ThawProof

1. One-liner

Rebuilds the one site-hour a slip-and-fall demand letter names, from your own logs plus certified NOAA weather.

2. Trend signal — why now?

Slip-and-fall is the single largest loss exposure in commercial snow removal, and the industry has quietly split into two halves: capture is solved, defense is not.

Every snow contractor of any size now runs a GPS-and-photo field app. Yeti Software starts at $95/mo and archives “every job, every photo, every timestamp — archived across every season.” QuoteIQ bundles timestamped before/after photos with GPS at $74.99–$149.99/mo. Housecall Pro and Nektyd sell the same promise. The 2020–2024 wave of field-service software did its job: the data exists.

What none of them do is the thing that actually happens next. A demand letter lands 12–23 months after the incident naming one address, one date, one approximate time. The contractor now has to pull the contract’s response-time clause, the service logs for that window, the salt/de-icer application record, the photos, the subcontractor assignment, and hourly certified weather — and assemble it into something an insurance adjuster or defense counsel can act on. Every one of those tools will happily export 4,000 rows of raw job history. None of them build the packet.

The expert-witness market prices this gap precisely. Forensic meteorologists charge $300–600/hr, with initial case reviews at $500–1,500, full reports $2,000–8,000, and total case costs $5,000–25,000. Certified NOAA data — the only weather record that clears the authentication bar for court — comes only from NCEI in Asheville, priced in 40-page increments with most orders starting at $191. A contractor facing a $10K nuisance settlement cannot justify a $8K meteorologist, so they settle. That’s the arbitrage.

The litigation pressure is not abstract. Late-January storms triggered more than 24,500 claims between two major insurers, producing $6.7 billion in insured losses. Brian Mahoney, Commercial Lines Manager at SWG Specialty, names the failure mode directly: “One common shortcoming is poor or nonexistent logging documentation, such as failing to log when snow was cleared or what time crews left a location.” Ontario’s 2021 amendment to the Occupiers Liability Act cut the notice period to 60 days, which compresses the window in which a contractor can still act on fresh memory.

Meanwhile money is visibly moving into the sector: a $13.4B snow and ice management services market in 2025, ~1,800 SIMA contractor members, and 16+ active private-equity roll-up platforms consolidating snow removal in 2026. PE-owned platforms carry real balance sheets and real claim reserves — they buy defensibility.

Provenance:

• Signal 1 (Demand): Snow contractors report slip-and-fall claims arriving up to 23 months post-incident; forum consensus that “9/10s of the fight is proper documentation” and cases “disappear rather quickly” when paperwork is complete — https://www.plowsite.com/threads/being-sued-for-a-slip-and-fall.122553/ — observed 2026-08-25
• Signal 2 (Feasibility): NCEI certified weather data is the court-authentication standard and is orderable programmatically; forensic meteorologists charge $300–600/hr and $2,000–8,000 per report, leaving a wide price gap under the expert-witness floor — https://www.ncei.noaa.gov/certification and https://weatherandclimateexpert.com/what-is-a-fair-price-for-an-expert-meteorologist-witness/ — observed 2026-08-25
• Signal 3 (Economic): $13.4B snow and ice management market (2025), ~1,800 SIMA contractor members, 16+ active PE roll-up platforms in 2026; one storm cluster produced 24,500 claims and $6.7B insured losses — https://ctacquisitions.com/guides/snow-removal-pe-rollup-tracker-2026/ and https://www.insurancebusinessmag.com/ca/news/commercial-liability/snow-removal-contractors-face-relentless-slipandfall-exposure-expert-warns-566589.aspx — observed 2026-08-25 Category: Underserved niche

3. The opportunity

The incumbents are field-service capture tools. Yeti, Nektyd, QuoteIQ, Housecall Pro — all of them optimize the moment the truck is on site. Their entire product surface is dispatch, route, photo, invoice. Yeti’s own liability pitch is data preservation: we keep it as long as you subscribe. That is a filing cabinet, not a defense.

