Resistance Survival Guide #317
A government agency denies someone medical care, food assistance, disability benefits, unemployment compensation, housing, or another essential service. The notice arrives late. The explanation makes no sense. The deadline is already breathing down their neck. When they call for help, a recording tells them that every representative is busy experiencing an unusually high call volume that has apparently lasted since the invention of hold music.
Then comes the final insult. The agency tells them they may appeal.
An appeal process is supposed to protect people from incorrect government decisions. It should provide clear notice, access to the evidence, a meaningful opportunity to respond, and an impartial review. However, an appeal system can exist on paper while remaining practically unusable. A right buried beneath confusing instructions, inaccessible technology, missing records, impossible deadlines, and months of delay is not much of a right.
This Resistance Survival Guide explains how to investigate whether an appeal process corrects mistakes or simply exhausts the people who discover them.
Why Appeal Systems Deserve Investigation
Automated government systems can decide who receives assistance, whose account is flagged, which claim receives additional scrutiny, and whether an individual must repay money. Yet people affected by these systems may never learn that automation influenced the decision.
The Markup found that obtaining basic information about government algorithms can require extraordinary effort. That secrecy creates a direct problem for due process. A person cannot meaningfully challenge a decision without knowing how it was made, what information was used, or what allegedly went wrong.
The appeal itself may conceal another layer of automation. Software can sort cases, assign priorities, generate hearing packets, recommend outcomes, detect supposed fraud, or identify appeals for dismissal. Contractors may operate call centers, document portals, or case management systems. A human reviewer may technically approve the final result while having little time or authority to question what the system produced.
The agency will still call this human review. Resistance Kitty calls it decorative accountability.
What a Meaningful Appeal Should Provide
A legitimate appeal process should give a person understandable notice of the decision and its factual basis. It should explain the deadline, the method for filing, the available hearing options, and whether benefits continue while the appeal is pending. It should also provide access to the evidence used by the agency and enough time to prepare a response.
The constitutional requirements vary according to the program and the interests involved. However, the basic principles of notice and an opportunity to be heard are deeply rooted in American due process law. These protections become hollow when notices are vague, evidence remains hidden, filing systems malfunction, or hearings occur after the damage can no longer be repaired.
The United States Department of Labor has recognized that delays in unemployment appeals can cause serious hardship precisely when people most need assistance. That observation is painfully obvious, yet some agencies continue treating time as if it affects only the spreadsheet.
Warning Signs That the Process May Be Designed To Fail
One bad experience does not prove that an entire appeal system is deliberately obstructive. Investigators should look for repeated patterns across cases, locations, demographic groups, decision types, and time periods.
Warning signs include notices that do not identify the evidence supporting the decision, appeal instructions written differently across agency webpages, deadlines calculated from the date a notice was created rather than received, inaccessible online forms, missing language assistance, long telephone waits, unexplained dismissals, and hearings scheduled after benefits have already stopped.
Other signs may appear in the data. Look for unusually low appeal rates despite high denial rates, large numbers of appeals dismissed before a hearing, substantial delays, sharp differences between offices, and high reversal rates once a person finally reaches an independent reviewer.
A high reversal rate does not automatically prove misconduct. It may reflect new evidence or different legal interpretations. However, it can also indicate that the original decision system is producing unreliable results. When an agency repeatedly gets the answer wrong and forces each affected person to fight alone, the appeal process becomes part of the harm.
Step by Step Guide
Step One: Define the Decision and the Appeal Path
Begin with one specific government decision. Identify the program, agency, office, contractor, decision date, notice type, appeal deadline, and every available level of review.
Create a simple sequence showing what happens after the original decision. Record whether the person must request reconsideration, attend an informal conference, appear before a hearing officer, seek review from an administrative board, or file in court.
Do not assume the agency website describes the complete process. Compare the notice with statutes, regulations, policy manuals, hearing guides, administrative rules, and legal aid materials. If the instructions conflict, save every version and record when each one was accessed.
This first map reveals where people can be lost. An appeal path with six doors is not accessible merely because the agency labeled one of them assistance.
Step Two: Collect Real Notices
Gather denial, suspension, termination, overpayment, and fraud notices from affected people who consent to share them. Remove names, addresses, identification numbers, medical information, account numbers, barcodes, and other sensitive details before storing or publishing anything.
For each notice, record the decision type, date printed, date received, response deadline, stated reason, legal authority, evidence mentioned, appeal instructions, language, reading level, and available accommodations.
Compare several notices side by side. Determine whether they explain what happened in concrete language. A notice stating that someone failed to meet program requirements does not identify which requirement was involved or what evidence supposedly established the failure.
Watch for notices that tell people to submit evidence without explaining what evidence would matter. That is not guidance. It is bureaucratic charades with groceries and medical care on the line.
