CertPreflight — pre-submission death-certificate validator for US funeral homes

1. One-liner

Catches the demographic and jurisdiction errors that get a death certificate kicked back, before the director submits it to the state.

2. Trend signal — why now?

Three things converged in the last 12 months.

The error problem is documented and expensive. Death certificates get rejected or held for single, mundane mistakes — name mismatch vs. ID, missing suffix, blank SSN, wrong registration jurisdiction (death registered where it occurred, not where the person lived), skipped authorization signatures, and out-of-sequence filing (some states require the death record filed before the cremation authorization). Funeral.com and a Continental Funeral Home operations post both lead with the same line: “accuracy on the first submission” is the only way to avoid the spiral. NYC Health quotes 12 weeks to process a death-certificate correction. A held certificate freezes the family’s life-insurance payout, pension, probate, and account closures — the consequences land on a grieving family, not the funeral home, which makes the director the one chasing it.

The labor side broke. Industry reporting (Carolina News, Funeral Director Daily) says >60% of US funeral directors plan to retire by 2028, mortuary enrollment is not backfilling, and small/rural homes increasingly run with one licensed director on call 24/7/365, doing 80–100 cases a year at 60+ hour weeks. There is no second pair of eyes to catch the bad field before submit.

The tech got cheap and a market formed around it. Voice→structured-data extraction and rule-checking became off-the-shelf in 2025–26. The funeral software market is ~$450M in 2026 heading to ~$1B by 2033 (~10% CAGR), and AI is already entering deathcare — Afterword launched “Grace” (photo → digital case file), ClosureMD runs AI voice agents for inbound family calls and outbound death-certificate follow-ups with physicians. Money and attention are moving here, but nobody is selling the thin pre-submission reject-rule check — the incumbents store the case, the new entrants chase the physician signature, and the demographic/jurisdiction error is still unguarded.

Provenance:

• Signal 1 (demand): Documented common death-certificate rejection causes + 12-week NYC correction timeline + family financial consequences — https://funeral.com/blogs/the-journal/why-death-certificates-get-delayed-and-how-to-prevent-it , https://www.continentalfuneralhome.com/blog-posts/13357/common-mistakes-that-delay-death-certificates-and-how-to-avoid-them , https://www.nyc.gov/site/doh/services/birth-death-records-corrections-death.page — observed 2026-05-16
• Signal 2 (economic): Funeral software market $450M→$1B by 2033 (~10% CAGR); >60% of funeral directors retiring by 2028 leaving solo/understaffed homes — https://www.cognitivemarketresearch.com/funeral-home-software-market-report , https://carolinanewsandreporter.cic.sc.edu/more-than-60-of-funeral-directors-nationwide-are-about-to-retire-do-you-want-this-job/ — observed 2026-05-16
• Signal 3 (feasibility): AI entering deathcare in 2025–26 — Afterword “Grace” case-file builder and ClosureMD AI voice agents — confirming voice/LLM extraction is now off-the-shelf and a buyer category exists — https://www.funeralvision.com/afterword-introduces-grace-the-first-ai-assistant-built-for-funeral-homes/ , https://afterword.com/ — observed 2026-05-16 Category: Underserved niche

3. The opportunity

Every incumbent (SRS Computing, Passare, CRÄKN, Osiris) is a full practice-management suite. They store the case and render the death-certificate worksheet. None of them validate the worksheet against the specific state EDRS’s rejection rules before the director hits submit. That’s the gap.

The two AI new entrants miss it from different sides:

• Afterword “Grace” turns arrangement worksheets into a digital case file. It captures data; it does not check whether that data will be rejected. Per the launch coverage, it addresses no death-certificate validation of any kind.
• ClosureMD runs voice agents for inbound family calls and outbound physician follow-up on the cause-of-death/signature side. That’s the medical-certifier delay, not the funeral-director-owned demographic block (items 1–23 / 51–55 on the certificate — name, SSN, jurisdiction, parents’ names, marital status, authorization signatures).

The opportunity: a thin, fast pre-flight checker that a one-person funeral home runs in 60 seconds before submitting in their state EDRS, that turns “submit and pray” into “submit clean the first time.” It doesn’t replace the suite — it sits beside it. The 10× isn’t a prettier UI; it’s converting a multi-week, family-facing correction crisis into a caught error before submission.

4. Target market

• Primary customer: Owner/operator or sole licensed funeral director at an independent US funeral home — 50–300 cases/year, 1–4 staff, no dedicated administrative person, disproportionately rural and small-town.
• Why they buy: In their words, the pain is having to call a grieving family back days later because the registrar kicked the certificate for a missing SSN or a maiden-name mismatch — while the family’s life-insurance check is frozen and the burial slips. It is professionally embarrassing, emotionally brutal, and (per New York law) furnishing incorrect informant information is even a misdemeanor exposure. They already eat the rework hours; they have no tool that prevents the error.
• Rough TAM reasoning: ~19,000 independent US funeral homes (≈89% of ~21–24K total are family/independent-owned). At a realistic 8–12% reachable paying base over a few years and $150–250/mo, that’s a $3–7M ARR ceiling — exactly the bootstrapped band, too small to interest the VC-funded suites as a standalone.
• Why now for them: The director who used to have an experienced colleague double-check the worksheet is retiring and not being replaced; EDRS is now mandatory in most states so the paper fallback is gone; and the family’s expectation of a fast insurance payout has not relaxed.