The gap is temporal. The value of snow documentation is realized 12–24 months after capture, in an adversarial context, against a specific allegation, under a legal standard nobody in the contractor’s office understands. At that moment the contractor has:

• A field app full of raw records nobody has ever queried by site-and-hour
• A contract whose response-time clause they have not read since signing
• No certified weather record (their app shows a consumer weather widget, which is not admissible)
• An insurance adjuster who will settle for ~$10K rather than litigate, because settling is cheaper than building a defense

Attorneys handling these cases need nine document categories: incident reports, snow removal and maintenance records, certified weather records, snow removal contracts, property maintenance policies, site plans, photographs and video, witness statements, and insurance investigation files. The contractor is the sole source for four of them and the fastest source for two more. Today they assemble this by hand, badly, weeks late — and “waiting until later stages of litigation to collect important records can delay expert analysis and limit opportunities to investigate missing information.”

Doing this 10× better is not a modeling problem. It’s a retrieval-and-assembly problem against a legal template, which is exactly what current models are good at and exactly what a dispatch company will never build, because it sits outside their product identity and their sales motion.

Second, sharper opportunity: the same engine run forward is a scoring tool. Point it at last season’s records and it tells the contractor which sites have defense gaps before a claim lands — missing salt tickets, response times exceeding the contract clause, service windows with no photo. That converts a reactive purchase into a renewal-season subscription, which is what makes this a business rather than a per-incident service.

4. Target market

• Primary customer: Owner or operations manager at a commercial snow and ice management contractor running 5–75 sites in the US Snow Belt (Midwest, Northeast, Mountain West) and Ontario. $400K–$8M in seasonal revenue. Almost always a landscaping company in summer. Already pays for a field app. Carries $1M/$2M CGL, paying roughly $540–$1,260/yr in premium — trivial premium against a claim exposure that “can blow past a $1 million liability limit on its own.”

• Why they buy: In their words, from the forums — “the only thing to protect you from slip and falls is detailed records,” and “9/10s of the fight is proper documentation, if you have all your paperwork, with well documented times dates, and what you did, as the cases seem to disappear rather quickly.” One contractor described a claim for an incident “23 months ago” on a lot he “was not contracted to salt or sand… except on call out by landlord” — he won because he “kept very meticulous records of everything.” Another discovered his insurer had simply paid a claim without ever contacting him, which he only found out two years later while switching carriers. That last one is the emotional hook: your carrier settled in your name, raised your loss run, and never asked whether you had a defense.

• Rough TAM reasoning: ~1,800 SIMA contractor members is the identifiable, association-attached core, but SIMA membership undercounts the field — there is no clean census of pure-play commercial snow contractors, with real counts derived from cross-referencing SIMA rosters, state landscape-trade rosters, and IRS data. A realistic serviceable population of commercial contractors above the 5-site threshold in the US and Ontario is in the 8,000–15,000 range. At 600 customers and $2,400 ACV that’s $1.44M ARR — comfortably inside the target band without needing to win the category.

• Why now for them: Three things changed. Their data finally became machine-readable (the field-app wave completed). Certified NOAA weather became cheap and orderable relative to a $5K–25K expert. And PE consolidation means their competitors are now selling defensibility as a differentiator in bids — property managers increasingly demand service logs and exact on-site times, and “using vague invoices as proof: ‘Snow service completed’ won’t defend a serious claim.”

5. Product sketch (MVP)

• Claim intake: Paste or upload the demand letter / adjuster email. It extracts the site address, incident date, alleged time, and claimant details, and opens a case.
• Site-hour reconstruction: Pulls every record you hold for that address in the 72 hours around the incident — service visits, arrival/departure timestamps, GPS traces, photos, salt and de-icer application, subcontractor assignment, invoice lines.
• Certified weather ordering: Identifies the nearest qualifying NCEI station, generates the certification request, and attaches the hourly precipitation/temperature/freezing-rain record to the case. No consumer weather widgets.
• Contract clause matching: Reads the site’s snow contract and surfaces the operative terms — trigger depth, response-time window, salt-on-callout-only clauses, hold-harmless and indemnity language — then states plainly whether your documented response met the contract.
• Gap flagging: Tells you where the file is weak before your adjuster finds out. “No salt ticket for this event.” “Response logged at 3h12m against a 2h contractual window.” “Site visit has GPS but zero photos.”
• Defense packet export: One PDF, organized to the nine categories defense counsel actually asks for, with a plain-English timeline narrative on page one and exhibits behind it. Sent to your adjuster or attorney the same day the letter arrives.
• Off-season gap audit: Point it at all of last season. Ranked list of sites by defense exposure, so you fix documentation habits and renegotiate response-time clauses before the next winter.
• Integrations: Read-only pulls from the field apps contractors already run, plus a CSV/photo-folder path for the ones still on spreadsheets and camera rolls.

