the whole claim system in discovery, the Operation Guides by revision number, industrial-hygienist assignment letters and raw sampling data, and native (not PDF) Xactimate/XactContents files showing every edit, override, and price-list selection. The posts stress these are allegations under investigation, and that sampling only the "easiest, cleanest, most accessible" spots doesn't answer whether the whole home was contaminated.
Why it matters
The scope fight on smoke and soot is decided upstream by the hygienist's sampling plan and the adjuster's dashboard prompts, before you ever get consulted. And your own file is now discoverable: if your remediation protocol reads "wipe down walls" with no technical basis, it won't survive the same scrutiny being aimed at the carrier.
The read
The obvious fight on a smoke claim is coverage. The leverage sits in the metadata. A native estimate file carries its own version history, which shows every edit, override and price list choice that turned a scope into a smaller number. Keep your own files native and unflattened, save the sampling data behind each cleaning or replace call, and you become the expert the homeowner needs.
Two reads on the same paradox. Claims Journal reports the 2023 tort reforms (killed one-way attorney fees, tightened bad-faith standards, moved Citizens disputes to arbitration) coincided with claims-closed-without-payment jumping ~5 points to 40.3% at private insurers, the highest in the country during Helene and Milton, while average premiums rose 14.3% to $8,292 (national average is $2,948). Merlin adds the balance-sheet side: Florida domestics booked ~$1B in 2025 underwriting gains, but a state-commissioned study of 53 insurers found 41 used affiliated MGAs pulling 20-34% of premium (up to 63% in fees total), and HB 1399, which would force insurers to prove affiliate payments were fair, passed the House 106-3 and died in a Senate committee.
Why it matters
This is the carrier-side scoreboard your adjusters are working from. "The market is healthy" is now the official line, which means less appetite to negotiate scope, water caps stuck at $10,000, and more files pushed toward arbitration where you have less leverage.
The read
Reforms worked and the market stabilized is the official line, and the balance sheets back it. The profit did not come from fewer lawsuits. It came from money leaving through affiliate management fees, and from smaller settlement checks on the claim side, with two in five claims closing without payment. Price Florida work for that reality: arbitration terms, deductible math and the water cap explained on day one.
RIA formally submitted a request to establish a new six-digit code (624231, Emergency Restoration Services) in the 2027 NAICS review, covering water extraction, structural drying, odor control, fire/smoke restoration, microbial and mold remediation, trauma/crime-scene cleanup, and contents. IICRC filed supporting comments; RIA's Darren Hudema is fronting the push. The Economic Classification Policy Committee makes recommendations to OMB by March 2026, with updated codes published January 2027. Today the work is scattered across construction and remediation buckets that capture neither the specialty nor the emergency nature.
Why it matters
A real code means real data (verifiable industry size, revenue, and margin benchmarks instead of guesses) plus a cleaner path to disaster-response recognition, field-specific safety data, and lobbying weight. It's plumbing, but it's the plumbing that every "what's a fair multiple" and "who counts as essential in a disaster" conversation eventually needs.
The read
A classification request reads as bureaucratic noise, and it will not make one carrier pay a dollar more. It is the plumbing under every argument that follows. A federal code produces real revenue and margin data, which is what valuations, disaster contracting and insurance recognition all end up leaning on. File a supporting comment during the review window, and keep your own numbers in a shape you can benchmark.
Dillon argues the signed agreement has to govern the whole time-and-materials process, because invoices are increasingly hitting automated reviews, AI assessments, and "opaque reduction logic" that ignore contract terms. His fix: get the signed contract in the reviewer's hands first, keep daily narrative records tying labor/materials/equipment back to the agreement, and align the file to S500/S520 so a thin file can't be quietly discounted.
Why it matters
T&M is where large-loss margin lives and dies, and the thing reducing your invoice now may not be an adjuster reading it, it may be software scoring it. Contractors with agreement-anchored daily documentation defend charges; contractors with thin files get trimmed automatically.
The read
Better paperwork sounds like the answer a vendor would sell you, and some of it is. The part that is true anyway: the review side automated faster than the billing side, so software now scores your invoice before a person reads it. Anchor every daily entry to the signed agreement and the standard you worked to, and make the file machine-legible before it goes out.
Citing industry survey data, Diem reports 84% of restoration now embraces or is open to AI, up from roughly 70% three years ago. Email drafting is still the top use case at every company size; 36% name estimating and jobsite documentation as their priority AI application. Notably, hiring fell from the #1 industry problem to #2, with payment now ranked first. His thesis: adoption lags appetite because of the team conversation nobody's had, not the tech.
Why it matters
The "should we use AI" debate is basically over, the gap now is execution. If a third of the industry is pointing AI at estimating and documentation, the operators still doing it by hand aren't cautious, they're falling behind on the two tasks that decide margin.
The read
An adoption survey from a vendor selling adoption deserves a squint, and open to AI is not the same as using it. The number underneath is the one that matters: payment just passed hiring as the industry's top problem. That tells you where to point any tool you buy. Start on the collections side, draft the follow-up letters, and measure days to payment before and after.
Helixco released ROM Wizard, a cloud tool inside its commercial platform that walks an estimator through guided questions and loss measurements and auto-generates a rough-order-of-magnitude scope and estimate, including equipment selection, power distribution, and generator fuel math. CEO Mike Allred frames it as removing "a barrier to entry for newer estimators." The release names no Xactimate integration and doesn't call itself AI, it's a guided workflow with automated calculations.
Why it matters
The commercial large-loss ROM is where inexperienced estimators lose money fast, and vendors keep racing to productize that judgment. Whether it plugs into your Xactimate workflow or forces a parallel one is the unanswered question that decides if it's useful or just another silo.
The read
Another guided estimating tool sounds like more automation, which the trade generally treats as progress. The question is not whether it produces a number. It is whether it produces the file a carrier's software will read, because a standalone wizard that does not export into your estimate workflow just adds a screen. Before you buy, ask for a sample export and open it in the platform you bill from.
The NWS preliminarily rated a July 27 tornado in northeastern Wisconsin an EF3 that stayed on the ground more than 20 minutes, damaging homes and businesses. Separately, torrential rain drove flash flooding in West Virginia that killed two people, per the governor's office.
Why it matters
These are the mid-summer demand drivers that don't make national headlines but generate real water and structural mitigation work regionally. EF3 wind + flash-flood water is exactly the mixed-peril workload that fills a Midwest/Appalachia shop's board in August.
The read
Two regional storms are not a market signal, and building strategy around every declaration is how shops overextend. Stack them up across the summer though and the pattern is clear: the steady revenue this year is convective and inland, not coastal fire. That rewards local density and fast response. Check your Midwest and Appalachia coverage, and make sure a truck can be on site within two hours.