5. Product sketch (MVP)

• Records (or accepts typed/photographed notes from) the arrangement conference and auto-drafts the vital-statistics worksheet — names, SSN, DOB, birthplace, parents’ legal names, marital status, informant relationship.
• Runs a state-specific reject-rule check: flags blank required fields, name/suffix/maiden-name inconsistencies, SSN format problems, wrong registration jurisdiction (occurred vs. resided), and missing authorization signatures.
• Enforces filing sequence for the selected state (e.g., “this state requires the death record filed before the cremation authorization — you are out of order”).
• Produces a one-page pre-flight report: red (will be rejected), amber (commonly queried), green (clean), each line citing the specific state rule.
• “Ask the family this” punch-list: the exact follow-up questions to close amber/red items in a single call, scripted for a grieving informant.
• Plain export/copy-out so the director keys the verified data into whatever EDRS/suite they already use — no integration required to get value on day one.
• Per-state rule packs, starting with the 5–8 highest-volume states, each independently togglable.

6. AI angle — what’s load-bearing

Two AI jobs, both load-bearing. (1) Extraction: turning a messy, emotional, non-linear arrangement conversation (or scanned worksheet) into clean structured vital-stats fields — this is the work that otherwise eats the director’s evening and introduces transcription errors. (2) Reasoning over inconsistency: “the ID says “Robert A. Smith Jr.”, the medical record says “Bob Smith”, the informant said “Smith” — this triggers a registrar query” is fuzzy cross-document judgment, not a regex. Remove the AI and you’re back to a static PDF checklist nobody fills in under stress — which is exactly the status quo that fails today. The codified state reject-rule library is the deterministic spine; AI is what makes a tired solo director actually use it.

7. Localization angle (if any)

N/A — this is a US-only play by design. The entire moat is the state-by-state EDRS rejection-rule knowledge; “localization” here means per-state rule packs, not per-country. A generic global product is impossible because death registration rules are jurisdictionally fragmented within the US itself — that fragmentation is the wedge, not a weakness.

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

• Pricing: $179/mo per funeral home (single location), flat. Multi-location/group add-on at $99/mo per extra location. Positioned as an add-on beneath the $200–400/mo PM suite, not a replacement.
• ACV: ~$2,150/year for a single-location home.
• Rough math to $1M ARR: ~465 single-location homes at $179/mo ≈ $1.0M ARR. Out of ~19,000 independents that’s 2.4% penetration.
• Rough math to $5M ARR: ~2,000 homes (≈10.5% of independents) plus modest multi-location expansion. Requires the rule library to cover ~20 states and a referral/association engine working — credible but not trivial.
• Expansion path: more state rule packs (each unlock = renewal/upsell justification); a “family-facing intake link” so the informant pre-fills demographics before the conference; an annual “audit pack” showing the home’s clean-submission rate for risk/insurance and franchise-quality conversations.

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

• State funeral director associations. ~50 state associations run paid CE and member newsletters. Sponsor/present a 30-minute CE-style session: “The 9 fields that get your certificates kicked back in [State].” Domain-credible, association-distributed, gets you in front of exactly the solo operators. Target 3 states first.
• The rule-pack-as-lead-magnet. Publish a free, genuinely accurate “[State] death-certificate rejection checklist” PDF per launch state. This is high-intent SEO no incumbent bothers to write because admitting their intake errors out is off-brand for them. Gate the interactive version.
• Direct outreach to the long tail. ~19,000 homes are in public state-license registries and directories. Cold email/postcard the single-location independents with a one-line hook and a 60-second Loom of a real worksheet being caught — expect low single-digit reply but the list is large and the pain is specific and named.
• Trade press + the existing AI-deathcare narrative. FuneralVision and Funeral Director Daily are already covering AI in deathcare (Grace, ClosureMD). A sharply-positioned “the thing that catches the error before you submit” angle is a natural follow-on story in a beat that’s actively running.

10. Build complexity — justification

Medium. The extraction and reasoning are off-the-shelf LLM/voice APIs; the web app is standard. The real work is building and maintaining the per-state reject-rule library — that’s domain-research labor (reading EDRS funeral-director handbooks, state vital-records bulletins, calling registrars), not novel engineering. A pair can ship a credible 5-state v1 in ~12–16 weeks; the rule library is the ongoing moat-building cost, not a v1 blocker. No integrations required for the wedge (copy-out is fine), which keeps v1 honest and fast.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	A pre-submission advisory checker; does not file or certify anything itself.
Ethical — no harm / dark patterns	✅	Reduces errors that harm grieving families; clearly positioned as advisory, “verify against your state”.
Market exists (evidence above)	✅	Documented rejection causes, 19K independents, funded adjacent AI entrants.
1–5 person team can build this	✅	Off-the-shelf AI + standard web; rule library is research labor, not R&D.
Launchable with <$50K / ₹40L	✅	Solo/pair, API costs minimal at low volume, no capex.