6. AI angle — what’s load-bearing

Remove the AI and this product does not exist. Three places it does real work:

Unstructured contract reading. Snow contracts are non-standard, per-property, often marked-up PDFs of a property manager’s template. The operative facts — trigger depth, response window, whether salting is scheduled or callout-only, who holds the indemnity — sit in prose that varies by every customer. Extracting “was this contractor obligated to be on site within two hours” from an arbitrary commercial services agreement is a language task. That single clause frequently decides the case: liability often hinges on response time rather than quality of work, and “if your contract specifies a two-hour response time and someone is injured at hour three, the quality of plowing doesn’t matter because you missed the contractual timeline.”

Cross-source timeline synthesis. GPS pings, photo EXIF, invoice lines, subcontractor texts, and hourly weather observations are five clocks that disagree. Fusing them into one defensible narrative — and being explicit about what is documented versus inferred — is judgment work over messy multimodal input.

Narrative generation to a legal standard. The output is a written timeline an adjuster reads in four minutes. Getting the register right — factual, non-speculative, no admissions, no overclaiming — is precisely what a well-prompted model does well and what a template engine does badly.

What is not AI: the certified weather record. That is a deliberate design choice. The weather exhibit must be an unmodified certified NCEI document, because its whole value is authentication. The AI reads and summarizes it; it never restates it as fact.

7. Localization angle (if any)

N/A as a language play — this is US and Ontario English. But there is a real jurisdictional localization that functions the same way:

Ontario is the sharpest single wedge. The 2021 Occupiers Liability Act amendment cut the notice period to 60 days, which means Ontario contractors get told about incidents while the season is still live and records are fresh — a much tighter, more urgent loop than the two-year statute of limitations common in US states like Illinois. Ontario is a concentrated, association-dense market where a 60-day clock makes the product feel like an emergency service rather than a filing habit.

Beyond that, state-by-state variation in premises liability standards (natural accumulation rules, ongoing-storm doctrines) is the substance of the “moat by accumulation” argument in section 3 — not a translation problem, but a per-jurisdiction knowledge base that has to be built and maintained.

8. Business model — path to $1M–$5M ARR

• Pricing: Seasonal SaaS, priced against the field app they already buy and the settlement they’re trying to avoid.

  • Solo/small (5–15 sites): $99/mo, 7-month winter season = $693/yr
  • Standard (15–50 sites): $249/mo seasonal, or $199/mo annual = $2,388/yr
  • Multi-branch / PE platform (50–300 sites): $599–1,200/mo annual = $7,200–14,400/yr
  • Per-claim rush packet (non-subscribers): $750 flat. This is the door-opener and it prices itself: it is 10% of a low-end forensic meteorologist report and roughly 7% of a typical $10K nuisance settlement.

• ACV: $2,400 blended. Realistic because it sits below the pain threshold — a single avoided nuisance settlement (~$10K) pays for four years, and contractors already spend $95–150/mo on the capture tool that doesn’t do this.

• Rough math to $1M ARR: 420 customers × $2,400 = $1.008M. Against a serviceable base of 8,000–15,000 contractors, that is 3–5% penetration. Achievable without category dominance.

• Rough math to $5M ARR: 1,400 customers × $2,400 = $3.36M, plus ~$800K from 60–80 multi-branch/PE platform accounts at $10K+, plus ~$500K in per-claim packets and off-season audits, plus the insurance channel below. Requires winning the PE roll-up platforms — which is plausible precisely because they’re consolidating and standardizing tooling anyway.