Step Three: Test Every Way To File an Appeal
Document every filing method offered by the agency. This may include an online portal, telephone line, email address, fax number, mailed form, or physical office.
Test public portions of the process without submitting false information or impersonating an applicant. Record whether instructions are easy to locate, whether forms work on mobile devices, whether confirmation numbers are provided, and whether the system clearly explains what happens next.
Check accessibility with screen readers, keyboard navigation, text enlargement, and translation tools when possible. Ask affected users and disability advocates about barriers that automated testing may miss.
Document business hours and telephone wait times at different times of day. Save screenshots of errors. Record disconnected calls and unanswered messages. Note whether the agency offers an alternative when the primary system fails.
A right that requires a printer, fax machine, desktop computer, flawless internet connection, spare weekday afternoon, and emotional immunity to hold music may have been designed for an imaginary citizen.
Step Four: Measure the Deadline Trap
Identify when the appeal clock begins. It may start when the agency issues the notice, mails it, delivers it, or when the recipient receives it. Those dates are not interchangeable.
Compare the official deadline with the actual delivery date. Ask whether weekends and holidays count. Determine whether the agency accepts late appeals for good cause and how often that exception is granted.
Look for notices sent close to the deadline, returned mail, incorrect addresses, portal messages without separate notification, and delays caused by contractors. Examine whether people lose benefits before a hearing can occur.
Calculate how many usable days a person actually receives after delivery. The agency may advertise a thirty day appeal period while the person receives the notice with twelve days remaining. Mathematics is often where administrative generosity goes to die.
Step Five: Request the Performance Data
Search agency budgets, annual reports, legislative presentations, meeting minutes, audits, inspector general reports, administrative dashboards, and procurement records.
Then request records showing the number of initial decisions, denials, appeals filed, appeals withdrawn, appeals dismissed, hearings completed, decisions affirmed, decisions reversed, and cases returned for additional review.
Request the median and average time between each stage. Ask for the oldest pending case, the number of cases exceeding required time limits, the number of hearing officers, vacancies, caseloads, turnover, contractor staffing, and complaint records.
Break the information down by office, county, decision type, program, month, language, disability accommodation, race, gender, age, or other available categories. Agencies may refuse some categories or claim they do not maintain them. Record those answers too. A government cannot manage inequity it has chosen not to measure.
Step Six: Calculate the Rates That Matter
Calculate the appeal rate by dividing the number of appeals filed by the number of appealable adverse decisions. A very low rate may mean the original decisions are accurate. It may also mean people do not understand the notice, cannot complete the process, or have already given up.
Next, calculate dismissal, withdrawal, affirmation, reversal, and remand rates. Compare results across offices and hearing officers when legally and ethically appropriate.
Separate procedural dismissals from decisions reached after reviewing the facts. If a large percentage of people lose because they missed a deadline, used the wrong form, failed to attend a hearing they did not know about, or could not submit a document through a broken portal, the process is filtering people rather than evaluating their claims.
Study how outcomes change when appellants have lawyers, advocates, interpreters, or other assistance. A dramatic representation gap can reveal that the system is too complex for ordinary people to navigate alone.
Step Seven: Identify Every Contractor
Search contracts, amendments, purchase orders, bid documents, invoices, corporate records, privacy notices, and technology descriptions.
Determine whether private companies created the original decision system, operate the appeal portal, staff the call center, scan documents, host records, prepare hearing packets, or provide analytics used during review.
Read the contract for performance standards. Look for required response times, error thresholds, complaint procedures, accessibility requirements, audit rights, payment incentives, penalties, and ownership of appeal data.
Find out whether a contractor earns more by processing cases quickly, identifying overpayments, reducing benefit costs, or meeting production targets. Financial incentives do not prove that individual decisions were improper. They do identify questions worth asking before everyone admires the efficiency dashboard.
Step Eight: Examine the Evidence Available to Appellants
Determine what evidence the agency uses and when the affected person may see it. Ask whether the file contains database matches, risk scores, internal notes, call recordings, medical reviews, identity verification results, employment records, location information, or material supplied by another agency.
Compare the information available to the decision maker with what the appellant receives. Look for unexplained codes, incomplete files, missing attachments, undisclosed scoring rules, and records added shortly before a hearing.
An agency may claim that a vendor’s methods are proprietary. That claim should be documented and challenged through records requests, hearing procedures, elected officials, oversight bodies, and qualified attorneys. The government should not be permitted to remove a benefit using evidence that the affected person is forbidden to understand.
Step Nine: Interview the People Inside the Process
Speak with appellants, legal aid attorneys, advocates, former employees, union representatives, hearing officers, interpreters, caseworkers, and social service providers.
