What changed
FACT (Federal Register, 2026-05-29): OMB proposes to revise the government-wide Guidance for Federal Financial Assistance (2 CFR 200), changing transparency, accountability and oversight requirements for Federal awards, adding recipient accountability standards and nondiscrimination conditions, clarifying the regulatory status of OMB requirements, and directing listed grant-making agencies (HHS, USDA, State, USAID, VA, DOE, Treasury and others) to make conforming changes to their adopting regulations. HYPOTHESIS: the conforming-regulation cascade means a recipient does not face one diff but a dozen agency-specific diffs.
Why now
FACT: this is a proposed rule in an open comment period — the substance is public but the compliance date has not attached. HYPOTHESIS (the core timing bet): the window between proposal and final adoption is exactly when a solo operator can build the diff engine, seed it with the proposed text, and be the only vendor already shipping on the day the final rule publishes. Every prior Uniform Guidance revision (2014 promulgation, 2020 revision, 2024 revision) triggered a wave of policy-manual rewrites — INFERENCE from the structure of the rule, not asserted by this document.
Converging signals
Three things meet at one point: (1) a rule that changes what recipient policy manuals must say; (2) a defined, enumerable filer class — every recipient and subrecipient of federal financial assistance, plus state pass-through agencies that must re-paper subrecipient monitoring; (3) an artifact that must change in response (the policy manual, the subaward template, the risk-assessment form). The weakness versus a classic forced-filer play: there is no single portal with a submit button. The obligation is proven at audit (Single Audit / 2 CFR 200 Subpart F), not at filing. That makes this a compliance-documentation product, not a per-transaction submission product.
Customer pain
HYPOTHESIS (not established by the source text): a grants-compliance officer at a mid-size nonprofit, county, or university has a 60-200 page policy manual written against the prior Uniform Guidance. When the rule changes, she does not know which specific paragraphs are now non-conforming. Today she either reads the final rule herself against the manual, waits for her audit firm to tell her at the next Single Audit (i.e. after the finding), or buys a generic template that does not reflect her org's actual text. The source document does not evidence this pain; it is inferred from the structure of the obligation.
Who pays
Primary: the grants-compliance or finance director at organizations with $750k+ in annual federal expenditures (the Single Audit threshold), who has budget authority for sub-$10k software on a purchasing card. Secondary and higher-value: state pass-through agencies that must re-issue conforming subaward terms to hundreds of subrecipients. Tertiary and probably the fastest first dollar: the grants-consulting and CPA firms who currently do this manually and would white-label the diff.
Solved today
HYPOTHESIS, all of it — none is asserted by the source: (a) subscription guidance publishers (Thompson Grants' Federal Grants Management Handbook, Bloomberg/Wolters Kluwer equivalents) sell interpretive updates but do not touch the customer's own document; (b) grants-management platforms (eCivis/Euna, AmpliFund, Streamlyne, Sage Intacct) manage awards and drawdowns but do not diff policy text; (c) CPA and grants-consulting firms bill hourly for policy-manual updates; (d) national associations (NGMA, NCURA, GFOA) publish free crosswalks that are generic, not customer-specific.
Why current solutions are bad
A guidance subscription tells you the rule changed; it does not tell you that section 4.2 of *your* procurement policy now conflicts. A consultant will tell you, at $200-350/hr, once. Neither produces a durable, re-runnable artifact when the agency conforming regulations land months apart. The generic crosswalk is free and therefore also worthless as a defense at audit, because the auditor tests your policy, not the crosswalk.
Proposed product
Ingest the customer's existing policy manual (PDF/DOCX). Maintain a structured, section-level representation of 2 CFR 200 as proposed, as final, and per agency adopting regulation. On each rule change, run a section-to-section semantic mapping between the rule and the customer's document, and output: (1) a conformance report — every rule section, mapped to the customer text that satisfies it, or flagged as unaddressed; (2) redline suggestions with conforming language; (3) a dated, exportable audit-ready memo showing what changed, when the customer was notified, and what was remediated. Plus a free public change-monitor (email alert on any docket movement) as the top of funnel.
MVP version
Two weeks of the diff engine and a manual delivery model. Scrape the eCFR API for 2 CFR 200 (structured XML, free, no auth) and the Federal Register API for the docket and agency conforming rules (both free, both stable — no platform risk). Section-split with a deterministic parser, embed, and mutual-kNN map rule sections to customer policy sections. Human-in-the-loop review of every mapping before delivery. First ten customers get a PDF conformance report that Charles reviews personally. Do not build a portal, a workflow, or a user-management system before revenue.
30-day build
Build the free change-monitor: an eCFR + Federal Register watcher on docket 2026-10817 and every listed agency's adopting regulation, publishing a public changelog page and an email alert. File a substantive public comment on the proposed rule under the founder's name — it becomes a permanent, citable, indexed artifact that establishes standing with exactly the audience being sold to. Simultaneously, hand-build conformance reports for three real policy manuals (municipal, nonprofit, university — pull them from the web; most are public documents) to prove the mapping quality and to use as sales artifacts.