• Expansion path: Sites are the natural meter, and site counts grow every season. Then three upsells: the off-season gap audit (converts one winter purchase into year-round revenue), contract review at renewal season (flagging response-time clauses that are unwinnable before they’re signed), and the real prize — the carrier/broker channel. Insurers and specialty brokers writing snow CGL have a direct interest in contractors who can defend claims. A broker who bundles ThawProof into a policy is buying lower loss ratios. That’s a distribution channel that pays for itself and it’s the most likely path past $5M.

9. Go-to-market wedge — first 100 customers

This is a sales-heavy, list-driven motion into a small, association-dense, seasonally-panicked market. All four channels are enumerable today.

• The SIMA roster, worked by hand. ~1,800 contractor members, publicly identifiable, attending known events. SIMA runs certification and standards programs — this audience has already self-selected as the ones who care about doing it right. Work the list directly: personalized outreach naming their state’s premises liability standard and offering a free gap audit on last season’s records. At a 4% close that’s 72 customers from this list alone. Show up in person at the SIMA Snow & Ice Symposium — this is a trade-show market and the buyers are all in one room once a year.

• PlowSite and the forum tier. The forums are where these contractors already narrate their claim experiences in public — there are long-running threads titled “Being Sued for a slip and fall,” “Anyone ever been sued for a slip and fall and LOST?”, “Slip and Fall Insurance question.” These are not cold prospects; they are people describing this exact pain in writing, dated and searchable. Answer substantively with genuinely useful jurisdiction-specific information, offer the per-claim packet at $750 to anyone with a live demand letter. Every rush packet is a case study and a conversion path to seasonal subscription.

• Specialty insurance brokers writing snow CGL. A small, nameable set of specialty programs and brokers (SWG Specialty, Mitchell Joseph, E.G. Bowman, NIP Group and their peers) write this book. Their contractors’ loss runs are their problem. Approach 40 of them with a co-marketing offer: they introduce it to their snow book, contractors get a discount, the broker gets defensible files and better renewals. One broker with 200 contractor clients converting at 10% is 20 customers from a single relationship.

• Demand-letter timing, off the field apps. Contractors on Yeti/Nektyd/QuoteIQ are pre-qualified: they already pay for documentation and already believe in it. Build the read-only integrations first, then target their user communities with the honest pitch — “your app stored it, we defend with it.” This is a complement, not a competitor, which makes partnership conversations plausible rather than hostile.

The seasonality is a feature for GTM, not a bug: October–December is when contractors sign tooling for the season, and January–March is when demand letters and incidents peak. Two distinct annual sales windows, both predictable.

10. Build complexity — justification

Medium. Roughly 14–18 weeks to a credible v1 for a pair.

Off-the-shelf: document extraction and contract clause reading, timeline narrative generation, PDF assembly, photo/EXIF handling, standard web stack. The NCEI certified-data path is a defined public process with published ordering instructions, not a research problem.

The genuine custom work is threefold: read-only integrations against 3–4 field-service apps that were not designed to be integrated with (this is the real time sink and the reason it isn’t 6 weeks); a per-jurisdiction knowledge base of premises liability standards and notice periods, which requires paid legal review rather than model output; and getting the packet format right, which needs a defense attorney and a claims adjuster on retainer as design partners from week one. Not technically hard — but you cannot fake the domain, and shipping this without an attorney reviewing the output register is irresponsible.

