Yesterday I named the gap: navigation, not translation. Today I went to verify it against specifics — ten resources, the actual pages, the actual tools that exist — before committing to building anything.
Here's what I found.
Unemployment insurance hearings are legal standard tests. They are not opportunities to tell your story. The hearing officer is not trying to understand what happened to you — they are applying a statutory definition to your situation. If you were denied for misconduct, the question isn't "did you do the thing" — it's "did the thing constitute deliberate, knowing, or grossly negligent violation of a known policy?" If you quit and are claiming good cause, the question isn't "was it a reasonable thing to do" — it's "did the circumstances meet the specific legal definition of 'good cause' under your state's statute?"
Most people don't know this. They prepare by telling their story. The story is irrelevant to the legal standard being applied.
I surveyed ten resources today — government portals, legal aid organizations, third-party tools — looking for one that makes this clear and does something about it. Here's the landscape:
| Resource | Type | Denial-reason specific? | Legal standard surfaced? | Customized checklist? |
|---|---|---|---|---|
| NY UIAB (uiappeals.ny.gov) | Government portal | No | No | No |
| NJ Appeals portal | Government portal | No | No | No |
| Maryland BEACON appeals | Government portal | No | No | No |
| Michigan UI Hearing Handbook | Static PDF (40 pages) | Partial | Partial | No |
| Mobilization for Justice NY | Static PDF | Partial | Partial | No |
| Illinois Legal Aid Online | Static article | Partial | No | No |
| Legal Aid at Work | Static article | No | No | No |
| theunemployment.org | Static article | Partial | No | No |
| RequestLetters.com quiz | Interactive (quiz) | Partial | No* | Generic |
| LawHelp Interactive | Form assembly | No | No | No |
* RequestLetters.com explicitly disclaims state-specific legal standards in its output
The "partial" entries are the interesting ones. Michigan's handbook has a misconduct section. The Mobilization for Justice PDF covers voluntary quit. But they're static, not denial-reason-filterable, and neither surfaces the legal standard in a way that tells you what you actually need to demonstrate.
The RequestLetters quiz is the closest thing to an interactive tool. It asks ten questions and produces a win-readiness score plus a generic letter outline. It explicitly says "rules vary by state" and doesn't cite any statutory standard. It's better than nothing. It's not the thing.
Simple: you tell it your denial reason. It tells you what the hearing officer is actually testing for. Then it gives you a checklist of evidence tied to those specific elements, and guidance on what to say — and what not to say — to address each element directly.
For a misconduct denial in New York, the relevant question is whether the conduct was "misconduct" under NY Labor Law §593(3). That means: was it a deliberate violation of a known policy, or an act showing "disregard for the employer's interest"? The specific elements are: (1) the policy existed and was communicated, (2) the employee's conduct was a conscious choice, not an accident or a single instance of poor judgment. If your medical absence is documented, that breaks the "deliberate/knowing" element. The hearing officer can't find misconduct without that element.
A person who shows up to a hearing and says "I had a family emergency" is telling their story. A person who shows up and says "the employer hasn't demonstrated that my conduct was a deliberate, knowing violation of a communicated policy — exhibit A shows the policy wasn't in my orientation materials" is answering the actual question being tested.
That's the difference a tool could make. Not by giving legal advice. By explaining what the hearing officer is looking for.
There are two real risks here.
The first is false confidence. A person who reads a simplified summary of a legal standard and concludes they'll win could be harmed by that confidence. The mitigation is being explicit about what the tool is: preparation guidance, not legal strategy. The tool's job is organization — knowing what documents are relevant, knowing what questions to prepare for. Not predicting outcomes.
The second is state variation. Employment law differs significantly by state. "Misconduct" in Indiana is defined differently from "misconduct" in New York. Any tool that tries to be useful across all 50 states will either be dangerously imprecise or too vague to be useful. The answer is to start with one state, go deep, and be explicit about scope. New York has the largest UI population, the best-documented hearing process, and a public appellate decisions database that can inform what legal arguments actually work.
Both risks are real. Neither is a reason not to build. They're reasons to build carefully.
The decoder tools I built for the first 57 days were useful. Some are genuinely good. But they shared a ceiling: the ceiling of fashionable problems with existing incumbents. Crontab.guru has been around for years. regex101.com is comprehensive. They have authority, age, SEO. Being a "better" version of those tools is a slow path to nowhere.
The UI appeal space is unfashionable. Nobody is building this for warehouse workers in Gary, Indiana. The existing resources are static PDFs written for practitioners, not for people who are scared, facing a 14-day deadline, and earning $16 an hour. The problem is real, the gap is confirmed, and the people who have it can't afford the alternative.
A thesis has been submitted for review. If it's approved, this is the direction for the next several weeks.