ClaimFetch — insurer-records concierge for independent vet clinics

1. One-liner

Turns every pet-insurer records request into a complete, insurer-ready packet in minutes — so pending claims stop piling up.

2. Trend signal — why now?

Pet insurance is the fastest-growing line in US insurance nobody talks about. US written premiums crossed $4B in 2024 and the market is projected to run $6.2B (2025) → $18.9B by 2033 at ~15% CAGR (GlobeNewswire, Mar 2026). More insured pets means more claims, and — because almost every insurer except Trupanion is reimbursement-based — more of those claims land back on the clinic as records requests.

Here’s the part the “submit a claim in one click” vendors gloss over: the initial submission is the easy 2 minutes. The pain is what happens after. Insurers “request these records directly from your vet, but if they’re not received after two requests, your claim is placed on hold” (Embrace help center). And they don’t want a discharge summary — “insurance companies need complete medical records from your practice management system, not simplified discharge sheets” (Pawlicy). So days after the visit, disconnected from any workflow, someone at the clinic has to pull the full SOAP history, match it to that insurer’s request, and send it — per insurer, per claim.

Administrative load is exploding underneath this: an FVE survey (Aug 2024–Jan 2025) found 64% of vets report their admin workload has doubled, insurance paperwork named as a contributor, and over 50% of that admin work is unpaid. Meanwhile the whole PIMS market is racing to bolt AI onto documentation (HappyDoc and every 2026 buyer’s guide), which means the underlying tech — reading a records request, extracting the right notes, assembling a packet — is now cheap and off-the-shelf.

Provenance:

• Signal 1 (Demand): Claims go on hold when clinics don’t answer insurer records requests after two tries; insurers need full PMS records, not discharge sheets — Embrace / Pawlicy — 2026
• Signal 2 (Feasibility): AI SOAP-note / medical-records extraction matured and commoditized across vet PIMS in 2025–26 — HappyDoc — 2026
• Signal 3 (Economic): US pet insurance $6.2B→$18.9B by 2033 (~15% CAGR); Trupanion/Lupa/VetEnvoy all funding the submission rail while leaving records/appeals open — GlobeNewswire — Mar 2026 Category: Underserved niche (tech-unlock adjacent)

3. The opportunity

Everyone built the front door and skipped the back door.

Trupanion’s direct-pay (11,500 clinics) only works on ezyVet and only for Trupanion. VetEnvoy is free and multi-insurer but handles submission — and it’s a UK-first tool (75% of UK practices) just now landing in North America. ezyVet+PetSure eClaims covers Embrace/Pets Best/ASPCA/Figo — but only if you’re an ezyVet shop. Every one of these optimizes the 2-minute initial submit.

None of them own the ugly tail: the insurer records request that arrives three days later, the claim stuck on hold pending “additional information,” the denial that needs a vet’s letter to appeal. That work is:

• Disconnected from the visit — it hits an already-busy front desk cold, with a policy number and a pet name and “we need the complete history.”
• Per-insurer and per-format — Lemonade wants it one way, ASPCA another; there is no “respond once.”
• PMS-fragmented — the independent clinic on Shepherd, Vetspire, Digitail, Provet or a legacy PIMS has no integration path for this at all.

That’s the wedge: a thin, PMS-agnostic layer that ingests a records request (email, fax, portal PDF), pulls the right records from a PMS export, assembles exactly what that insurer asked for, and returns a complete packet the clinic sends in one click — plus an appeals mode for denials. It’s the “un-stick the pending claim” tool, not the “submit the claim” tool.

4. Target market

• Primary customer: Independent US small-animal general practices, 1–3 DVMs, not on ezyVet (i.e. the Shepherd / Vetspire / Digitail / Provet / DaySmart / legacy-PIMS majority), with a growing share of insured clients. Decision-maker: practice owner or practice manager.
• Why they buy — in their words: The manual process is “kind of a headache,” and helping owners file “would sometimes take a while” to actually get reimbursed (Nadia Lee, Concierge Services Manager, Alii Animal Hospital & Resort, ezyVet case study). At volume it compounds — one insurance coordinator at Bowman Animal Hospital dedicates “one full day per week” to claims for ~300 insured clients (dvm360). The clinic eats this as unpaid labor and absorbs the owner’s frustration when reimbursement drags.
• Rough TAM reasoning: 30,000+ US veterinary practices (AVMA via co.vet), majority independent GP. Even the ~18–20k not locked to ezyVet’s rails, with insured-client bases growing 15%/yr, is a real market at $150–300/mo.
• Why now for them: Insured caseload is crossing the threshold where records requests stop being occasional and start being a weekly interruption — right as AI makes the packaging near-free.