All five pass.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	16/20	Painful, recurring, financially and emotionally costly — but per-case, not literally daily for a solo home; strong workaround-pain rather than hair-on-fire every morning.
Demand evidence	15	12/15	Multiple independent documented signals + funded adjacent AI entrants prove a buying category; lacking direct verbatim “I’d pay for this” quotes (deathcare professionals are low-visibility online).
Build feasibility	15	12/15	Off-the-shelf AI + standard stack; the state rule library is real, ongoing, unglamorous work that gates breadth, not v1.
Distribution clarity	15	11/15	Named channels (state associations, license registries, trade press) with credible math, but conversion of conservative solo operators is uncertain.
Revenue mechanics	15	11/15	Pricing benchmarked below the suites; $1M path needs only 2.4% penetration; $5M needs wide state coverage + retention assumptions.
Time to first revenue	10	7/10	A useful 3–5 state v1 in ~12–16 weeks; first paid likely 8–12 weeks post-launch via association demo.
Defensibility	10	5/10	The state reject-rule library is a real accumulating moat, but Afterword/ClosureMD are funded, adjacent, and could extend into this lane — execution + niche focus, not a hard wall.
Total	100	74/100

13. Qualitative modifiers

Founder-fit tags

domain-expertise-required · technical-heavy

A founder with (or paired with) a licensed funeral director or vital-records insider builds the rule library 5× faster and with credibility that survives the association stage. Pure engineers without that access will produce a checklist that registrars laugh at.

Key assumptions to validate (3–5)

1. Assumption: Solo/small funeral directors will pay ~$179/mo for an error-prevention add-on on top of their existing PM suite. How to test: 25–30 structured calls with single-location operators via two state associations; ask the rework-frequency and willingness-to-pay questions directly.
2. Assumption: Demographic/jurisdiction rejections (the funeral-director-owned fields) are frequent and painful enough to drive purchase — not a rare edge case dwarfed by cause-of-death/signature delays. How to test: Ask 30 directors for their last-12-months count of kicked-back/corrected certificates and the cause split.
3. Assumption: Per-state reject rules can be codified accurately enough to be trusted without becoming an unmaintainable treadmill. How to test: Fully build and registrar-verify 3 states; measure research hours per state and rule volatility over one quarter.
4. Assumption: State associations will give a CE/newsletter slot to an outside vendor. How to test: Pitch 5 state associations; book ≥2 sessions.

Risk flags

1. Competitive encroachment: Afterword and ClosureMD are funded and one feature-release away from this lane. Mitigation: go deep on reject-rule breadth and association trust faster than they bother to.
2. Liability/positioning risk: If a customer treats the green light as a guarantee and a certificate still gets rejected, blame can land on the tool. Mitigation: advisory framing, “verify against your state”, and a track-record/audit framing rather than a warranty.
3. Maintenance treadmill: States change EDRS rules and forms; a stale rule pack is worse than none. Mitigation: per-state versioning, registrar relationships, and pricing that funds the upkeep.
4. Market timing: Deathcare adopts conservatively and slowly; the retirement wave that creates the need also thins the buyer pool and removes the very people who appreciate the problem.

14. Structured verdict

Score:                  74/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Technical founder paired with a licensed funeral director or vital-records insider
Time to revenue:        8–12 weeks post-launch (≈3–4 months from start incl. 5-state v1)
Capital to launch:      $8–15K (solo/pair, API + domain research time)
Top 3 assumptions to validate first:
  1. WTP ~$179/mo as a suite add-on — 25–30 association-sourced operator calls
  2. Demographic/jurisdiction rejection frequency is purchase-driving — last-12-months count + cause split from 30 directors
  3. Per-state rules are codifiable and stable — fully build + registrar-verify 3 states, measure hours and volatility
Kill criteria:
  - Abandon if <20% of 30 interviewed directors report ≥3 demographic/jurisdiction kickbacks in the last 12 months
  - Abandon if zero of 5 pitched state associations grant a CE/newsletter slot within 8 weeks
  - Abandon if Afterword or ClosureMD ships an equivalent pre-submission validator before your v1 reaches 3 verified states

15. Next step — 1-week validation sprint

• Day 1–2: Build the codified reject-rule pack for ONE high-volume state from its EDRS funeral-director handbook + vital-records bulletins. Call that state’s vital-records office once to confirm the top rejection reasons. Falsifiable output: a rule list a registrar agrees with.
• Day 3–4: Get the list of that state’s licensed funeral homes from the public registry. Call 25–30 single-location operators with three questions: how many certificates got kicked back in the last 12 months, what for (demographic vs. cause-of-death), and would they pay $179/mo for a pre-submit checker.
• Day 5: Decide. Go only if ≥20% report ≥3 demographic/jurisdiction kickbacks in 12 months AND ≥6 of 30 say they’d pay at $179/mo. Anything less = the pain is real but not purchase-driving — VALIDATE or PASS, do not build.