Ask each person where cases become delayed or disappear. Find out whether staff are evaluated by volume, speed, accuracy, reversals, or customer complaints. Ask whether employees can override automated recommendations and whether doing so requires additional approval.
Do not ask workers to disclose protected personal information. Focus on procedures, training, staffing, software, pressure, and recurring failure patterns.
Compare testimony with documents. One account can identify a lead. Repeated accounts supported by records can establish a pattern.
Step Ten: Follow Reversals Back to the Original Failure
A reversal is not the end of the investigation. Determine why the original decision was wrong.
Was the information inaccurate? Did the system confuse two people? Was a document scanned but never attached to the case? Did a database update arrive late? Did the algorithm misclassify a household? Did a worker follow a flawed policy? Did the reviewer have evidence the original decision maker never considered?
Group reversals by cause. Then ask whether the agency corrected the underlying system or merely fixed the individual case.
An appeal system designed to correct errors should feed lessons back into the original process. If the same error continues harming new people, the agency is not learning. It is operating a subscription service for preventable suffering.
Step Eleven: Compare Policy With Reality
Create a table comparing each formal requirement with actual practice. Include notice standards, filing methods, deadlines, access to records, hearing time limits, accommodations, interpreter access, continuation of benefits, and decision deadlines.
For every requirement, identify the legal source, agency promise, observed practice, supporting evidence, and consequence of failure.
This comparison keeps the investigation grounded. It distinguishes frustrating experiences from documented violations and systemic patterns. It also makes the findings easier for journalists, advocates, inspectors general, legislators, and attorneys to evaluate.
Step Twelve: Publish the Pattern Without Exposing the People
Explain the system through documents, aggregated data, carefully verified examples, and the experiences of people who gave informed consent.
Do not publish sensitive files merely because an agency failed to protect them. Remove metadata and hidden personal information. Allow sources to review how their private circumstances will be described, but retain independent control over factual conclusions.
State what the evidence proves, what it suggests, and what remains unknown. Avoid claiming deliberate sabotage unless records or testimony support intent. A process can systematically obstruct people even when no official wrote a memo titled Operation Make Everyone Miserable.
End with specific accountability questions. Who designed the process? Who approved it? Who monitors errors? Who reviews dismissals? Who benefits financially? Who knew about the barriers? What changed after the agency learned people were being harmed?
How To Recognize a Systemic Failure
The strongest investigation will combine several forms of evidence.
Individual stories demonstrate the human consequences. Notices reveal what the agency communicated. Performance data show scale. Contracts identify private involvement. Regulations establish what should have happened. Interviews explain how the machinery works. Reversal records expose where it breaks.
No single number tells the entire story. A low appeal rate may reflect barriers. A high reversal rate may expose defective original decisions. A long processing time may conceal staffing shortages, procedural complexity, or intentional neglect. The job is to connect the measurements without forcing them to say more than they prove.
That discipline matters. We are investigating government failure, not auditioning for its public relations department or manufacturing outrage it has not already earned.
Questions Every Agency Should Answer
- What percentage of adverse decisions are appealed?
- How many appeals are dismissed without a review of the underlying evidence?
- How long does each stage take?
- How many people lose assistance while waiting?
- What information is withheld from appellants?
- Can a reviewer override the automated system?
- How often does that happen?
- Are contractors involved in the original decision or appeal?
- What performance incentives govern their work?
- What errors appear most frequently in reversed decisions?
- What has the agency changed in response?
If officials cannot answer these questions, that is itself an important finding. If they can answer but refuse, document the refusal.
In Closing
An appeal is not meaningful merely because an agency printed instructions at the bottom of a denial notice. People need understandable reasons, access to evidence, sufficient time, functional filing methods, impartial review, and a decision before the damage becomes permanent.
When those protections disappear, the appeal process stops acting as a safeguard. It becomes an endurance test imposed on people who are already dealing with illness, unemployment, disability, hunger, housing insecurity, or another crisis.
Investigate the notices. Test the systems. Count the dismissals. Measure the delays. Follow the contractors. Read the reversals. Listen to the people forced to navigate the maze.
Then ask the deliciously ugly question.
Is this process correcting government mistakes, or merely exhausting the humans who found them?
Sources
- The Markup: It Takes a Small Miracle to Learn Basic Facts About Government Algorithms
- The Markup: Online Unemployment Benefits Systems Are Buckling Under a Wave of Applications
- The Markup: Can a Robot Decide My Medical Treatment?
- United States Department of Labor: Appeals Timeliness Measures
- United States Department of Labor: Benefits Timeliness and Quality Measures
- Government Accountability Office: Social Security Disability Appeal Productivity Expectations
- Government Accountability Office: Social Security Disability Backlog Reduction Efforts
- Center on Budget and Policy Priorities: Introduction to Medicaid