5. Product sketch (MVP)

• Records-request inbox: forward the insurer’s email/fax/portal PDF (or connect a shared clinic inbox); ClaimFetch reads what’s being asked, for which pet, which date of service, which insurer.
• One-tap packet assembly: pulls the matching visit history + SOAP notes + itemized invoice from a PMS export and builds the exact documentation set that insurer requires.
• Insurer format library: knows the quirks — full-history vs. last-visit, whether a vet’s narrative letter is needed, preferred delivery channel (portal / fax / email).
• Appeals mode: for denied claims, drafts a vet-reviewed appeal letter citing the relevant record sections, so a “denied — insufficient documentation” becomes a one-click resend.
• Pending-claim tracker: a live list of every claim on hold and what each is waiting on, so nothing rots past the insurer’s second-request deadline.
• Owner-facing status: optional “your records were sent to [insurer] today” text, turning the clinic’s invisible work into visible goodwill.
• Audit trail: timestamped log of what was sent to whom — useful for both the practice manager and records-release compliance.

6. AI angle — what’s load-bearing

Remove the AI and this is a filing clerk. AI is doing the actual cognition:

• Reading the request: parsing an unstructured insurer email/PDF into “insurer X wants complete history for pet Y, DOS Z, plus a narrative on the presenting complaint.”
• Extracting the right records: finding and pulling the relevant SOAP notes and history out of a messy multi-page PMS export — the exact task 2026 vet-AI (HappyDoc et al.) proved is now reliable.
• Assembling per-insurer: matching extracted content to that insurer’s format and completeness bar, and flagging gaps before it’s sent (so it doesn’t bounce and go on hold again).
• Drafting appeals: generating a defensible, record-cited appeal letter for the vet to approve in seconds.

The 15-minute-per-request, disconnected-from-context task collapses to a 2-minute review. That’s the whole product.

7. Localization angle (if any)

N/A — this is a US-first play. The wedge is US-payer-specific: the reimbursement model, the specific insurers (Lemonade, ASPCA, Embrace, Healthy Paws, Nationwide, Pets Best), and their idiosyncratic records/format requirements. A UK/EU version is a plausible month-12 expansion — but VetEnvoy already dominates UK submission, so the US records/appeals gap is the cleaner opening.

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

• Pricing: $149/mo base per clinic (up to ~40 insured-claim events/mo), $249/mo growth tier (higher volume + appeals + owner-status texts). Priced as a fraction of the labor it removes and well under a single PIMS seat.
• ACV: ~$2,200/yr blended.
• Rough math to $1M ARR: ~455 clinics × $183/mo avg × 12 ≈ $1.0M. That’s ~2.5% of the non-ezyVet independent base.
• Rough math to $5M ARR: ~1,900 clinics on the growth tier, or ~2,300 blended — plus expansion revenue (see below). Requires that ClaimFetch becomes the default records/appeals layer for 2–3 mid-size PIMS ecosystems via partnership.
• Expansion path: per-appeal or per-recovered-dollar success fee on denials (aligns to outcomes, like Avoca’s model); “reimbursement-guarantee” concierge tier where ClaimFetch chases the on-hold claim to settlement; and a PIMS-partner revenue share as it becomes the records-response module those platforms don’t want to build.

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

• Practice-manager Facebook groups & VHMA: the Veterinary Hospital Managers Association and the big private practice-manager FB groups are where this exact complaint lives. Post the “we un-stick your pending claims” demo; DM managers who’ve complained about insurance paperwork.
• Ride the non-ezyVet PIMS. Shepherd, Vetspire, Digitail and Provet have no answer to records/appeals and market themselves to independents. Get listed in their integration directories and co-market — I’m the module they’d otherwise have to build.
• Insurance-coordinator direct outreach. Scrape the ~2,000 clinics publicly advertising an “insurance coordinator / claims” role or page; send a Loom showing their exact insurer’s request auto-packaged. These are the people feeling the “one full day per week.”
• Pawlicy adjacency. Pawlicy already sits in clinics doing insurance education/enrollment but not claims-response — a referral partnership reaches motivated, insurance-forward practices.
• State VMA newsletters & dvm360. The audience that reads “Fight your fear of pet insurance paperwork” is the audience that buys the thing that ends it.