Nothing here needs a novel model, proprietary data, or capital beyond two salaries and some legal hours.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Document assembly and organization, not legal advice. Hard product line: it states what your records show and where they’re weak — it never advises on liability or strategy. Attorney-reviewed disclaimers and output register; no UPL exposure if that line holds.
Ethical — no harm / dark patterns	✅	It surfaces accurate contemporaneous records, including records that hurt the contractor. Gap flagging tells them when they’re actually exposed. It does not fabricate, backdate, or alter — and the certified weather exhibit is passed through unmodified by design. A tool that made records look better than they were would be evidence tampering; this one is explicitly the opposite.
Market exists (evidence above)	✅	$13.4B market, ~1,800 SIMA members, 16+ PE roll-ups, forum threads spanning years, a priced expert-witness alternative at $5K–25K.
1–5 person team can build this	✅	Two people, 14–18 weeks, plus retained attorney/adjuster advisors.
Launchable with <$50K / ₹40L	✅	Two founder salaries deferred; ~$15–25K covers legal review, advisor retainers, SIMA event presence, and integration work.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	17/20	Genuine hair-on-fire at the moment of the demand letter — six-figure exposure against a $1M limit, and carriers settling in your name without asking for your defense. Docked 3 because the pain is episodic, not daily: a contractor may go two seasons without a claim, which weakens the “pay this week” urgency between incidents. The off-season audit is the deliberate answer to that.
Demand evidence	15	12/15	Multiple independent signals: years of dated forum threads, a named industry expert diagnosing exactly this failure, a priced substitute at $5K–25K, $13.4B market, active PE consolidation. Docked 3 because nobody is currently paying for this specific product — the $750 packet price is inferred from adjacent substitutes, not observed.
Build feasibility	15	11/15	Standard stack and off-the-shelf extraction, but 3–4 uncooperative field-app integrations plus a jurisdiction knowledge base requiring paid legal review push it past the 12-week line.
Distribution clarity	15	12/15	Four named, enumerable channels with real conversion math — a 1,800-member roster, public forum threads full of people describing the pain, a nameable set of specialty brokers, and an annual trade show. Docked 3 because it’s high-touch outbound into a market with no self-serve habit, and the broker channel is unproven.
Revenue mechanics	15	11/15	Pricing is benchmarked in both directions — above the $95–150/mo capture tools, far below the $5K–25K expert. $1M needs only 420 customers. Docked 4 for seasonality: seasonal-only pricing creates renewal cliffs every autumn, and the off-season audit that fixes this is an unvalidated assumption.
Time to first revenue	8	8/10	The $750 per-claim packet can be sold manually to a contractor with a live demand letter before the product is finished — the first dollar does not wait for v1. Docked 2 because subscription revenue is gated on the October–December buying window, so timing the launch wrong costs a full season.
Defensibility	10	5/10	Honest score. At month 3 this is copyable — it’s assembly and prompting. At month 12 the moat is the accumulated jurisdiction knowledge base, the packet formats that specific adjusters and defense firms have come to expect, and broker relationships. Real but soft. The genuine risk is Yeti or Nektyd shipping this as a feature; the counter is that it sits outside their product identity and sales motion, which buys time, not safety.
Total	100	76/100

13. Qualitative modifiers

Founder-fit tags

sales-heavy · domain-expertise-required

This is not a build-it-and-they-come product. It needs someone who will work a 1,800-name list by hand, stand in a booth at a snow trade show, and build relationships with specialty insurance brokers. It also needs real domain access — a defense attorney and a claims adjuster as design partners from week one, not as a later validation step. A purely technical founder will ship an elegant PDF generator that no adjuster respects.

Key assumptions to validate (3–5)

1. Assumption: Contractors will pay $750 for a rush defense packet when a demand letter is live. How to test: Find 10 contractors with active claims via PlowSite threads and SIMA contacts. Build the packet manually — no product, just work. Charge real money. If fewer than 3 of 10 pay, the acute pain doesn’t convert.
2. Assumption: The off-season gap audit is worth paying for, converting an episodic purchase into a subscription. How to test: Offer a free audit on last season’s records to 25 contractors, then ask for $249/mo to keep it running. Measure conversion. This is the single assumption the ARR math most depends on.
3. Assumption: Adjusters and defense counsel will actually accept and prefer this packet format. How to test: Take three manually-built packets to five claims adjusters and three defense attorneys. Ask directly whether it changes their reserve or settlement posture. If the answer is “nice, but we’d rebuild it anyway,” the value prop collapses.
4. Assumption: Read-only integration with Yeti/Nektyd/QuoteIQ is technically and contractually feasible. How to test: Attempt integration with two of them in week one. If all paths are blocked, the product falls back to CSV import, which raises friction significantly.
5. Assumption: Specialty brokers will co-market. How to test: Pitch 10 brokers. If none will make an introduction to their snow book, the $5M path loses its strongest channel.

