KennelClear — inspection-prep desk for USDA-licensed animal facilities

1. One-liner

Keeps a USDA-licensed breeder or kennel’s records audit-ready and flags the gaps an inspector cites before they arrive.

2. Trend signal — why now?

Three things converged in the last 16 months and they all point at the same boring, terrifying moment: an APHIS inspector standing in your kennel asking for paperwork you don’t have.

The federal crackdown is real and recent. In February 2025 the USDA Office of Inspector General published an audit slamming APHIS’s Animal Care program for going too soft. The headline finding: 80% of the dog breeders investigators visited still had uncorrected violations they’d already been cited for, and common non-compliant items (NCIs) were records and veterinary-care documentation gaps — missing exercise plans, missing medical records, no written Program of Veterinary Care. The political response to “USDA is too lax” is not “relax further.” APHIS conducted 9,700+ site inspections in 2024 with a rising share unannounced.

States are piling on, fast. Oklahoma HB1953 (2025) mandates unannounced inspections. Massachusetts now requires kennels to file annual reports to MDAR (starting June 2025) with rabies, spay/neuter, vet-care and breeding records attached. Indiana HB1165 (Jan 2026) toughened animal statutes. Florida launched its DAWG breeder program (April 2026) with mandatory licensing being debated. The number of facilities that now owe somebody a records binder is climbing every legislative session.

The tooling that exists is the wrong shape. The breeder software market — BreederHQ, Dog Breeder Pro ($5/mo), Empowered Breeder ($30/mo), Good Dog — is built for the show/hobby breeder: pedigrees, COI scores, genetics, puppy-buyer CRM, marketplace listings. None of them is built around the federal forms (APHIS 7005 acquisition, 7006 disposition, 7002A Program of Veterinary Care) and the specific NCIs an inspector writes up. There is a clean gap between “pretty breeding-program app” and “will this survive an unannounced USDA inspection.”

Provenance:

• Signal 1 (demand): USDA OIG audit — 80% of inspected breeders had uncorrected violations, records/vet-care among top NCIs; states adding unannounced inspections and annual reporting — https://usdaoig.oversight.gov/reports/audit/animal-care-program-oversight-dog-breeder-inspections + https://www.dogbreederlicensing.org/inspection-guide — Feb 2025 / 2025–2026
• Signal 2 (feasibility): Cheap multimodal vision/LLM (DeepSeek V4 Vision ~10× cheaper image cost, Gemini 2.5 Flash $0.15/$0.60) makes auto-structuring photographed vet docs, receipts and handwritten logs into 7005/7006/7002A fields trivially affordable — https://inference.net/content/llm-api-pricing-comparison/ — 2026
• Signal 3 (economic): APHIS oversees 17,500+ AWA-regulated entities and ran 9,700+ inspections in 2024; active breeder/kennel CRM market (Empowered Breeder $30/mo, BreederHQ won a 2026 G2 award) shows these operators already pay for software — https://www.akc.org/legislative-alerts/aphis-releases-2024-impact-report-enforcement-activity-summary-data/ — 2024 Category: Regulatory arbitrage

3. The opportunity

The incumbents sell a breeding program. KennelClear sells not losing your license. Those are different buyers’ moments even when they’re the same person.

What the breeder CRMs do badly, specifically: they store data the operator chooses to enter, in the shape the operator likes, with no opinion about whether that data would survive 9 CFR 2.40 / 2.75 scrutiny. They have no model of “your written Program of Veterinary Care is missing the parasite-control schedule,” “you have 14 dogs on premises but only 9 have current acquisition records,” or “your last documented vet exam for dam #7 is 13 months old and the standard is annual.” That gap-detection is the entire product. An AWA inspector walks a mental checklist; KennelClear runs that checklist continuously and tells you what’s red before the truck pulls up.

This is a regulatory-arbitrage play: the rules are public and the forms are standardized, but encoding them into a living “are you inspection-ready right now?” engine is work nobody serving this niche has bothered to do, because the niche looked too small and too unsexy for the venture-backed CRMs to specialize into.

4. Target market

• Primary customer: Owner-operators of USDA Class A breeders, Class B dealers, Class C exhibitors (petting farms, small zoos, sanctuaries) and state-licensed commercial kennels/boarding facilities — 1–10 staff, the person who is simultaneously the breeder, the bookkeeper, and the compliance officer.
• Why they buy: In their world, an NCI isn’t a paperwork ding — repeat uncorrected violations are how you get your license suspended and your animals confiscated. The OIG audit guarantees inspectors will be pushed to write more, not fewer. They are not worried about a fine; they are worried about losing the business. That’s a hair-on-fire problem the week before (and the week of) an inspection, and a low-grade anxiety the rest of the year.
• Rough TAM reasoning: APHIS oversees ~17,500 AWA-regulated entities; the licensed breeder/dealer core alone is ~3,200, exhibitors another ~2,400, plus a fast-growing state-licensed commercial-kennel layer (every state writing its own records + annual-report regime). A serviceable base of 8,000–12,000 facilities that genuinely face inspection is realistic.
• Why now for them: Unannounced inspections are spreading (OK), annual document filing is appearing (MA), and the federal posture flipped from lax to “the OIG is watching us.” The grace period of a friendly, scheduled, paperwork-light inspection is ending.