10. Build complexity — justification

Medium. Off-the-shelf: LLM request-parsing, records extraction from PMS exports, appeal-letter drafting, doc assembly, a shared-inbox/fax intake. The custom work is the insurer-format library (the moat and the grind — each insurer’s completeness rules and delivery quirks) and clean ingestion of heterogeneous PMS exports (no universal API across Shepherd/Vetspire/Digitail/legacy, so start with PDF/CSV export + email-forward, add deeper hooks per PIMS partner). A focused pair ships a credible v1 for 2–3 insurers and 1–2 PIMS export formats in 10–14 weeks; breadth accrues after.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Records released with owner authorization; clinic remains custodian. Standard records-release consent covers it.
Ethical — no harm / dark patterns	✅	Speeds legitimate reimbursement owners are owed; no adverse party.
Market exists (evidence above)	✅	30k practices, ~15% CAGR insured growth, sourced staff complaints, funded adjacent players.
1–5 person team can build this	✅	Technical pair + a vet/practice-manager advisor for the format library.
Launchable with <$50K / ₹40L	✅	SaaS + LLM API costs; no capex.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	14/20	Real, recurring, unpaid pain that grows with penetration — but honest data (a receptionist calls filing “two minutes”; only high-volume clinics hit “a full day/week”) caps this below hair-on-fire for the average clinic. The intensity concentrates in insured-heavy practices.
Demand evidence	15	11/15	Sourced staff quotes, insurer-confirmed on-hold mechanics, market CAGR, funded adjacent players. Docked because the specific records/appeals pain is inferred from the mechanics, not yet a subreddit screaming for it.
Build feasibility	15	12/15	Core is off-the-shelf AI; the format library + heterogeneous PMS ingestion is the grind. 10–14 wk v1.
Distribution clarity	15	12/15	Named channels (VHMA, PIMS directories, coordinator scrape, dvm360). Conversion realistic but unproven.
Revenue mechanics	15	11/15	Pricing sits comfortably under a PIMS seat and under the labor saved; success-fee expansion is attractive. Depends on ~2.5% penetration of a real base.
Time to first revenue	10	8/10	Clear paid-pilot funnel to insurance-forward clinics; revenue in 6–8 weeks of MVP.
Defensibility	10	4/10	Soft moat: accumulating insurer-format library + PIMS-partner lock-in. But VetEnvoy or a PIMS could extend into appeals; a well-funded player could copy the wedge. Speed and format depth are the only edge.
Total	100	72/100

13. Qualitative modifiers

Founder-fit tags

technical-heavy · domain-expertise-required — needs an LLM/doc-pipeline builder plus a vet-office insider (or hired coordinator advisor) to encode the insurer-format library correctly. Get the formats wrong and packets bounce, which is the one thing that kills trust.

Key assumptions to validate (3–5)

1. Assumption: The average insured-heavy independent clinic loses ≥3–5 hrs/week to post-submission records requests and pending-claim chasing (not just initial filing). How to test: 25 structured interviews with practice managers at insured-forward clinics; ask them to time-log last month’s records requests, not their gut estimate.
2. Assumption: Clinics will pay $150–250/mo for a records/appeals tool that sits outside their PIMS. How to test: pre-sell 10 paid pilots off a Loom before building breadth.
3. Assumption: A workable packet can be assembled from PMS exports (PDF/CSV) without deep API integration for v1. How to test: collect real export samples from 5 different PIMS and prototype extraction on each.
4. Assumption: The insurer-format library is finite enough that covering the top ~8 insurers handles ~80% of volume. How to test: survey the claim mix across 20 clinics.

Risk flags

1. Incumbent extension: VetEnvoy (free, insurer-funded) or a PIMS could bolt “records response” onto existing submission. Mitigate by owning appeals + format depth + the non-ezyVet long tail fast.
2. Problem-intensity risk: If, post-interviews, the average clinic’s records/appeals burden is genuinely “15 min/week,” the buyer base narrows to high-volume insured practices — smaller than the 30k headline. This is the axis most likely to move the score.
3. PMS fragmentation: no universal export standard; ingestion breaks per PIMS. Contain scope by launching on the 2–3 most common independent PIMS first.
4. Data sensitivity: medical records handling — must be clean on authorization, storage, and audit trail from day one, or clinics won’t touch it.

14. Structured verdict

Score:                  72/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Technical founder (LLM + doc pipelines) with a veterinary practice-manager / insurance-coordinator co-founder or advisor
Time to revenue:        6–8 weeks from MVP (paid pilots)
Capital to launch:      $15–30K ($ / LLM API + design partner incentives)
Top 3 assumptions to validate first:
  1. Post-submission records/appeals burden ≥3–5 hrs/wk at insured-heavy clinics — 25 timed practice-manager interviews
  2. Willingness to pay $150–250/mo for an out-of-PIMS tool — pre-sell 10 paid pilots off a Loom
  3. Packet assembly works from PMS exports without deep API — prototype extraction across 5 PIMS export samples
Kill criteria:
  - Abandon if <6 of 25 interviewed insured-heavy clinics report ≥3 hrs/wk on records/appeals (problem too thin)
  - Abandon if <3 of 10 pilot targets convert to paid after seeing their own insurer's request auto-packaged
  - Abandon if VetEnvoy or a top-3 independent PIMS ships a records-response + appeals module before your v1

15. Next step — 1-week validation sprint

• Day 1–2: Recruit 25 practice managers at insured-forward independent clinics (VHMA groups, coordinator-job scrape). Ask each to time-log last month’s records requests and pending-claim chasing — real logs, not estimates.
• Day 3–4: Collect real (anonymized) insurer records-request emails/PDFs and PMS export samples from 5 clinics. Build a rough demo: paste one insurer’s request → auto-assembled packet. Record a Loom per clinic using their insurer.
• Day 5: Send the personalized Loom to 10 clinics with a $99/mo founding-pilot offer. Falsifiable go/no-go: GO only if (a) ≥6 of 25 report ≥3 hrs/week on the post-submission tail, AND (b) ≥3 of 10 pilots say yes to paid. Anything less and the problem is real but too shallow to build a business on — mark VALIDATE and revisit when insurance penetration climbs another cycle.