What changed
FACT: Grants.gov opportunity EDA-DISASTER-2025 (CFDA 11.307, Economic Development Administration) is posted with rolling submission and no stated close date (source: grants.gov/search-results-detail/359225). FACT: USAspending records a Department of Homeland Security disaster grant obligation of $35,301,159,434.96 to a 'Governor's Authorized Representative' for repair or replacement of disaster-damaged facilities. HYPOTHESIS: the eligible-entity list (EDDs, counties, municipalities, tribes, non-profits in presidentially declared disaster areas) and the specific paperwork chain (ED-900, scope of work, budget narrative, environmental review, Davis-Bacon certified payrolls, semi-annual ED-915) are inferred from the standard CFDA 11.307 construction-award shape, not asserted in the provided source text. That inference must be verified against the actual NOFO PDF before a line of code is written.
Why now
The rolling posting means applications are being assembled continuously rather than in one annual spike, so there is no single deadline to miss and no single deadline to sell against. HYPOTHESIS: disaster supplemental appropriations recur after every major declaration cycle, so the filer population refreshes rather than exhausts. The honest counterpoint: 'why now' is weak here. ED-900 is not a new form and the Davis-Bacon certified-payroll obligation is decades old. This is not a regulation that just changed and created a fresh forced-filer class overnight — it is a standing obligation with a new pot of money attached. Treat the 'why now' as money-driven, not rule-driven.
Converging signals
Three signals meet: (1) an open, rolling federal construction-grant opportunity aimed at local governments; (2) a very large adjacent federal obligation flowing to state-level pass-through entities for the same disaster-recovery purpose; (3) an inferred multi-year reporting tail (certified payroll during construction, semi-annual performance reports after) attached to every award. The convergence is real but shallow: the money and the paperwork are both long-standing. What converges is opportunity and capital availability, not a novel structural gap.
Customer pain
HYPOTHESIS (no PAIN evidence in the input — no complaint threads, no forum posts, no hiring evidence for grant administrators was supplied): a town of 4,000 people that wins a $3M EDA construction award has no grants administrator, must produce weekly certified payrolls for every contractor and subcontractor for the life of construction, and must file semi-annual performance reports for years. Getting this wrong triggers findings, disallowed costs, or clawback. Note clearly: the provided demand_evidence contains zero direct testimony of this pain. The FORCED BUYER item establishes that a filer class exists and must file; it does not establish that they are unhappy with how they file today, and it does not establish that they will pay a stranger to change it.
Who pays
Three candidate buyers, ranked by reachability: (1) Economic Development Districts — regional bodies that already administer grants for member municipalities, budget for software, and are a concentrated list of roughly 400 nationally; (2) grant-writing and grant-administration consultants who bill municipalities a percentage of the award and would use the tool to raise their own margin; (3) the municipalities themselves — the worst buyer, because purchase requires a council vote, a budget line, and often a procurement process, on a timeline uncorrelated with your runway. The municipalities are who the convergence description names. They are the wrong buyer.
Solved today
Assembled by hand in Word and Excel by a consultant or a clerk, with the ED-900 filled out from the prior award's copy. Certified payroll is where existing software already lives: LCPtracker, eMars, Points North, eComply, and Elation Systems have sold Davis-Bacon certified-payroll compliance to public agencies and contractors for well over a decade. Post-award grant reporting is likewise an established software category (AmpliFund/StreamLink, eCivis, Submittable, ZoomGrants). HYPOTHESIS on all vendor claims — none of these vendors appear in the provided sources; they are asserted from general knowledge and must be verified.
Why current solutions are bad
The consultant fee (commonly cited as a percentage of the award; HYPOTHESIS, unverified) is expensive relative to the marginal work of assembling a repeatable form. The incumbent certified-payroll vendors are priced and shaped for state DOTs and large prime contractors, not for a four-person public works department on one grant. That is a genuine seam. But 'the incumbent is priced for someone bigger' is the most commonly overestimated wedge in software: the incumbent can discount, and the small municipality's actual alternative is not a cheaper tool, it is the consultant they already trust.
Proposed product
Narrow it hard. Do not build all three legs. Build the ED-900 assembler alone: a structured intake that captures scope, budget narrative, and beneficiary/economic-impact data once, validates against the NOFO's stated requirements, and emits the submission-ready package for Grants.gov. Sell it to EDDs and consultants as a throughput tool — they file many of these, they feel the cost per filing, and they buy without a council vote. Treat certified payroll and ED-915 as a later expansion contingent on the first leg selling, not as MVP scope.