5. Product sketch (MVP)

• Inspection-readiness score — a single red/yellow/green dashboard answering “if APHIS walked in today, what gets cited?” mapped to actual 9 CFR records and vet-care standards.
• Snap-to-record capture — photograph a handwritten log, vet invoice, or rabies certificate; AI extracts and files it into the correct 7005/7006/7002A field with the date and animal ID.
• Living Program of Veterinary Care — guided builder that produces an APHIS-7002A-shaped document and then monitors it: flags overdue exams, lapsed vaccinations, missing parasite-control schedule.
• Acquisition/disposition ledger — every dog in, every dog out, with the federally required fields, reconciled against animals currently on premises so counts never drift.
• Gap alerts — “Dam #7’s exam is 13 months old,” “3 disposition records missing for May,” pushed weekly so corrections happen before, not during, an inspection.
• Corrective-action tracker — if you do get cited, log the NCI and the fix with dated evidence (the exact thing the OIG says breeders fail to document).
• Inspection-day export — one clean PDF/printout bundle an inspector can review, plus a “what they’ll ask for” prep checklist.
• State-pack add-ons — preloaded record/reporting requirements for high-activity states (OK, MO, MA, FL) layered on top of the federal baseline.

6. AI angle — what’s load-bearing

Two places, both load-bearing:

1. Unstructured-to-structured capture. These operators keep records on paper, in barn notebooks, and as photographed vet receipts. Cheap multimodal vision turns that mess into structured 7005/7006/7002A fields automatically — that’s the difference between “another data-entry chore I’ll skip” and “I take a photo and it’s filed.” Remove the AI and you’re back to a manual form-filler nobody maintains, which is exactly why current spreadsheets fail an inspection.
2. Gap reasoning over the regs. Encoding “is this facility inspection-ready?” means reasoning across records, dates, animal counts, and the written vet-care program against codified standards, then explaining each gap in plain English. That’s an LLM-shaped judgment task, not a SQL query.

If you removed the AI, you’d have a digital binder operators won’t keep current — which is the status quo that’s failing them.

7. Localization angle (if any)

N/A — this is a US-only regulatory play. The wedge is the localization: it’s built around US federal AWA / 9 CFR forms and US state animal-facility law. There’s no India/LatAm version because the regulatory regime is the product. (The same shape could later be ported to UK/EU animal-licensing regimes, but that’s a separate build, not a translation.)

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

• Pricing: $49/mo solo facility (federal baseline), $99/mo “Pro” (1 state pack + corrective-action tracking + inspection-day export), $149/mo multi-facility/multi-state. Annual discount to lock retention. This sits 2–5× above the hobby CRMs — justified because the buying trigger is license risk, not convenience.
• ACV: ~$700–900 blended (most on the $49–99 tiers).
• Rough math to $1M ARR: ~1,200 facilities at ~$70/mo blended = $1.0M. Against a serviceable base of 8,000–12,000, that’s ~12–15% penetration — aggressive but not fantasy for a category-of-one tool.
• Rough math to $5M ARR: Needs ~5,000 facilities or a higher ACV. Path: add the larger Class C exhibitors and research-adjacent facilities (richer recordkeeping, $300–500/mo), sell state-association bulk licenses, and add a “managed prep” services tier for operators who want a human to review before inspection.
• Expansion path: state packs (per-state upsell), animal-type expansion (the AWA covers more than dogs — cats, rabbits, exotics for exhibitors), and a per-inspection “readiness review” add-on.

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

• The license list is public. APHIS publishes the active-licensee list and the Public Search Tool exposes every facility plus its inspection history. Scrape it, segment to facilities with recent NCIs (they’re scared right now), and send a personalized “here are the 3 record types you were cited for — here’s how we keep them ready” email/letter. Recent-NCI facilities are the warmest 2,000 leads in the country.
• Go where they already gather. Pitch state breeder/kennel associations and the AKC’s Breeder Education / government-relations channels; these orgs actively push USDA compliance education and want a tool to recommend. One association webinar = a room full of exactly-right buyers.
• Ride the consultants. A small ecosystem of USDA inspection-prep consultants and attorneys already exists (e.g., inspection-prep guides online). Affiliate them — they bill hourly for prep work; KennelClear makes their clients easier and gives them recurring referral revenue.
• State-law moment marketing. When a state (OK, MA, FL…) passes a new records/reporting rule, that’s a dated event with press; show up in that state’s breeder Facebook groups and forums the week it lands with “here’s what you now have to keep — and a tool that does it.”