Risk flags

1. Incumbent feature risk: Yeti, Nektyd, or a PE-backed consolidator ships “claim defense export” as a checkbox. Most likely path to death. Mitigation is depth — jurisdiction knowledge and adjuster-accepted formats are harder to clone than a PDF button, and speed matters more than the moat.
2. Regulatory / UPL risk: The line between “organizing your records” and “giving legal advice” is real and state-specific. Cross it and the product becomes unauthorized practice of law. Requires attorney-designed output constraints, not a disclaimer bolted on at the end.
3. Seasonality / cash-flow risk: Revenue concentrates October–March. Renewal happens once a year in a narrow window, and a bad autumn is a lost year. The off-season audit exists specifically to blunt this, and it’s unproven.
4. Episodic-pain risk: A contractor who goes two claim-free seasons will churn, because the product’s value is invisible when nothing goes wrong. Insurance has this problem too and solves it with mandate and habit — this product has neither yet. The broker channel is the strongest available answer.
5. Evidence-integrity risk: A product that assembles litigation evidence must be provably incapable of altering it. Any perception of backdating or enhancement destroys the company and exposes customers. Requires immutable audit logging and passing certified weather through untouched — a design constraint from day one, not a v2 feature.

14. Structured verdict

Score:                  76/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Sales-driven operator with snow/landscaping or insurance-defense
                        domain access, paired with one technical co-founder. Retained
                        defense attorney and claims adjuster as design partners.
Time to revenue:        4–6 weeks to first manual $750 packet; subscription revenue
                        gated on the Oct–Dec buying window
Capital to launch:      $15–25K (legal review, advisor retainers, SIMA presence,
                        integration work) plus deferred founder salaries
Top 3 assumptions to validate first:
  1. Acute-pain conversion — manually build and sell 10 rush packets at $750 to
     contractors with live demand letters; need 3+ closes
  2. Subscription conversion — free gap audit for 25 contractors, then ask $249/mo;
     this is what the entire ARR model rests on
  3. Adjuster acceptance — put 3 manual packets in front of 5 adjusters and 3 defense
     attorneys; ask whether it moves their reserve or settlement posture
Kill criteria:
  - Abandon if fewer than 3 of 10 contractors with live demand letters pay $750 for a
    manually-built packet
  - Abandon if fewer than 15% of free-gap-audit recipients convert to a paid seasonal
    subscription
  - Abandon if adjusters and defense counsel say they would rebuild the file regardless
    of format — that means the packet has no purchaser downstream
  - Abandon if Yeti or Nektyd ships an equivalent claim-defense export before v1 launch

15. Next step — 1-week validation sprint

The whole sprint is designed to sell something before building anything. No product, no landing page, no code.

• Day 1–2: Mine PlowSite, SIMA contacts, and specialty broker referrals for contractors with a live slip-and-fall claim — not a hypothetical one. Target a list of 30. The qualifying question is one line: “Do you have a demand letter or adjuster request open right now?”
• Day 3–4: For the first 3 who say yes, build the defense packet by hand. Order the certified NCEI record. Read their contract. Assemble the timeline. Spend 6–8 hours per packet. Charge $750 up front, not on delivery.
• Day 5: Take those packets to 5 claims adjusters and 3 defense attorneys. One question, asked flatly: does this change your reserve or your settlement posture? Separately, ask the 27 contractors who didn’t have a live claim whether they’d pay $249/mo for an off-season audit.

Go / no-go: Proceed only if ≥3 contractors paid the full $750 in advance and ≥3 of 8 adjusters/attorneys say the packet changes their posture. If contractors pay but the downstream professionals shrug, this is a feel-good product with no economic buyer — that’s a no-go, and it’s better to learn it in week one for the price of three manual packets than after 16 weeks of integration work.