MVP version
One thing: ingest the actual EDA-DISASTER-2025 NOFO and ED-900 form, build a guided intake that produces a complete, validated application package. Ship a real filled ED-900 from a real prior award as the demo. No portal integration in v1 — Grants.gov submission is a workflow you can hand back to the user as a downloadable package; the automated-submission layer is a v2 that requires understanding Grants.gov's S2S API and the applicant's SAM.gov registration, and it is not where the first dollar comes from.
30-day build
Pure validation, no build. Pull the actual NOFO and ED-900 instructions and confirm or falsify every inferred item above (is ED-915 the real reporting form; is Davis-Bacon actually triggered; who is genuinely eligible). Get the list of ~400 EDDs. Call thirty of them and ask one question: who assembled your last ED-900 and what did it cost you. If the answer is consistently 'a consultant, for a percentage,' you have a business. If it is consistently 'our staff, in about a week, and it was fine,' kill this and do not spend the capital.
60-day build
Only if the calls validated. Build the assembler against the real form. Recruit two EDDs or two consultants as design partners who will run a live application through it. Charge them — a discounted price is fine, free is not, because a free pilot proves nothing about willingness to pay.
90-day revenue plan
HYPOTHESIS, low confidence: first paid application fee from a design partner. Do not model revenue at 90 days. Federal construction applications move on the grantee's timeline, not yours, and a rolling deadline removes the one forcing function that would compress the sales cycle. Realistic first meaningful revenue is 150-210 days.
Distribution path
The EDD list and the NADO (National Association of Development Organizations) membership are the entire addressable channel for leg one, and they are enumerable — this is cold outreach to a known list, not marketing. Publish a genuinely useful free artifact first: a parsed, annotated breakdown of the EDA-DISASTER-2025 requirements. The founder sells through demonstrated value; this is the demonstration. Conferences (NADO annual) matter here more than the founder would like, because public-sector buyers validate through peers.
Pricing hypothesis
HYPOTHESIS: $2,500-$5,000 per application package for the assembler, positioned explicitly against a consultant fee. Per-seat subscription for the reporting tail is premature to price before the assembler sells.
Technical difficulty
Low. Structured intake, validation rules, document generation. The hard part is not the code, it is the domain encoding — correctly representing what the NOFO requires, which is a reading-and-verification problem the founder is well suited to.
Legal / regulatory risk
Moderate and worth naming. Assembling a federal grant application on a recipient's behalf sits near, though probably not inside, the space where errors become the recipient's False Claims Act exposure. Certified payroll is worse: a wrong WH-347 is a Davis-Bacon violation with real consequences. If the certified-payroll leg is ever built, the liability posture needs a lawyer, not an assumption. This is a reason to sequence the assembler first.
Platform dependency
None meaningful. Grants.gov is a government system with no platform owner who can deplatform the tool. If v2 does automated submission, it depends on Grants.gov S2S remaining available, which is a low but nonzero risk.
Founder fit
Genuinely high, and the highest-confidence claim in this brief. The founder has already shipped a production app that submits certificates into a federal portal (FMCSA Training Provider Registry) and charges per upload. This is structurally the same motion: read a federal requirement, identify who is compelled to file, build the layer, charge per filing. The lesson at 0.80 confidence ('government-portal mandate opportunities fit this founder best') applies directly and is corroborated by the shipped ELDT product rather than merely asserted.
Breakout potential
Bounded. The ED-900 assembler alone is a low-hundreds-of-customers business at low-thousands per filing — a real, fundable, solo-scale product, not a breakout. The expansion thesis (same rail, different federal construction grant; then state pass-throughs, where the founder's own thesis correctly notes there are 50 near-identical markets) is where the ceiling lifts, and it is credible because the assembler pattern generalizes across NOFOs.
Final recommendation
PROCEED TO VALIDATION ONLY — do not build. The founder-fit is real and rests on a shipped product rather than a hypothesis, and the ED-900 assembler leg is a plausible, correctly-sized solo business. But the idea as delivered is three products stapled together, aimed at the wrong buyer, supported by one mismatched citation and a stack of inferences. The certified-payroll leg should be abandoned outright. The remaining leg is worth thirty phone calls and zero lines of code this month. If the calls reveal that EDDs and consultants are paying real money to assemble these packages, this becomes a build. If they reveal that staff handle it in a week, the correct outcome is to kill it and route the capital to a mandate with an actual deadline — which is what the founder's own thesis says to look for, and which this opportunity conspicuously lacks.
Next action
Download the EDA-DISASTER-2025 NOFO and the current ED-900 form from grants.gov/search-results-detail/359225, and verify — before anything else — whether ED-915 and Davis-Bacon certified payroll are actually required by this program. Then call thirty Economic Development Districts and ask what their last ED-900 cost them to assemble and who did it.