10. Build complexity — justification

Medium. The capture (multimodal extraction) and the dashboard are off-the-shelf AI + standard web stack. The genuine work is domain encoding: faithfully modeling 9 CFR records rules and the Program-of-Veterinary-Care standard into a gap engine, then layering state packs — that’s research-and-rules work, not novel ML, and it needs a domain advisor (a USDA-savvy vet or compliance consultant) to get right. Small team, ~3–4 months to a credible federal-baseline v1; state packs added incrementally after.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	It’s a recordkeeping/compliance tool; forms and regs are public. Not practicing law/vet medicine.
Ethical — no harm / dark patterns	✅	Improves animal-welfare recordkeeping; aligned with the regulation’s intent.
Market exists (evidence above)	✅	17,500+ regulated entities, active paid CRM market, OIG-driven enforcement pressure.
1–5 person team can build this	✅	Off-the-shelf AI + web stack + domain advisor.
Launchable with <$50K / ₹40L	✅	No capex; inference costs trivial at low volume.

All five pass.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	16/20	License-loss fear is hair-on-fire at inspection time; lower-grade the rest of the year. Real, recurring, documented by OIG.
Demand evidence	15	11/15	Strong indirect signal (OIG audit, state laws, paid CRM market). Weak on direct “I’d pay for this specific tool” verbatim demand — needs validation.
Build feasibility	15	12/15	Standard stack + AI capture; the rules-encoding and state packs are the real effort and need domain input.
Distribution clarity	15	10/15	Public license + NCI list is a gift; association/consultant channels plausible but conversion unproven.
Revenue mechanics	15	11/15	Pricing power from license risk; $1M needs ~12–15% penetration of a finite base — achievable but not trivial.
Time to first revenue	10	7/10	Warm scraped list + acute pain → first paid pilots in 6–8 weeks of a working v1.
Defensibility	10	5/10	Moat is accumulated rules/state-pack depth + record lock-in, not tech. A funded breeder CRM could add a compliance module — head start and focus are the defense.
Total	100	72/100

13. Qualitative modifiers

Founder-fit tags

domain-expertise-required · content-heavy (state-by-state rules content is the moat; pair a builder with a USDA-compliance/vet advisor).

Key assumptions to validate (3–5)

1. Assumption: Operators will pay $49–99/mo specifically for inspection-readiness, not just lump it in with their existing $30 CRM. How to test: Cold-email 200 recent-NCI facilities from the APHIS list with a paid-pilot offer; target ≥5% reply, ≥15 pilots.
2. Assumption: The federal records + vet-care rules can be encoded into a gap engine that operators trust enough to act on. How to test: Build the gap checklist with a USDA-compliance consultant and run it against 10 real (public) inspection reports — does it predict the actual NCIs cited?
3. Assumption: Associations/consultants will channel customers. How to test: Pitch 5 state associations + 5 prep consultants; secure ≥2 referral/affiliate commitments before build is finished.
4. Assumption: The serviceable base is genuinely 8,000+ inspection-facing facilities, not just the ~3,200 federal breeders. How to test: Quantify state-licensed commercial kennels in the top 8 states from public license rolls.

Risk flags

1. Niche-size risk: Finite, slow-growing base. $1M is reachable; $5M needs ACV expansion (exhibitors, services tier, state bulk deals). Don’t model this as hypergrowth.
2. Regulatory-whiplash risk: Enforcement intensity is political. A future administration could gut APHIS enforcement again and soften the buying trigger — though state-level rules diversify that risk.
3. Platform/data risk: Capture quality depends on third-party vision models; rules encoding must be kept current as 9 CFR and state laws change — content debt that compounds if neglected.
4. Incumbent-response risk: A breeder CRM with an existing base (BreederHQ) could bolt on a compliance module. Speed, depth of state packs, and compliance-first positioning are the only defense.

14. Structured verdict

Score:                  72/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Solo/duo technical builder paired with a USDA-compliance or veterinary-practice advisor
Time to revenue:        6–8 weeks after a federal-baseline v1 (warm scraped list + acute pain)
Capital to launch:      $8–15K (mostly advisor time + a few months of build)
Top 3 assumptions to validate first:
  1. Recent-NCI operators will pay $49–99/mo for inspection-readiness — cold-email pilot, target 15 paid pilots
  2. The gap engine predicts real cited NCIs — backtest against 10 public inspection reports with a compliance advisor
  3. Associations/consultants will refer — secure 2 affiliate commitments pre-launch
Kill criteria:
  - Abandon if <5% of 200 recent-NCI cold emails reply with interest
  - Abandon if the gap engine can't predict the NCIs in 7 of 10 backtested public inspection reports
  - Abandon if a funded breeder CRM ships a credible compliance module before your v1

15. Next step — 1-week validation sprint

• Day 1–2: Pull the APHIS active-licensee list + Public Search Tool inspection histories. Segment to ~500 facilities with NCIs in the last 18 months. Build the cold-outreach list with the specific records each was cited for.
• Day 3–4: Sit with a USDA-compliance consultant or AWA-savvy vet for two hours; draft the federal gap checklist and backtest it against 10 randomly chosen public inspection reports — does it predict the actual cited NCIs?
• Day 5: Send 100 personalized “you were cited for X; here’s how we’d keep it ready” emails offering a $49 founding-pilot. Go/no-go: ≥5 replies expressing interest AND the gap checklist correctly anticipated the NCIs in ≥7 of 10 reports. Both must hit, or rework before building.

The result is falsifiable: either scared, recently-cited operators raise their hand at a price, and the rules genuinely predict real citations — or they don’t.