60-day build
Sell the manual service, not the software. Target the consulting and CPA layer first: firms with a federal-grants practice already have the client list and the billing relationship, and they buy tools that make an hourly engagement faster. Ten discovery calls sourced from NGMA and GFOA member directories and from the comment docket's own commenter list (public, named, self-selected as caring about this rule). Price the first engagements at $2,500-5,000 as a one-time conformance assessment. Concurrently harden the diff engine on the manuals those first customers supply.
90-day revenue plan
HYPOTHESIS: $10-30k in one-time assessment revenue by day 90, from 4-8 customers, with zero recurring revenue yet. The recurring line does not switch on until the final rule publishes and the agency conforming regulations begin to land — that is the event that converts a one-time assessment into a monitoring subscription. Be honest that the 90-day number here is services revenue that de-risks and finances the product, not product revenue.
Distribution path
Three channels, in order of expected yield. (1) The comment docket itself: every entity that files a comment on 2026-10817 has publicly identified itself as a federal award recipient that cares enough about this rule to write about it, with a named contact. That is a free, precise, legally public lead list, and it is the single strongest distribution asset in this idea. (2) The free change-monitor as content — be the canonical 'what changed in the Uniform Guidance' page, which is a search-durable asset the moment the final rule drops. (3) Grants-consulting and CPA firms as a white-label channel. Notably absent: no ad spend, no network effects, no marketplace approval.
Pricing hypothesis
Free change-monitor (email alerts, public changelog). $2,500-5,000 one-time conformance assessment per organization. $3,600/yr recurring monitoring tier for a single recipient. $15,000-40,000/yr pass-through tier for a state agency, priced on subrecipient count. Wedge against the consulting fee, which is the existing spend being displaced.
Technical difficulty
Moderate and mostly bounded. eCFR and Federal Register both expose free, well-documented APIs with structured section-level data — the rule side is easy. The hard part is the customer side: policy manuals are unstructured PDFs with idiosyncratic numbering, and a mapping that is 85% right is worse than useless in a compliance context because the 15% is precisely where the audit finding lives. This is why the MVP must be human-reviewed. HYPOTHESIS: the engine reaches 'reviewable in 30 minutes instead of 8 hours' quickly, and never reaches 'unattended'. Price accordingly.
Legal / regulatory risk
Real and worth naming, though not disqualifying. Generating suggested policy language that a customer adopts, and characterizing it as conforming, edges toward unauthorized practice of law and toward assuming liability for an audit finding. Mitigation: sell it as a document-comparison and gap-identification tool, ship suggested language as excerpts of the regulation's own text rather than as novel drafting, disclaim in the MSA, and carry E&O. No licensure is required to operate — this is not a case where the founder must become certified, so `heavy_compliance` is not flagged.
Platform dependency
None meaningful. eCFR and the Federal Register are government systems with public APIs and no terms that permit deplatforming. There is no app store, no marketplace review, no platform owner. Customer documents arrive by upload. This is a genuine structural strength.
Founder fit
High but not maximal, and the gap matters. It matches the proven ELDT shape in three of four respects — read a federal mandate, identify who is forced to act, build the tooling, sell without relationship selling. It differs in the fourth and most important: ELDT terminated in a submission to a government portal with a per-upload fee and a binary success state. This one terminates in a document that an auditor may later disagree with. That is a softer product, a longer sales conversation, and a fuzzier value proof. The founder's public-records and systems-thinking strengths apply directly; the fast-revenue preference is stressed.
Breakout potential
The diff engine is regulation-agnostic. Once it maps any regulation's section tree onto any organization's policy corpus, the same machinery serves state grant pass-through rules across 50 states — the founder's stated replication thesis — and then any rule-to-policy conformance problem. That is the real asset. The Uniform Guidance revision is the wedge that funds building it, not the terminal market.
Final recommendation
CONDITIONAL PURSUE, with the conditions doing real work. Build the free change-monitor and file the public comment in the next 30 days — that costs under a week of effort, creates the lead list and the search-durable asset, and is worth doing even if the product never ships. Hand-build three conformance reports to test whether the mapping is good enough to be worth money. Then STOP and check two things before investing further: (1) do consulting firms bite at a paid assessment before the final rule exists, and (2) has OMB signaled a final-rule timeline. If the final rule slips past ~12 months, the incumbents will have caught up by the time the deadline arrives and the wedge closes. The honest framing is that this is a well-shaped bet on a rule that has not landed yet, sold into a buyer class that prefers to buy assurance from a person. It is a good fit for the founder's skills and a mediocre fit for his 30-180 day revenue window. Do not let the FORCED BUYER label in the input do more work than it has earned — a proposed rule with no compliance date is a forced buyer with no clock, and the clock is the whole reason forced-buyer plays work.
Next action
Within 7 days: stand up a public 'Uniform Guidance Watch' page backed by the eCFR API (2 CFR 200) and the Federal Register API (docket 2026-10817 plus the listed agencies' conforming rules), with an email alert signup. Then read the comment docket weekly and export every named commenter to a lead list. That is the single highest-value, lowest-cost, most reversible action available, and it is worth doing before any judgment about the product is final.